[Runner] - Start a new experiment
2023-08-11 02:23:47 | INFO | s3prl.upstream.hubert.hubconf | Converting a fairseq checkpoint: /home/victor/checkpoints/hubert_d2v2_800k.pt
2023-08-11 02:23:47 | INFO | s3prl.upstream.hubert.hubconf | To: /home/victor/checkpoints/hubert_d2v2_800k.converted.pt
2023-08-11 02:23:47 | INFO | fairseq.models.hubert.hubert | HubertModel Config: HubertConfig(_name='hubert', label_rate=50.0, extractor_mode='default', encoder_layers=12, encoder_embed_dim=768, encoder_ffn_embed_dim=3072, encoder_attention_heads=12, activation_fn='gelu', layer_type='transformer', dropout=0.1, attention_dropout=0.1, activation_dropout=0.0, encoder_layerdrop=0.05, dropout_input=0.1, dropout_features=0.1, final_dim=256, untie_final_proj=True, layer_norm_first=False, conv_feature_layers='[(512,10,5)] + [(512,3,2)] * 4 + [(512,2,2)] * 3', conv_bias=False, logit_temp=0.1, target_glu=False, feature_grad_mult=0.1, mask_length=10, mask_prob=0.8, mask_selection='static', mask_other=0.0, no_mask_overlap=False, mask_min_space=1, mask_channel_length=10, mask_channel_prob=0.0, mask_channel_selection='static', mask_channel_other=0.0, no_mask_channel_overlap=False, mask_channel_min_space=1, conv_pos=128, conv_pos_groups=16, latent_temp=[2.0, 0.5, 0.999995], skip_masked=False, skip_nomask=False, checkpoint_activations=False, required_seq_len_multiple=2, depthwise_conv_kernel_size=31, attn_type='', pos_enc_type='abs', fp16=False)
[Featurizer] - Take a list of 13 features and weighted sum them.
[(512, 10, 5), (512, 3, 2), (512, 3, 2), (512, 3, 2), (512, 3, 2), (512, 2, 2), (512, 2, 2), (512, 2, 2)] 640
[Featurizer] - The selected feature hidden_states's downsample rate is 640

overall:   0%|          | 0/4000 [00:00<?, ?it/s][LibriPhone] - Taking the first phoneme sequences for a deterministic behavior.


word -> phonemes:   0%|          | 0/3515 [00:00<?, ?it/s]

word -> phonemes:   6%|▌         | 207/3515 [00:00<00:01, 2068.32it/s]

word -> phonemes:  12%|█▏        | 414/3515 [00:00<00:01, 1844.85it/s]

word -> phonemes:  17%|█▋        | 601/3515 [00:00<00:01, 1848.77it/s]

word -> phonemes:  22%|██▏       | 787/3515 [00:00<00:01, 1850.64it/s]

word -> phonemes:  28%|██▊       | 987/3515 [00:00<00:01, 1901.15it/s]

word -> phonemes:  34%|███▍      | 1190/3515 [00:00<00:01, 1940.68it/s]

word -> phonemes:  41%|████▏     | 1455/3515 [00:00<00:00, 2167.66it/s]

word -> phonemes:  48%|████▊     | 1673/3515 [00:00<00:00, 2092.24it/s]

word -> phonemes:  54%|█████▎    | 1887/3515 [00:00<00:00, 2105.99it/s]

word -> phonemes:  62%|██████▏   | 2192/3515 [00:01<00:00, 2388.84it/s]

word -> phonemes:  70%|███████   | 2469/3515 [00:01<00:00, 2503.08it/s]

word -> phonemes:  79%|███████▊  | 2766/3515 [00:01<00:00, 2642.42it/s]

word -> phonemes:  86%|████████▋ | 3032/3515 [00:01<00:00, 2616.96it/s]

word -> phonemes:  94%|█████████▎| 3295/3515 [00:01<00:00, 2295.34it/s]
word -> phonemes: 100%|██████████| 3515/3515 [00:01<00:00, 2209.41it/s]


train:   0%|          | 0/3515 [00:00<?, ?it/s]
overall:   0%|          | 1/4000 [00:04<4:44:33,  4.27s/it]

train:   0%|          | 1/3515 [00:01<1:36:11,  1.64s/it]
overall:   0%|          | 2/4000 [00:04<2:07:01,  1.91s/it]

train:   0%|          | 2/3515 [00:01<48:16,  1.21it/s]  
overall:   0%|          | 3/4000 [00:04<1:16:32,  1.15s/it]

train:   0%|          | 3/3515 [00:02<32:50,  1.78it/s]
overall:   0%|          | 4/4000 [00:05<53:40,  1.24it/s]  

train:   0%|          | 4/3515 [00:02<26:20,  2.22it/s]
overall:   0%|          | 5/4000 [00:05<40:28,  1.65it/s]

train:   0%|          | 5/3515 [00:02<22:14,  2.63it/s]
overall:   0%|          | 6/4000 [00:05<32:55,  2.02it/s]

train:   0%|          | 6/3515 [00:02<20:08,  2.90it/s]
overall:   0%|          | 7/4000 [00:05<28:01,  2.38it/s]

train:   0%|          | 7/3515 [00:03<18:42,  3.13it/s]
overall:   0%|          | 8/4000 [00:06<24:30,  2.72it/s]

train:   0%|          | 8/3515 [00:03<17:30,  3.34it/s]
overall:   0%|          | 9/4000 [00:06<22:21,  2.97it/s]

train:   0%|          | 9/3515 [00:03<16:52,  3.46it/s]
overall:   0%|          | 10/4000 [00:07<29:37,  2.24it/s]

train:   0%|          | 10/3515 [00:04<24:06,  2.42it/s]
overall:   0%|          | 11/4000 [00:07<25:55,  2.56it/s]

train:   0%|          | 11/3515 [00:04<21:27,  2.72it/s]
overall:   0%|          | 12/4000 [00:07<23:03,  2.88it/s]

train:   0%|          | 12/3515 [00:04<19:20,  3.02it/s]
overall:   0%|          | 13/4000 [00:07<20:43,  3.21it/s]

train:   0%|          | 13/3515 [00:05<17:33,  3.32it/s]
overall:   0%|          | 14/4000 [00:08<19:49,  3.35it/s]

train:   0%|          | 14/3515 [00:05<16:57,  3.44it/s]
overall:   0%|          | 15/4000 [00:08<18:40,  3.56it/s]

train:   0%|          | 15/3515 [00:05<16:05,  3.63it/s]
overall:   0%|          | 16/4000 [00:08<18:20,  3.62it/s]

train:   0%|          | 16/3515 [00:05<15:53,  3.67it/s]
overall:   0%|          | 17/4000 [00:08<18:02,  3.68it/s]

train:   0%|          | 17/3515 [00:06<15:41,  3.71it/s]
overall:   0%|          | 18/4000 [00:09<16:59,  3.91it/s]

train:   1%|          | 18/3515 [00:06<14:49,  3.93it/s]
overall:   0%|          | 19/4000 [00:09<17:09,  3.87it/s]

train:   1%|          | 19/3515 [00:06<14:59,  3.88it/s]train loss: 4.2257795333862305
train per: 0.9310266332802185

overall:   0%|          | 20/4000 [00:09<17:29,  3.79it/s]

train:   1%|          | 20/3515 [00:06<15:18,  3.80it/s]
overall:   1%|          | 21/4000 [00:09<17:10,  3.86it/s]

train:   1%|          | 21/3515 [00:07<15:02,  3.87it/s]
overall:   1%|          | 22/4000 [00:10<17:10,  3.86it/s]

train:   1%|          | 22/3515 [00:07<15:02,  3.87it/s]
overall:   1%|          | 23/4000 [00:10<17:17,  3.83it/s]

train:   1%|          | 23/3515 [00:07<15:10,  3.84it/s]
overall:   1%|          | 24/4000 [00:10<16:56,  3.91it/s]

train:   1%|          | 24/3515 [00:07<14:51,  3.92it/s]
overall:   1%|          | 25/4000 [00:10<17:14,  3.84it/s]

train:   1%|          | 25/3515 [00:08<15:07,  3.85it/s]
overall:   1%|          | 26/4000 [00:11<17:19,  3.82it/s]

train:   1%|          | 26/3515 [00:08<15:11,  3.83it/s]
overall:   1%|          | 27/4000 [00:11<16:38,  3.98it/s]

train:   1%|          | 27/3515 [00:08<14:36,  3.98it/s]
overall:   1%|          | 28/4000 [00:11<16:37,  3.98it/s]

train:   1%|          | 28/3515 [00:09<14:35,  3.98it/s]
overall:   1%|          | 29/4000 [00:11<16:50,  3.93it/s]

train:   1%|          | 29/3515 [00:09<14:47,  3.93it/s]
overall:   1%|          | 30/4000 [00:12<17:08,  3.86it/s]

train:   1%|          | 30/3515 [00:09<15:03,  3.86it/s]
overall:   1%|          | 31/4000 [00:12<16:45,  3.95it/s]

train:   1%|          | 31/3515 [00:09<14:42,  3.95it/s]
overall:   1%|          | 32/4000 [00:12<16:48,  3.93it/s]

train:   1%|          | 32/3515 [00:10<14:45,  3.93it/s]
overall:   1%|          | 33/4000 [00:12<17:13,  3.84it/s]

train:   1%|          | 33/3515 [00:10<15:06,  3.84it/s]
overall:   1%|          | 34/4000 [00:13<21:02,  3.14it/s]

train:   1%|          | 34/3515 [00:10<18:27,  3.14it/s]
overall:   1%|          | 35/4000 [00:13<20:02,  3.30it/s]

train:   1%|          | 35/3515 [00:11<17:35,  3.30it/s]
overall:   1%|          | 36/4000 [00:13<19:24,  3.40it/s]

train:   1%|          | 36/3515 [00:11<17:02,  3.40it/s]
overall:   1%|          | 38/4000 [00:14<15:32,  4.25it/s]

train:   1%|          | 38/3515 [00:11<13:38,  4.25it/s]
overall:   1%|          | 39/4000 [00:14<16:05,  4.10it/s]

train:   1%|          | 39/3515 [00:11<14:06,  4.10it/s]train loss: 1.51618230342865
train per: 0.40246567532899424

overall:   1%|          | 40/4000 [00:14<16:38,  3.97it/s]

train:   1%|          | 40/3515 [00:12<14:36,  3.96it/s]
overall:   1%|          | 41/4000 [00:15<16:52,  3.91it/s]

train:   1%|          | 41/3515 [00:12<14:47,  3.91it/s]
overall:   1%|          | 42/4000 [00:15<17:02,  3.87it/s]

train:   1%|          | 42/3515 [00:12<14:57,  3.87it/s]
overall:   1%|          | 43/4000 [00:15<17:01,  3.87it/s]

train:   1%|          | 43/3515 [00:12<14:55,  3.88it/s]
overall:   1%|          | 44/4000 [00:15<16:59,  3.88it/s]

train:   1%|▏         | 44/3515 [00:13<14:54,  3.88it/s]
overall:   1%|          | 45/4000 [00:16<16:55,  3.89it/s]

train:   1%|▏         | 45/3515 [00:13<14:51,  3.89it/s]
overall:   1%|          | 46/4000 [00:16<13:55,  4.73it/s]

train:   1%|▏         | 46/3515 [00:13<12:12,  4.73it/s]
overall:   1%|          | 47/4000 [00:16<14:58,  4.40it/s]

train:   1%|▏         | 47/3515 [00:13<13:08,  4.40it/s]
overall:   1%|          | 48/4000 [00:16<15:36,  4.22it/s]

train:   1%|▏         | 48/3515 [00:14<13:41,  4.22it/s]
overall:   1%|          | 49/4000 [00:16<14:59,  4.39it/s]

train:   1%|▏         | 49/3515 [00:14<13:08,  4.39it/s]
overall:   1%|▏         | 50/4000 [00:17<15:49,  4.16it/s]

train:   1%|▏         | 50/3515 [00:14<13:53,  4.16it/s]
overall:   1%|▏         | 51/4000 [00:17<16:15,  4.05it/s]

train:   1%|▏         | 51/3515 [00:14<14:15,  4.05it/s]
overall:   1%|▏         | 52/4000 [00:17<16:30,  3.98it/s]

train:   1%|▏         | 52/3515 [00:15<14:29,  3.98it/s]
overall:   1%|▏         | 54/4000 [00:17<11:29,  5.73it/s]

train:   2%|▏         | 54/3515 [00:15<10:04,  5.73it/s]
overall:   1%|▏         | 55/4000 [00:18<13:00,  5.05it/s]

train:   2%|▏         | 55/3515 [00:15<11:24,  5.05it/s]
overall:   1%|▏         | 56/4000 [00:18<14:18,  4.59it/s]

train:   2%|▏         | 56/3515 [00:15<12:33,  4.59it/s]
overall:   1%|▏         | 57/4000 [00:18<15:16,  4.30it/s]

train:   2%|▏         | 57/3515 [00:16<13:23,  4.30it/s]
overall:   1%|▏         | 58/4000 [00:19<21:38,  3.04it/s]

train:   2%|▏         | 58/3515 [00:16<18:58,  3.04it/s]
overall:   1%|▏         | 59/4000 [00:19<20:01,  3.28it/s]

train:   2%|▏         | 59/3515 [00:16<17:33,  3.28it/s]train loss: 1.041218876838684
train per: 0.28108165768841004

overall:   2%|▏         | 61/4000 [00:19<15:52,  4.14it/s]

train:   2%|▏         | 61/3515 [00:17<13:54,  4.14it/s]
overall:   2%|▏         | 62/4000 [00:20<16:23,  4.01it/s]

train:   2%|▏         | 62/3515 [00:17<14:22,  4.01it/s]
overall:   2%|▏         | 63/4000 [00:20<16:38,  3.94it/s]

train:   2%|▏         | 63/3515 [00:17<14:35,  3.94it/s]
overall:   2%|▏         | 64/4000 [00:20<16:46,  3.91it/s]

train:   2%|▏         | 64/3515 [00:18<14:42,  3.91it/s]
overall:   2%|▏         | 65/4000 [00:20<16:49,  3.90it/s]

train:   2%|▏         | 65/3515 [00:18<14:44,  3.90it/s]
overall:   2%|▏         | 66/4000 [00:21<16:32,  3.96it/s]

train:   2%|▏         | 66/3515 [00:18<14:30,  3.96it/s]
overall:   2%|▏         | 68/4000 [00:21<14:02,  4.67it/s]

train:   2%|▏         | 68/3515 [00:18<12:18,  4.67it/s]
overall:   2%|▏         | 69/4000 [00:21<14:57,  4.38it/s]

train:   2%|▏         | 69/3515 [00:19<13:07,  4.38it/s]
overall:   2%|▏         | 70/4000 [00:22<15:23,  4.26it/s]

train:   2%|▏         | 70/3515 [00:19<13:29,  4.26it/s]
overall:   2%|▏         | 71/4000 [00:22<15:44,  4.16it/s]

train:   2%|▏         | 71/3515 [00:19<13:47,  4.16it/s]
overall:   2%|▏         | 72/4000 [00:22<16:17,  4.02it/s]

train:   2%|▏         | 72/3515 [00:19<14:16,  4.02it/s]
overall:   2%|▏         | 73/4000 [00:22<16:27,  3.98it/s]

train:   2%|▏         | 73/3515 [00:20<14:25,  3.98it/s]
overall:   2%|▏         | 74/4000 [00:23<16:20,  4.00it/s]

train:   2%|▏         | 74/3515 [00:20<14:19,  4.00it/s]
overall:   2%|▏         | 75/4000 [00:23<16:44,  3.91it/s]

train:   2%|▏         | 75/3515 [00:20<14:40,  3.91it/s]
overall:   2%|▏         | 77/4000 [00:23<14:22,  4.55it/s]

train:   2%|▏         | 77/3515 [00:21<12:35,  4.55it/s]
overall:   2%|▏         | 78/4000 [00:23<14:58,  4.37it/s]

train:   2%|▏         | 78/3515 [00:21<13:07,  4.37it/s]
overall:   2%|▏         | 79/4000 [00:24<15:25,  4.24it/s]

train:   2%|▏         | 79/3515 [00:21<13:30,  4.24it/s]train loss: 0.8834659457206726
train per: 0.24703772807676463

overall:   2%|▏         | 80/4000 [00:24<16:10,  4.04it/s]

train:   2%|▏         | 80/3515 [00:21<14:11,  4.04it/s]
overall:   2%|▏         | 81/4000 [00:24<16:34,  3.94it/s]

train:   2%|▏         | 81/3515 [00:22<14:31,  3.94it/s]
overall:   2%|▏         | 82/4000 [00:25<22:20,  2.92it/s]

train:   2%|▏         | 82/3515 [00:22<19:34,  2.92it/s]
overall:   2%|▏         | 83/4000 [00:25<20:38,  3.16it/s]

train:   2%|▏         | 83/3515 [00:22<18:04,  3.16it/s]
overall:   2%|▏         | 85/4000 [00:25<16:26,  3.97it/s]

train:   2%|▏         | 85/3515 [00:23<14:24,  3.97it/s]
overall:   2%|▏         | 86/4000 [00:26<16:40,  3.91it/s]

train:   2%|▏         | 86/3515 [00:23<14:36,  3.91it/s]
overall:   2%|▏         | 87/4000 [00:26<16:58,  3.84it/s]

train:   2%|▏         | 87/3515 [00:23<14:52,  3.84it/s]
overall:   2%|▏         | 88/4000 [00:26<16:39,  3.91it/s]

train:   3%|▎         | 88/3515 [00:24<14:35,  3.91it/s]
overall:   2%|▏         | 89/4000 [00:26<16:49,  3.88it/s]

train:   3%|▎         | 89/3515 [00:24<14:44,  3.88it/s]
overall:   2%|▏         | 90/4000 [00:27<16:50,  3.87it/s]

train:   3%|▎         | 90/3515 [00:24<14:44,  3.87it/s]
overall:   2%|▏         | 91/4000 [00:27<16:59,  3.83it/s]

train:   3%|▎         | 91/3515 [00:24<14:53,  3.83it/s]
overall:   2%|▏         | 92/4000 [00:27<17:04,  3.81it/s]

train:   3%|▎         | 92/3515 [00:25<14:57,  3.81it/s]
overall:   2%|▏         | 93/4000 [00:27<14:04,  4.62it/s]

train:   3%|▎         | 93/3515 [00:25<12:19,  4.62it/s]
overall:   2%|▏         | 94/4000 [00:28<14:32,  4.48it/s]

train:   3%|▎         | 94/3515 [00:25<12:43,  4.48it/s]
overall:   2%|▏         | 95/4000 [00:28<15:06,  4.31it/s]

train:   3%|▎         | 95/3515 [00:25<13:13,  4.31it/s]
overall:   2%|▏         | 96/4000 [00:28<15:45,  4.13it/s]

train:   3%|▎         | 96/3515 [00:25<13:47,  4.13it/s]
overall:   2%|▏         | 98/4000 [00:28<13:37,  4.78it/s]

train:   3%|▎         | 98/3515 [00:26<11:55,  4.78it/s]
overall:   2%|▏         | 99/4000 [00:29<14:23,  4.52it/s]

train:   3%|▎         | 99/3515 [00:26<12:36,  4.52it/s]train loss: 0.8121033906936646
train per: 0.23004193860585156

overall:   2%|▎         | 100/4000 [00:29<15:28,  4.20it/s]

train:   3%|▎         | 100/3515 [00:26<13:34,  4.20it/s]
overall:   3%|▎         | 101/4000 [00:29<15:55,  4.08it/s]

train:   3%|▎         | 101/3515 [00:27<13:56,  4.08it/s]
overall:   3%|▎         | 102/4000 [00:30<16:11,  4.01it/s]

train:   3%|▎         | 102/3515 [00:27<14:10,  4.01it/s]
overall:   3%|▎         | 103/4000 [00:30<16:31,  3.93it/s]

train:   3%|▎         | 103/3515 [00:27<14:28,  3.93it/s]
overall:   3%|▎         | 104/4000 [00:30<16:31,  3.93it/s]

train:   3%|▎         | 104/3515 [00:27<14:28,  3.93it/s]
overall:   3%|▎         | 105/4000 [00:30<16:37,  3.90it/s]

train:   3%|▎         | 105/3515 [00:28<14:33,  3.90it/s]
overall:   3%|▎         | 106/4000 [00:31<22:53,  2.84it/s]

train:   3%|▎         | 106/3515 [00:28<20:02,  2.84it/s]
overall:   3%|▎         | 107/4000 [00:31<18:57,  3.42it/s]

train:   3%|▎         | 107/3515 [00:28<16:36,  3.42it/s]
overall:   3%|▎         | 108/4000 [00:31<18:17,  3.55it/s]

train:   3%|▎         | 108/3515 [00:29<16:01,  3.55it/s]
overall:   3%|▎         | 109/4000 [00:32<18:14,  3.56it/s]

train:   3%|▎         | 109/3515 [00:29<15:57,  3.56it/s]
overall:   3%|▎         | 110/4000 [00:32<17:31,  3.70it/s]

train:   3%|▎         | 110/3515 [00:29<15:20,  3.70it/s]
overall:   3%|▎         | 111/4000 [00:32<17:12,  3.77it/s]

train:   3%|▎         | 111/3515 [00:29<15:03,  3.77it/s]
overall:   3%|▎         | 112/4000 [00:32<17:04,  3.80it/s]

train:   3%|▎         | 112/3515 [00:30<14:56,  3.80it/s]
overall:   3%|▎         | 113/4000 [00:33<17:13,  3.76it/s]

train:   3%|▎         | 113/3515 [00:30<15:04,  3.76it/s]
overall:   3%|▎         | 114/4000 [00:33<17:03,  3.80it/s]

train:   3%|▎         | 114/3515 [00:30<14:56,  3.80it/s]
overall:   3%|▎         | 115/4000 [00:33<17:02,  3.80it/s]

train:   3%|▎         | 115/3515 [00:30<14:55,  3.80it/s]
overall:   3%|▎         | 116/4000 [00:33<16:57,  3.82it/s]

train:   3%|▎         | 116/3515 [00:31<14:50,  3.82it/s]
overall:   3%|▎         | 117/4000 [00:34<16:50,  3.84it/s]

train:   3%|▎         | 117/3515 [00:31<14:44,  3.84it/s]
overall:   3%|▎         | 118/4000 [00:34<16:51,  3.84it/s]

train:   3%|▎         | 118/3515 [00:31<14:45,  3.84it/s]
overall:   3%|▎         | 119/4000 [00:34<16:33,  3.91it/s]

train:   3%|▎         | 119/3515 [00:32<14:29,  3.91it/s]train loss: 0.7602599859237671
train per: 0.21385698757030105

overall:   3%|▎         | 120/4000 [00:34<15:07,  4.27it/s]

train:   3%|▎         | 120/3515 [00:32<13:15,  4.27it/s]
overall:   3%|▎         | 121/4000 [00:35<15:55,  4.06it/s]

train:   3%|▎         | 121/3515 [00:32<13:55,  4.06it/s]
overall:   3%|▎         | 122/4000 [00:35<16:23,  3.94it/s]

train:   3%|▎         | 122/3515 [00:32<14:20,  3.94it/s]
overall:   3%|▎         | 123/4000 [00:35<14:19,  4.51it/s]

train:   3%|▎         | 123/3515 [00:32<12:31,  4.51it/s]
overall:   3%|▎         | 124/4000 [00:35<15:04,  4.28it/s]

train:   4%|▎         | 124/3515 [00:33<13:11,  4.28it/s]
overall:   3%|▎         | 125/4000 [00:36<15:42,  4.11it/s]

train:   4%|▎         | 125/3515 [00:33<13:44,  4.11it/s]
overall:   3%|▎         | 126/4000 [00:36<16:02,  4.02it/s]

train:   4%|▎         | 126/3515 [00:33<14:02,  4.02it/s]
overall:   3%|▎         | 127/4000 [00:36<16:11,  3.99it/s]

train:   4%|▎         | 127/3515 [00:33<14:09,  3.99it/s]
overall:   3%|▎         | 128/4000 [00:36<16:29,  3.91it/s]

train:   4%|▎         | 128/3515 [00:34<14:25,  3.91it/s]
overall:   3%|▎         | 129/4000 [00:37<17:05,  3.77it/s]

train:   4%|▎         | 129/3515 [00:34<14:57,  3.77it/s]
overall:   3%|▎         | 130/4000 [00:37<23:06,  2.79it/s]

train:   4%|▎         | 130/3515 [00:35<20:12,  2.79it/s]
overall:   3%|▎         | 131/4000 [00:37<20:44,  3.11it/s]

train:   4%|▎         | 131/3515 [00:35<18:08,  3.11it/s]
overall:   3%|▎         | 132/4000 [00:38<18:38,  3.46it/s]

train:   4%|▍         | 132/3515 [00:35<16:18,  3.46it/s]
overall:   3%|▎         | 133/4000 [00:38<18:31,  3.48it/s]

train:   4%|▍         | 133/3515 [00:35<16:12,  3.48it/s]
overall:   3%|▎         | 134/4000 [00:38<17:06,  3.77it/s]

train:   4%|▍         | 134/3515 [00:36<14:57,  3.77it/s]
overall:   3%|▎         | 135/4000 [00:38<16:52,  3.82it/s]

train:   4%|▍         | 135/3515 [00:36<14:45,  3.82it/s]
overall:   3%|▎         | 136/4000 [00:39<14:47,  4.35it/s]

train:   4%|▍         | 136/3515 [00:36<12:56,  4.35it/s]
overall:   3%|▎         | 137/4000 [00:39<15:35,  4.13it/s]

train:   4%|▍         | 137/3515 [00:36<13:38,  4.13it/s]
overall:   3%|▎         | 138/4000 [00:39<16:08,  3.99it/s]

train:   4%|▍         | 138/3515 [00:36<14:06,  3.99it/s]
overall:   3%|▎         | 139/4000 [00:39<16:36,  3.87it/s]

train:   4%|▍         | 139/3515 [00:37<14:31,  3.87it/s]train loss: 0.7551872134208679
train per: 0.21316277707255152

overall:   4%|▎         | 140/4000 [00:40<16:47,  3.83it/s]

train:   4%|▍         | 140/3515 [00:37<14:41,  3.83it/s]
overall:   4%|▎         | 141/4000 [00:40<17:03,  3.77it/s]

train:   4%|▍         | 141/3515 [00:37<14:54,  3.77it/s]
overall:   4%|▎         | 142/4000 [00:40<17:04,  3.76it/s]

train:   4%|▍         | 142/3515 [00:38<14:55,  3.77it/s]
overall:   4%|▎         | 143/4000 [00:40<16:54,  3.80it/s]

train:   4%|▍         | 143/3515 [00:38<14:47,  3.80it/s]
overall:   4%|▎         | 144/4000 [00:41<17:00,  3.78it/s]

train:   4%|▍         | 144/3515 [00:38<14:51,  3.78it/s]
overall:   4%|▎         | 145/4000 [00:41<17:03,  3.76it/s]

train:   4%|▍         | 145/3515 [00:38<14:55,  3.77it/s]
overall:   4%|▎         | 146/4000 [00:41<17:04,  3.76it/s]

train:   4%|▍         | 146/3515 [00:39<14:55,  3.76it/s]
overall:   4%|▎         | 147/4000 [00:42<17:18,  3.71it/s]

train:   4%|▍         | 147/3515 [00:39<15:08,  3.71it/s]
overall:   4%|▎         | 148/4000 [00:42<17:22,  3.70it/s]

train:   4%|▍         | 148/3515 [00:39<15:10,  3.70it/s]
overall:   4%|▎         | 149/4000 [00:42<17:25,  3.68it/s]

train:   4%|▍         | 149/3515 [00:39<15:13,  3.68it/s]
overall:   4%|▍         | 150/4000 [00:42<17:06,  3.75it/s]

train:   4%|▍         | 150/3515 [00:40<14:56,  3.75it/s]
overall:   4%|▍         | 151/4000 [00:43<17:04,  3.76it/s]

train:   4%|▍         | 151/3515 [00:40<14:55,  3.76it/s]
overall:   4%|▍         | 152/4000 [00:43<16:52,  3.80it/s]

train:   4%|▍         | 152/3515 [00:40<14:44,  3.80it/s]
overall:   4%|▍         | 153/4000 [00:43<17:24,  3.68it/s]

train:   4%|▍         | 153/3515 [00:41<15:13,  3.68it/s]
overall:   4%|▍         | 154/4000 [00:44<23:27,  2.73it/s]

train:   4%|▍         | 154/3515 [00:41<20:30,  2.73it/s]
overall:   4%|▍         | 155/4000 [00:44<21:34,  2.97it/s]

train:   4%|▍         | 155/3515 [00:41<18:51,  2.97it/s]
overall:   4%|▍         | 156/4000 [00:44<18:58,  3.38it/s]

train:   4%|▍         | 156/3515 [00:42<16:34,  3.38it/s]
overall:   4%|▍         | 157/4000 [00:44<17:58,  3.56it/s]

train:   4%|▍         | 157/3515 [00:42<15:42,  3.56it/s]
overall:   4%|▍         | 158/4000 [00:45<17:30,  3.66it/s]

train:   4%|▍         | 158/3515 [00:42<15:18,  3.66it/s]
overall:   4%|▍         | 159/4000 [00:45<17:12,  3.72it/s]

train:   5%|▍         | 159/3515 [00:42<15:01,  3.72it/s]train loss: 0.711365818977356
train per: 0.2024185068349106

overall:   4%|▍         | 160/4000 [00:45<17:26,  3.67it/s]

train:   5%|▍         | 160/3515 [00:43<15:15,  3.67it/s]
overall:   4%|▍         | 161/4000 [00:45<17:23,  3.68it/s]

train:   5%|▍         | 161/3515 [00:43<15:11,  3.68it/s]
overall:   4%|▍         | 162/4000 [00:46<17:14,  3.71it/s]

train:   5%|▍         | 162/3515 [00:43<15:03,  3.71it/s]
overall:   4%|▍         | 163/4000 [00:46<16:57,  3.77it/s]

train:   5%|▍         | 163/3515 [00:43<14:49,  3.77it/s]
overall:   4%|▍         | 164/4000 [00:46<17:03,  3.75it/s]

train:   5%|▍         | 164/3515 [00:44<14:53,  3.75it/s]
overall:   4%|▍         | 165/4000 [00:47<17:04,  3.74it/s]

train:   5%|▍         | 165/3515 [00:44<14:54,  3.74it/s]
overall:   4%|▍         | 166/4000 [00:47<17:09,  3.73it/s]

train:   5%|▍         | 166/3515 [00:44<14:58,  3.73it/s]
overall:   4%|▍         | 167/4000 [00:47<17:16,  3.70it/s]

train:   5%|▍         | 167/3515 [00:44<15:05,  3.70it/s]
overall:   4%|▍         | 168/4000 [00:47<17:17,  3.69it/s]

train:   5%|▍         | 168/3515 [00:45<15:06,  3.69it/s]
overall:   4%|▍         | 169/4000 [00:48<17:13,  3.71it/s]

train:   5%|▍         | 169/3515 [00:45<15:02,  3.71it/s]
overall:   4%|▍         | 170/4000 [00:48<17:10,  3.72it/s]

train:   5%|▍         | 170/3515 [00:45<15:00,  3.72it/s]
overall:   4%|▍         | 171/4000 [00:48<17:01,  3.75it/s]

train:   5%|▍         | 171/3515 [00:46<14:52,  3.75it/s]
overall:   4%|▍         | 172/4000 [00:48<17:08,  3.72it/s]

train:   5%|▍         | 172/3515 [00:46<14:57,  3.72it/s]
overall:   4%|▍         | 173/4000 [00:49<17:00,  3.75it/s]

train:   5%|▍         | 173/3515 [00:46<14:51,  3.75it/s]
overall:   4%|▍         | 174/4000 [00:49<17:08,  3.72it/s]

train:   5%|▍         | 174/3515 [00:46<14:57,  3.72it/s]
overall:   4%|▍         | 175/4000 [00:49<17:00,  3.75it/s]

train:   5%|▍         | 175/3515 [00:47<14:51,  3.75it/s]
overall:   4%|▍         | 176/4000 [00:50<16:59,  3.75it/s]

train:   5%|▌         | 176/3515 [00:47<14:50,  3.75it/s]
overall:   4%|▍         | 177/4000 [00:50<17:15,  3.69it/s]

train:   5%|▌         | 177/3515 [00:47<15:04,  3.69it/s]
overall:   4%|▍         | 178/4000 [00:50<23:21,  2.73it/s]

train:   5%|▌         | 178/3515 [00:48<20:24,  2.73it/s]
overall:   4%|▍         | 179/4000 [00:51<21:32,  2.96it/s]

train:   5%|▌         | 179/3515 [00:48<18:48,  2.96it/s]train loss: 0.7028231024742126
train per: 0.1956062056224542

overall:   4%|▍         | 180/4000 [00:51<20:09,  3.16it/s]

train:   5%|▌         | 180/3515 [00:48<17:36,  3.16it/s]
overall:   5%|▍         | 181/4000 [00:51<19:17,  3.30it/s]

train:   5%|▌         | 181/3515 [00:49<16:50,  3.30it/s]
overall:   5%|▍         | 182/4000 [00:51<18:20,  3.47it/s]

train:   5%|▌         | 182/3515 [00:49<16:00,  3.47it/s]
overall:   5%|▍         | 183/4000 [00:52<17:28,  3.64it/s]

train:   5%|▌         | 183/3515 [00:49<15:15,  3.64it/s]
overall:   5%|▍         | 184/4000 [00:52<17:05,  3.72it/s]

train:   5%|▌         | 184/3515 [00:49<14:55,  3.72it/s]
overall:   5%|▍         | 185/4000 [00:52<16:53,  3.76it/s]

train:   5%|▌         | 185/3515 [00:50<14:44,  3.76it/s]
overall:   5%|▍         | 186/4000 [00:52<16:44,  3.80it/s]

train:   5%|▌         | 186/3515 [00:50<14:36,  3.80it/s]
overall:   5%|▍         | 187/4000 [00:53<16:42,  3.80it/s]

train:   5%|▌         | 187/3515 [00:50<14:35,  3.80it/s]
overall:   5%|▍         | 188/4000 [00:53<16:49,  3.77it/s]

train:   5%|▌         | 188/3515 [00:50<14:41,  3.77it/s]
overall:   5%|▍         | 189/4000 [00:53<16:42,  3.80it/s]

train:   5%|▌         | 189/3515 [00:51<14:34,  3.80it/s]
overall:   5%|▍         | 190/4000 [00:54<16:53,  3.76it/s]

train:   5%|▌         | 190/3515 [00:51<14:44,  3.76it/s]
overall:   5%|▍         | 191/4000 [00:54<16:33,  3.83it/s]

train:   5%|▌         | 191/3515 [00:51<14:27,  3.83it/s]
overall:   5%|▍         | 192/4000 [00:54<16:41,  3.80it/s]

train:   5%|▌         | 192/3515 [00:51<14:33,  3.80it/s]
overall:   5%|▍         | 193/4000 [00:54<16:36,  3.82it/s]

train:   5%|▌         | 193/3515 [00:52<14:29,  3.82it/s]
overall:   5%|▍         | 194/4000 [00:55<16:45,  3.79it/s]

train:   6%|▌         | 194/3515 [00:52<14:37,  3.79it/s]
overall:   5%|▍         | 195/4000 [00:55<16:56,  3.74it/s]

train:   6%|▌         | 195/3515 [00:52<14:46,  3.74it/s]
overall:   5%|▍         | 196/4000 [00:55<14:44,  4.30it/s]

train:   6%|▌         | 196/3515 [00:52<12:51,  4.30it/s]
overall:   5%|▍         | 197/4000 [00:55<15:22,  4.12it/s]

train:   6%|▌         | 197/3515 [00:53<13:24,  4.12it/s]
overall:   5%|▍         | 198/4000 [00:56<15:45,  4.02it/s]

train:   6%|▌         | 198/3515 [00:53<13:44,  4.02it/s]
overall:   5%|▍         | 199/4000 [00:56<15:55,  3.98it/s]

train:   6%|▌         | 199/3515 [00:53<13:53,  3.98it/s]train loss: 0.6521700024604797
train per: 0.1859267734553776

overall:   5%|▌         | 200/4000 [00:56<16:14,  3.90it/s]

train:   6%|▌         | 200/3515 [00:53<14:10,  3.90it/s]
overall:   5%|▌         | 201/4000 [00:56<16:58,  3.73it/s]

train:   6%|▌         | 201/3515 [00:54<14:48,  3.73it/s]
overall:   5%|▌         | 202/4000 [00:57<22:40,  2.79it/s]

train:   6%|▌         | 202/3515 [00:54<19:46,  2.79it/s]
overall:   5%|▌         | 203/4000 [00:57<20:45,  3.05it/s]

train:   6%|▌         | 203/3515 [00:55<18:06,  3.05it/s]
overall:   5%|▌         | 204/4000 [00:57<19:23,  3.26it/s]

train:   6%|▌         | 204/3515 [00:55<16:54,  3.26it/s]
overall:   5%|▌         | 205/4000 [00:58<18:48,  3.36it/s]

train:   6%|▌         | 205/3515 [00:55<16:23,  3.36it/s]
overall:   5%|▌         | 206/4000 [00:58<18:14,  3.47it/s]

train:   6%|▌         | 206/3515 [00:55<15:54,  3.47it/s]
overall:   5%|▌         | 207/4000 [00:58<17:18,  3.65it/s]

train:   6%|▌         | 207/3515 [00:56<15:05,  3.65it/s]
overall:   5%|▌         | 208/4000 [00:58<16:06,  3.92it/s]

train:   6%|▌         | 208/3515 [00:56<14:02,  3.92it/s]
overall:   5%|▌         | 209/4000 [00:59<16:27,  3.84it/s]

train:   6%|▌         | 209/3515 [00:56<14:21,  3.84it/s]
overall:   5%|▌         | 210/4000 [00:59<15:53,  3.97it/s]

train:   6%|▌         | 210/3515 [00:56<13:51,  3.97it/s]
overall:   5%|▌         | 211/4000 [00:59<14:02,  4.50it/s]

train:   6%|▌         | 211/3515 [00:56<12:14,  4.50it/s]
overall:   5%|▌         | 212/4000 [00:59<14:55,  4.23it/s]

train:   6%|▌         | 212/3515 [00:57<13:01,  4.23it/s]
overall:   5%|▌         | 213/4000 [01:00<15:26,  4.09it/s]

train:   6%|▌         | 213/3515 [00:57<13:27,  4.09it/s]
overall:   5%|▌         | 214/4000 [01:00<15:55,  3.96it/s]

train:   6%|▌         | 214/3515 [00:57<13:53,  3.96it/s]
overall:   5%|▌         | 215/4000 [01:00<16:16,  3.88it/s]

train:   6%|▌         | 215/3515 [00:58<14:11,  3.88it/s]
overall:   5%|▌         | 216/4000 [01:00<16:27,  3.83it/s]

train:   6%|▌         | 216/3515 [00:58<14:21,  3.83it/s]
overall:   5%|▌         | 217/4000 [01:01<16:06,  3.92it/s]

train:   6%|▌         | 217/3515 [00:58<14:02,  3.92it/s]
overall:   5%|▌         | 218/4000 [01:01<16:07,  3.91it/s]

train:   6%|▌         | 218/3515 [00:58<14:03,  3.91it/s]
overall:   5%|▌         | 219/4000 [01:01<14:42,  4.28it/s]

train:   6%|▌         | 219/3515 [00:58<12:49,  4.28it/s]train loss: 0.6739255785942078
train per: 0.19066505441354292

overall:   6%|▌         | 220/4000 [01:01<15:33,  4.05it/s]

train:   6%|▋         | 220/3515 [00:59<13:33,  4.05it/s]
overall:   6%|▌         | 221/4000 [01:02<15:43,  4.01it/s]

train:   6%|▋         | 221/3515 [00:59<13:42,  4.01it/s]
overall:   6%|▌         | 222/4000 [01:02<16:15,  3.87it/s]

train:   6%|▋         | 222/3515 [00:59<14:09,  3.87it/s]
overall:   6%|▌         | 223/4000 [01:02<16:28,  3.82it/s]

train:   6%|▋         | 223/3515 [01:00<14:21,  3.82it/s]
overall:   6%|▌         | 224/4000 [01:02<16:42,  3.77it/s]

train:   6%|▋         | 224/3515 [01:00<14:33,  3.77it/s]
overall:   6%|▌         | 225/4000 [01:03<16:28,  3.82it/s]

train:   6%|▋         | 225/3515 [01:00<14:21,  3.82it/s]
overall:   6%|▌         | 226/4000 [01:03<21:24,  2.94it/s]

train:   6%|▋         | 226/3515 [01:01<18:39,  2.94it/s]
overall:   6%|▌         | 227/4000 [01:04<20:12,  3.11it/s]

train:   6%|▋         | 227/3515 [01:01<17:36,  3.11it/s]
overall:   6%|▌         | 228/4000 [01:04<19:12,  3.27it/s]

train:   6%|▋         | 228/3515 [01:01<16:43,  3.27it/s]
overall:   6%|▌         | 229/4000 [01:04<18:26,  3.41it/s]

train:   7%|▋         | 229/3515 [01:01<16:04,  3.41it/s]
overall:   6%|▌         | 230/4000 [01:04<17:51,  3.52it/s]

train:   7%|▋         | 230/3515 [01:02<15:33,  3.52it/s]
overall:   6%|▌         | 231/4000 [01:05<16:37,  3.78it/s]

train:   7%|▋         | 231/3515 [01:02<14:28,  3.78it/s]
overall:   6%|▌         | 232/4000 [01:05<16:41,  3.76it/s]

train:   7%|▋         | 232/3515 [01:02<14:32,  3.76it/s]
overall:   6%|▌         | 233/4000 [01:05<16:42,  3.76it/s]

train:   7%|▋         | 233/3515 [01:02<14:33,  3.76it/s]
overall:   6%|▌         | 234/4000 [01:05<16:49,  3.73it/s]

train:   7%|▋         | 234/3515 [01:03<14:39,  3.73it/s]
overall:   6%|▌         | 235/4000 [01:06<16:20,  3.84it/s]

train:   7%|▋         | 235/3515 [01:03<14:14,  3.84it/s]
overall:   6%|▌         | 237/4000 [01:06<13:41,  4.58it/s]

train:   7%|▋         | 237/3515 [01:03<11:55,  4.58it/s]
overall:   6%|▌         | 238/4000 [01:06<14:29,  4.32it/s]

train:   7%|▋         | 238/3515 [01:04<12:37,  4.32it/s]
overall:   6%|▌         | 239/4000 [01:06<14:41,  4.27it/s]

train:   7%|▋         | 239/3515 [01:04<12:47,  4.27it/s]train loss: 0.6812464594841003
train per: 0.1912321550365719

overall:   6%|▌         | 240/4000 [01:07<15:33,  4.03it/s]

train:   7%|▋         | 240/3515 [01:04<13:34,  4.02it/s]
overall:   6%|▌         | 242/4000 [01:07<13:36,  4.60it/s]

train:   7%|▋         | 242/3515 [01:04<11:50,  4.60it/s]
overall:   6%|▌         | 243/4000 [01:07<14:13,  4.40it/s]

train:   7%|▋         | 243/3515 [01:05<12:23,  4.40it/s]
overall:   6%|▌         | 244/4000 [01:08<14:41,  4.26it/s]

train:   7%|▋         | 244/3515 [01:05<12:47,  4.26it/s]
overall:   6%|▌         | 245/4000 [01:08<14:52,  4.21it/s]

train:   7%|▋         | 245/3515 [01:05<12:57,  4.21it/s]
overall:   6%|▌         | 246/4000 [01:08<15:25,  4.06it/s]

train:   7%|▋         | 246/3515 [01:05<13:25,  4.06it/s]
overall:   6%|▌         | 247/4000 [01:08<15:33,  4.02it/s]

train:   7%|▋         | 247/3515 [01:06<13:33,  4.02it/s]
overall:   6%|▌         | 248/4000 [01:09<14:44,  4.24it/s]

train:   7%|▋         | 248/3515 [01:06<12:50,  4.24it/s]
overall:   6%|▌         | 249/4000 [01:09<15:43,  3.97it/s]

train:   7%|▋         | 249/3515 [01:06<13:41,  3.97it/s]
overall:   6%|▋         | 250/4000 [01:09<20:39,  3.02it/s]

train:   7%|▋         | 250/3515 [01:07<17:59,  3.02it/s]
overall:   6%|▋         | 251/4000 [01:10<17:17,  3.61it/s]

train:   7%|▋         | 251/3515 [01:07<15:02,  3.61it/s]
overall:   6%|▋         | 252/4000 [01:10<16:37,  3.76it/s]

train:   7%|▋         | 252/3515 [01:07<14:28,  3.76it/s]
overall:   6%|▋         | 253/4000 [01:10<16:55,  3.69it/s]

train:   7%|▋         | 253/3515 [01:07<14:43,  3.69it/s]
overall:   6%|▋         | 255/4000 [01:10<14:43,  4.24it/s]

train:   7%|▋         | 255/3515 [01:08<12:48,  4.24it/s]
overall:   6%|▋         | 256/4000 [01:11<15:15,  4.09it/s]

train:   7%|▋         | 256/3515 [01:08<13:17,  4.09it/s]
overall:   6%|▋         | 257/4000 [01:11<15:30,  4.02it/s]

train:   7%|▋         | 257/3515 [01:08<13:29,  4.02it/s]
overall:   6%|▋         | 258/4000 [01:11<14:50,  4.20it/s]

train:   7%|▋         | 258/3515 [01:09<12:54,  4.20it/s]
overall:   6%|▋         | 259/4000 [01:11<15:23,  4.05it/s]

train:   7%|▋         | 259/3515 [01:09<13:24,  4.05it/s]train loss: 0.6615642309188843
train per: 0.1890433486651078

overall:   6%|▋         | 260/4000 [01:12<16:03,  3.88it/s]

train:   7%|▋         | 260/3515 [01:09<13:58,  3.88it/s]
overall:   7%|▋         | 261/4000 [01:12<16:08,  3.86it/s]

train:   7%|▋         | 261/3515 [01:09<14:02,  3.86it/s]
overall:   7%|▋         | 262/4000 [01:12<16:20,  3.81it/s]

train:   7%|▋         | 262/3515 [01:10<14:12,  3.81it/s]
overall:   7%|▋         | 263/4000 [01:13<16:19,  3.82it/s]

train:   7%|▋         | 263/3515 [01:10<14:12,  3.82it/s]
overall:   7%|▋         | 264/4000 [01:13<14:10,  4.40it/s]

train:   8%|▊         | 264/3515 [01:10<12:19,  4.40it/s]
overall:   7%|▋         | 265/4000 [01:13<14:28,  4.30it/s]

train:   8%|▊         | 265/3515 [01:10<12:35,  4.30it/s]
overall:   7%|▋         | 267/4000 [01:13<12:34,  4.95it/s]

train:   8%|▊         | 267/3515 [01:11<10:56,  4.95it/s]
overall:   7%|▋         | 268/4000 [01:14<13:41,  4.54it/s]

train:   8%|▊         | 268/3515 [01:11<11:54,  4.54it/s]
overall:   7%|▋         | 269/4000 [01:14<14:11,  4.38it/s]

train:   8%|▊         | 269/3515 [01:11<12:20,  4.38it/s]
overall:   7%|▋         | 270/4000 [01:14<14:57,  4.16it/s]

train:   8%|▊         | 270/3515 [01:11<13:00,  4.16it/s]
overall:   7%|▋         | 271/4000 [01:14<15:17,  4.06it/s]

train:   8%|▊         | 271/3515 [01:12<13:18,  4.06it/s]
overall:   7%|▋         | 272/4000 [01:15<15:37,  3.98it/s]

train:   8%|▊         | 272/3515 [01:12<13:35,  3.98it/s]
overall:   7%|▋         | 273/4000 [01:15<15:26,  4.02it/s]

train:   8%|▊         | 273/3515 [01:12<13:25,  4.02it/s]
overall:   7%|▋         | 274/4000 [01:15<19:54,  3.12it/s]

train:   8%|▊         | 274/3515 [01:13<17:19,  3.12it/s]
overall:   7%|▋         | 275/4000 [01:16<18:17,  3.39it/s]

train:   8%|▊         | 275/3515 [01:13<15:54,  3.39it/s]
overall:   7%|▋         | 276/4000 [01:16<17:49,  3.48it/s]

train:   8%|▊         | 276/3515 [01:13<15:30,  3.48it/s]
overall:   7%|▋         | 277/4000 [01:16<17:23,  3.57it/s]

train:   8%|▊         | 277/3515 [01:13<15:07,  3.57it/s]
overall:   7%|▋         | 278/4000 [01:16<16:52,  3.67it/s]

train:   8%|▊         | 278/3515 [01:14<14:40,  3.67it/s]
overall:   7%|▋         | 279/4000 [01:17<17:52,  3.47it/s]

train:   8%|▊         | 279/3515 [01:14<15:32,  3.47it/s]train loss: 0.6760408878326416
train per: 0.1931085358018228

overall:   7%|▋         | 280/4000 [01:17<17:29,  3.55it/s]

train:   8%|▊         | 280/3515 [01:14<15:13,  3.54it/s]
overall:   7%|▋         | 281/4000 [01:17<17:08,  3.62it/s]

train:   8%|▊         | 281/3515 [01:15<14:53,  3.62it/s]
overall:   7%|▋         | 282/4000 [01:17<16:58,  3.65it/s]

train:   8%|▊         | 282/3515 [01:15<14:45,  3.65it/s]
overall:   7%|▋         | 283/4000 [01:18<16:39,  3.72it/s]

train:   8%|▊         | 283/3515 [01:15<14:29,  3.72it/s]
overall:   7%|▋         | 284/4000 [01:18<16:04,  3.85it/s]

train:   8%|▊         | 284/3515 [01:15<13:58,  3.85it/s]
overall:   7%|▋         | 285/4000 [01:18<15:59,  3.87it/s]

train:   8%|▊         | 285/3515 [01:16<13:54,  3.87it/s]
overall:   7%|▋         | 286/4000 [01:18<16:10,  3.83it/s]

train:   8%|▊         | 286/3515 [01:16<14:03,  3.83it/s]
overall:   7%|▋         | 287/4000 [01:19<16:03,  3.85it/s]

train:   8%|▊         | 287/3515 [01:16<13:57,  3.85it/s]
overall:   7%|▋         | 288/4000 [01:19<16:08,  3.83it/s]

train:   8%|▊         | 288/3515 [01:16<14:02,  3.83it/s]
overall:   7%|▋         | 289/4000 [01:19<15:30,  3.99it/s]

train:   8%|▊         | 289/3515 [01:17<13:28,  3.99it/s]
overall:   7%|▋         | 290/4000 [01:19<15:54,  3.89it/s]

train:   8%|▊         | 290/3515 [01:17<13:49,  3.89it/s]
overall:   7%|▋         | 291/4000 [01:20<13:57,  4.43it/s]

train:   8%|▊         | 291/3515 [01:17<12:08,  4.43it/s]
overall:   7%|▋         | 292/4000 [01:20<12:32,  4.93it/s]

train:   8%|▊         | 292/3515 [01:17<10:54,  4.93it/s]
overall:   7%|▋         | 293/4000 [01:20<10:39,  5.79it/s]

train:   8%|▊         | 293/3515 [01:17<09:16,  5.79it/s]
overall:   7%|▋         | 294/4000 [01:20<12:27,  4.96it/s]

train:   8%|▊         | 294/3515 [01:18<10:49,  4.96it/s]
overall:   7%|▋         | 295/4000 [01:20<13:11,  4.68it/s]

train:   8%|▊         | 295/3515 [01:18<11:27,  4.68it/s]
overall:   7%|▋         | 296/4000 [01:21<14:28,  4.26it/s]

train:   8%|▊         | 296/3515 [01:18<12:35,  4.26it/s]
overall:   7%|▋         | 298/4000 [01:21<15:14,  4.05it/s]

train:   8%|▊         | 298/3515 [01:19<13:14,  4.05it/s]
overall:   7%|▋         | 299/4000 [01:21<15:01,  4.10it/s]

train:   9%|▊         | 299/3515 [01:19<13:03,  4.10it/s]train loss: 0.7003909349441528
train per: 0.193797620638952

overall:   8%|▊         | 300/4000 [01:22<15:32,  3.97it/s]

train:   9%|▊         | 300/3515 [01:19<13:30,  3.96it/s]
overall:   8%|▊         | 301/4000 [01:22<15:59,  3.85it/s]

train:   9%|▊         | 301/3515 [01:19<13:53,  3.86it/s]
overall:   8%|▊         | 302/4000 [01:22<15:03,  4.09it/s]

train:   9%|▊         | 302/3515 [01:20<13:04,  4.10it/s]
overall:   8%|▊         | 303/4000 [01:23<16:47,  3.67it/s]

train:   9%|▊         | 303/3515 [01:20<14:34,  3.67it/s]
overall:   8%|▊         | 304/4000 [01:23<16:34,  3.72it/s]

train:   9%|▊         | 304/3515 [01:20<14:23,  3.72it/s]
overall:   8%|▊         | 305/4000 [01:23<16:15,  3.79it/s]

train:   9%|▊         | 305/3515 [01:20<14:07,  3.79it/s]
overall:   8%|▊         | 306/4000 [01:23<16:12,  3.80it/s]

train:   9%|▊         | 306/3515 [01:21<14:04,  3.80it/s]
overall:   8%|▊         | 307/4000 [01:23<13:19,  4.62it/s]

train:   9%|▊         | 307/3515 [01:21<11:34,  4.62it/s]
overall:   8%|▊         | 308/4000 [01:24<14:20,  4.29it/s]

train:   9%|▉         | 308/3515 [01:21<12:27,  4.29it/s]
overall:   8%|▊         | 309/4000 [01:24<12:11,  5.04it/s]

train:   9%|▉         | 309/3515 [01:21<10:35,  5.05it/s]
overall:   8%|▊         | 310/4000 [01:24<12:57,  4.74it/s]

train:   9%|▉         | 310/3515 [01:21<11:15,  4.74it/s]
overall:   8%|▊         | 311/4000 [01:24<13:36,  4.52it/s]

train:   9%|▉         | 311/3515 [01:22<11:49,  4.52it/s]
overall:   8%|▊         | 312/4000 [01:25<14:33,  4.22it/s]

train:   9%|▉         | 312/3515 [01:22<12:38,  4.22it/s]
overall:   8%|▊         | 313/4000 [01:25<14:04,  4.37it/s]

train:   9%|▉         | 313/3515 [01:22<12:13,  4.37it/s]
overall:   8%|▊         | 314/4000 [01:25<14:44,  4.17it/s]

train:   9%|▉         | 314/3515 [01:22<12:48,  4.17it/s]
overall:   8%|▊         | 315/4000 [01:25<15:06,  4.07it/s]

train:   9%|▉         | 315/3515 [01:23<13:06,  4.07it/s]
overall:   8%|▊         | 316/4000 [01:26<15:10,  4.05it/s]

train:   9%|▉         | 316/3515 [01:23<13:10,  4.05it/s]
overall:   8%|▊         | 317/4000 [01:26<13:52,  4.43it/s]

train:   9%|▉         | 317/3515 [01:23<12:02,  4.43it/s]
overall:   8%|▊         | 318/4000 [01:26<14:45,  4.16it/s]

train:   9%|▉         | 318/3515 [01:23<12:49,  4.16it/s]
overall:   8%|▊         | 319/4000 [01:26<15:22,  3.99it/s]

train:   9%|▉         | 319/3515 [01:24<13:21,  3.99it/s]train loss: 0.7119651436805725
train per: 0.1998917693140236

overall:   8%|▊         | 320/4000 [01:27<15:40,  3.91it/s]

train:   9%|▉         | 320/3515 [01:24<13:37,  3.91it/s]
overall:   8%|▊         | 321/4000 [01:27<15:53,  3.86it/s]

train:   9%|▉         | 321/3515 [01:24<13:47,  3.86it/s]
overall:   8%|▊         | 322/4000 [01:27<20:27,  3.00it/s]

train:   9%|▉         | 322/3515 [01:25<17:45,  3.00it/s]
overall:   8%|▊         | 323/4000 [01:28<19:04,  3.21it/s]

train:   9%|▉         | 323/3515 [01:25<16:33,  3.21it/s]
overall:   8%|▊         | 324/4000 [01:28<18:16,  3.35it/s]

train:   9%|▉         | 324/3515 [01:25<15:52,  3.35it/s]
overall:   8%|▊         | 325/4000 [01:28<17:56,  3.41it/s]

train:   9%|▉         | 325/3515 [01:26<15:34,  3.41it/s]
overall:   8%|▊         | 326/4000 [01:28<17:32,  3.49it/s]

train:   9%|▉         | 326/3515 [01:26<15:13,  3.49it/s]
overall:   8%|▊         | 327/4000 [01:29<17:43,  3.45it/s]

train:   9%|▉         | 327/3515 [01:26<15:22,  3.45it/s]
overall:   8%|▊         | 328/4000 [01:29<17:12,  3.56it/s]

train:   9%|▉         | 328/3515 [01:26<14:55,  3.56it/s]
overall:   8%|▊         | 329/4000 [01:29<16:39,  3.67it/s]

train:   9%|▉         | 329/3515 [01:27<14:27,  3.67it/s]
overall:   8%|▊         | 330/4000 [01:29<14:26,  4.23it/s]

train:   9%|▉         | 330/3515 [01:27<12:32,  4.23it/s]
overall:   8%|▊         | 331/4000 [01:30<15:03,  4.06it/s]

train:   9%|▉         | 331/3515 [01:27<13:03,  4.06it/s]
overall:   8%|▊         | 332/4000 [01:30<15:17,  4.00it/s]

train:   9%|▉         | 332/3515 [01:27<13:16,  4.00it/s]
overall:   8%|▊         | 333/4000 [01:30<15:18,  3.99it/s]

train:   9%|▉         | 333/3515 [01:28<13:16,  3.99it/s]
overall:   8%|▊         | 334/4000 [01:30<15:08,  4.04it/s]

train:  10%|▉         | 334/3515 [01:28<13:08,  4.04it/s]
overall:   8%|▊         | 335/4000 [01:31<15:24,  3.96it/s]

train:  10%|▉         | 335/3515 [01:28<13:22,  3.96it/s]
overall:   8%|▊         | 336/4000 [01:31<14:46,  4.13it/s]

train:  10%|▉         | 336/3515 [01:28<12:49,  4.13it/s]
overall:   8%|▊         | 337/4000 [01:31<15:28,  3.94it/s]

train:  10%|▉         | 337/3515 [01:29<13:25,  3.95it/s]
overall:   8%|▊         | 338/4000 [01:31<15:48,  3.86it/s]

train:  10%|▉         | 338/3515 [01:29<13:43,  3.86it/s]
overall:   8%|▊         | 339/4000 [01:32<15:32,  3.93it/s]

train:  10%|▉         | 339/3515 [01:29<13:29,  3.93it/s]train loss: 0.6389274597167969
train per: 0.17952687339751341

overall:   8%|▊         | 340/4000 [01:32<15:36,  3.91it/s]

train:  10%|▉         | 340/3515 [01:29<13:33,  3.91it/s]
overall:   9%|▊         | 341/4000 [01:32<15:50,  3.85it/s]

train:  10%|▉         | 341/3515 [01:30<13:44,  3.85it/s]
overall:   9%|▊         | 342/4000 [01:32<15:58,  3.82it/s]

train:  10%|▉         | 342/3515 [01:30<13:51,  3.82it/s]
overall:   9%|▊         | 343/4000 [01:33<15:53,  3.83it/s]

train:  10%|▉         | 343/3515 [01:30<13:47,  3.84it/s]
overall:   9%|▊         | 344/4000 [01:33<15:01,  4.06it/s]

train:  10%|▉         | 344/3515 [01:30<13:01,  4.06it/s]
overall:   9%|▊         | 345/4000 [01:33<15:13,  4.00it/s]

train:  10%|▉         | 345/3515 [01:31<13:12,  4.00it/s]
overall:   9%|▊         | 346/4000 [01:34<19:10,  3.18it/s]

train:  10%|▉         | 346/3515 [01:31<16:37,  3.18it/s]
overall:   9%|▊         | 347/4000 [01:34<16:08,  3.77it/s]

train:  10%|▉         | 347/3515 [01:31<13:59,  3.77it/s]
overall:   9%|▊         | 348/4000 [01:34<16:05,  3.78it/s]

train:  10%|▉         | 348/3515 [01:31<13:57,  3.78it/s]
overall:   9%|▊         | 349/4000 [01:34<16:26,  3.70it/s]

train:  10%|▉         | 349/3515 [01:32<14:15,  3.70it/s]
overall:   9%|▉         | 350/4000 [01:35<16:34,  3.67it/s]

train:  10%|▉         | 350/3515 [01:32<14:22,  3.67it/s]
overall:   9%|▉         | 351/4000 [01:35<16:43,  3.64it/s]

train:  10%|▉         | 351/3515 [01:32<14:30,  3.64it/s]
overall:   9%|▉         | 352/4000 [01:35<16:30,  3.68it/s]

train:  10%|█         | 352/3515 [01:33<14:18,  3.68it/s]
overall:   9%|▉         | 353/4000 [01:35<16:14,  3.74it/s]

train:  10%|█         | 353/3515 [01:33<14:04,  3.74it/s]
overall:   9%|▉         | 354/4000 [01:36<16:25,  3.70it/s]

train:  10%|█         | 354/3515 [01:33<14:14,  3.70it/s]
overall:   9%|▉         | 355/4000 [01:36<13:21,  4.55it/s]

train:  10%|█         | 355/3515 [01:33<11:34,  4.55it/s]
overall:   9%|▉         | 356/4000 [01:36<13:59,  4.34it/s]

train:  10%|█         | 356/3515 [01:33<12:07,  4.34it/s]
overall:   9%|▉         | 357/4000 [01:36<14:11,  4.28it/s]

train:  10%|█         | 357/3515 [01:34<12:17,  4.28it/s]
overall:   9%|▉         | 358/4000 [01:37<14:54,  4.07it/s]

train:  10%|█         | 358/3515 [01:34<12:55,  4.07it/s]
overall:   9%|▉         | 359/4000 [01:37<15:25,  3.93it/s]

train:  10%|█         | 359/3515 [01:34<13:22,  3.93it/s]train loss: 0.6760891675949097
train per: 0.19128867379441467

overall:   9%|▉         | 360/4000 [01:37<15:22,  3.95it/s]

train:  10%|█         | 360/3515 [01:34<13:20,  3.94it/s]
overall:   9%|▉         | 361/4000 [01:37<15:39,  3.87it/s]

train:  10%|█         | 361/3515 [01:35<13:34,  3.87it/s]
overall:   9%|▉         | 362/4000 [01:38<14:04,  4.31it/s]

train:  10%|█         | 362/3515 [01:35<12:12,  4.31it/s]
overall:   9%|▉         | 363/4000 [01:38<14:51,  4.08it/s]

train:  10%|█         | 363/3515 [01:35<12:52,  4.08it/s]
overall:   9%|▉         | 364/4000 [01:38<15:28,  3.92it/s]

train:  10%|█         | 364/3515 [01:35<13:24,  3.92it/s]
overall:   9%|▉         | 365/4000 [01:38<15:36,  3.88it/s]

train:  10%|█         | 365/3515 [01:36<13:31,  3.88it/s]
overall:   9%|▉         | 366/4000 [01:39<15:49,  3.83it/s]

train:  10%|█         | 366/3515 [01:36<13:42,  3.83it/s]
overall:   9%|▉         | 367/4000 [01:39<15:55,  3.80it/s]

train:  10%|█         | 367/3515 [01:36<13:47,  3.80it/s]
overall:   9%|▉         | 368/4000 [01:39<15:47,  3.83it/s]

train:  10%|█         | 368/3515 [01:37<13:41,  3.83it/s]
overall:   9%|▉         | 369/4000 [01:39<15:45,  3.84it/s]

train:  10%|█         | 369/3515 [01:37<13:38,  3.84it/s]
overall:   9%|▉         | 370/4000 [01:40<20:23,  2.97it/s]

train:  11%|█         | 370/3515 [01:37<17:39,  2.97it/s]
overall:   9%|▉         | 371/4000 [01:40<17:58,  3.37it/s]

train:  11%|█         | 371/3515 [01:38<15:34,  3.37it/s]
overall:   9%|▉         | 372/4000 [01:40<16:57,  3.57it/s]

train:  11%|█         | 372/3515 [01:38<14:41,  3.57it/s]
overall:   9%|▉         | 373/4000 [01:41<16:48,  3.60it/s]

train:  11%|█         | 373/3515 [01:38<14:33,  3.60it/s]
overall:   9%|▉         | 374/4000 [01:41<15:32,  3.89it/s]

train:  11%|█         | 374/3515 [01:38<13:28,  3.89it/s]
overall:   9%|▉         | 375/4000 [01:41<17:07,  3.53it/s]

train:  11%|█         | 375/3515 [01:39<14:50,  3.53it/s]
overall:   9%|▉         | 376/4000 [01:41<14:38,  4.13it/s]

train:  11%|█         | 376/3515 [01:39<12:40,  4.13it/s]
overall:   9%|▉         | 377/4000 [01:42<15:05,  4.00it/s]

train:  11%|█         | 377/3515 [01:39<13:04,  4.00it/s]
overall:   9%|▉         | 378/4000 [01:42<12:28,  4.84it/s]

train:  11%|█         | 378/3515 [01:39<10:48,  4.84it/s]
overall:   9%|▉         | 379/4000 [01:42<13:52,  4.35it/s]

train:  11%|█         | 379/3515 [01:39<12:01,  4.35it/s]train loss: 0.6388362646102905
train per: 0.17815628656871668

overall:  10%|▉         | 380/4000 [01:42<14:45,  4.09it/s]

train:  11%|█         | 380/3515 [01:40<12:47,  4.08it/s]
overall:  10%|▉         | 381/4000 [01:43<15:05,  4.00it/s]

train:  11%|█         | 381/3515 [01:40<13:04,  4.00it/s]
overall:  10%|▉         | 382/4000 [01:43<15:27,  3.90it/s]

train:  11%|█         | 382/3515 [01:40<13:22,  3.90it/s]
overall:  10%|▉         | 383/4000 [01:43<15:45,  3.82it/s]

train:  11%|█         | 383/3515 [01:40<13:39,  3.82it/s]
overall:  10%|▉         | 384/4000 [01:43<15:53,  3.79it/s]

train:  11%|█         | 384/3515 [01:41<13:45,  3.79it/s]
overall:  10%|▉         | 385/4000 [01:44<15:35,  3.86it/s]

train:  11%|█         | 385/3515 [01:41<13:30,  3.86it/s]
overall:  10%|▉         | 386/4000 [01:44<15:40,  3.84it/s]

train:  11%|█         | 386/3515 [01:41<13:33,  3.84it/s]
overall:  10%|▉         | 387/4000 [01:44<15:22,  3.92it/s]

train:  11%|█         | 387/3515 [01:41<13:18,  3.92it/s]
overall:  10%|▉         | 388/4000 [01:44<15:31,  3.88it/s]

train:  11%|█         | 388/3515 [01:42<13:26,  3.88it/s]
overall:  10%|▉         | 389/4000 [01:45<15:29,  3.88it/s]

train:  11%|█         | 389/3515 [01:42<13:24,  3.88it/s]
overall:  10%|▉         | 390/4000 [01:45<15:31,  3.88it/s]

train:  11%|█         | 390/3515 [01:42<13:25,  3.88it/s]
overall:  10%|▉         | 391/4000 [01:45<15:33,  3.87it/s]

train:  11%|█         | 391/3515 [01:43<13:28,  3.87it/s]
overall:  10%|▉         | 392/4000 [01:45<15:35,  3.86it/s]

train:  11%|█         | 392/3515 [01:43<13:30,  3.86it/s]
overall:  10%|▉         | 393/4000 [01:46<15:54,  3.78it/s]

train:  11%|█         | 393/3515 [01:43<13:46,  3.78it/s]
overall:  10%|▉         | 394/4000 [01:46<20:01,  3.00it/s]

train:  11%|█         | 394/3515 [01:44<17:19,  3.00it/s]
overall:  10%|▉         | 395/4000 [01:46<18:36,  3.23it/s]

train:  11%|█         | 395/3515 [01:44<16:06,  3.23it/s]
overall:  10%|▉         | 396/4000 [01:47<16:38,  3.61it/s]

train:  11%|█▏        | 396/3515 [01:44<14:24,  3.61it/s]
overall:  10%|▉         | 397/4000 [01:47<16:18,  3.68it/s]

train:  11%|█▏        | 397/3515 [01:44<14:06,  3.68it/s]
overall:  10%|▉         | 398/4000 [01:47<15:32,  3.86it/s]

train:  11%|█▏        | 398/3515 [01:45<13:26,  3.86it/s]
overall:  10%|▉         | 399/4000 [01:47<16:50,  3.56it/s]

train:  11%|█▏        | 399/3515 [01:45<14:34,  3.56it/s]train loss: 0.639264702796936
train per: 0.17582714050965098

overall:  10%|█         | 400/4000 [01:48<16:18,  3.68it/s]

train:  11%|█▏        | 400/3515 [01:45<14:07,  3.67it/s]
overall:  10%|█         | 401/4000 [01:48<15:00,  3.99it/s]

train:  11%|█▏        | 401/3515 [01:45<12:59,  4.00it/s]
overall:  10%|█         | 402/4000 [01:48<15:19,  3.91it/s]

train:  11%|█▏        | 402/3515 [01:46<13:15,  3.91it/s]
overall:  10%|█         | 403/4000 [01:48<15:23,  3.90it/s]

train:  11%|█▏        | 403/3515 [01:46<13:18,  3.90it/s]
overall:  10%|█         | 404/4000 [01:49<15:40,  3.82it/s]

train:  11%|█▏        | 404/3515 [01:46<13:33,  3.82it/s]
overall:  10%|█         | 405/4000 [01:49<15:48,  3.79it/s]

train:  12%|█▏        | 405/3515 [01:46<13:40,  3.79it/s]
overall:  10%|█         | 406/4000 [01:49<15:56,  3.76it/s]

train:  12%|█▏        | 406/3515 [01:47<13:47,  3.76it/s]
overall:  10%|█         | 407/4000 [01:50<16:10,  3.70it/s]

train:  12%|█▏        | 407/3515 [01:47<13:59,  3.70it/s]
overall:  10%|█         | 408/4000 [01:50<16:10,  3.70it/s]

train:  12%|█▏        | 408/3515 [01:47<13:59,  3.70it/s]
overall:  10%|█         | 409/4000 [01:50<15:43,  3.81it/s]

train:  12%|█▏        | 409/3515 [01:47<13:35,  3.81it/s]
overall:  10%|█         | 410/4000 [01:50<16:02,  3.73it/s]

train:  12%|█▏        | 410/3515 [01:48<13:52,  3.73it/s]
overall:  10%|█         | 411/4000 [01:51<14:24,  4.15it/s]

train:  12%|█▏        | 411/3515 [01:48<12:27,  4.15it/s]
overall:  10%|█         | 412/4000 [01:51<14:49,  4.03it/s]

train:  12%|█▏        | 412/3515 [01:48<12:49,  4.03it/s]
overall:  10%|█         | 414/4000 [01:51<12:57,  4.61it/s]

train:  12%|█▏        | 414/3515 [01:49<11:12,  4.61it/s]
overall:  10%|█         | 415/4000 [01:51<13:22,  4.47it/s]

train:  12%|█▏        | 415/3515 [01:49<11:33,  4.47it/s]
overall:  10%|█         | 416/4000 [01:52<13:51,  4.31it/s]

train:  12%|█▏        | 416/3515 [01:49<11:58,  4.31it/s]
overall:  10%|█         | 417/4000 [01:52<14:22,  4.15it/s]

train:  12%|█▏        | 417/3515 [01:49<12:26,  4.15it/s]
overall:  10%|█         | 418/4000 [01:52<18:53,  3.16it/s]

train:  12%|█▏        | 418/3515 [01:50<16:20,  3.16it/s]
overall:  10%|█         | 419/4000 [01:53<17:56,  3.33it/s]

train:  12%|█▏        | 419/3515 [01:50<15:30,  3.33it/s]train loss: 0.6552298069000244
train per: 0.18236683285226973

overall:  10%|█         | 420/4000 [01:53<17:14,  3.46it/s]

train:  12%|█▏        | 420/3515 [01:50<14:54,  3.46it/s]
overall:  11%|█         | 421/4000 [01:53<16:51,  3.54it/s]

train:  12%|█▏        | 421/3515 [01:51<14:34,  3.54it/s]
overall:  11%|█         | 422/4000 [01:53<16:27,  3.62it/s]

train:  12%|█▏        | 422/3515 [01:51<14:13,  3.62it/s]
overall:  11%|█         | 423/4000 [01:54<17:23,  3.43it/s]

train:  12%|█▏        | 423/3515 [01:51<15:02,  3.43it/s]
overall:  11%|█         | 424/4000 [01:54<17:00,  3.50it/s]

train:  12%|█▏        | 424/3515 [01:51<14:42,  3.50it/s]
overall:  11%|█         | 425/4000 [01:54<16:42,  3.57it/s]

train:  12%|█▏        | 425/3515 [01:52<14:26,  3.57it/s]
overall:  11%|█         | 426/4000 [01:55<16:58,  3.51it/s]

train:  12%|█▏        | 426/3515 [01:52<14:39,  3.51it/s]
overall:  11%|█         | 427/4000 [01:55<16:49,  3.54it/s]

train:  12%|█▏        | 427/3515 [01:52<14:32,  3.54it/s]
overall:  11%|█         | 428/4000 [01:55<16:05,  3.70it/s]

train:  12%|█▏        | 428/3515 [01:53<13:54,  3.70it/s]
overall:  11%|█         | 429/4000 [01:55<16:02,  3.71it/s]

train:  12%|█▏        | 429/3515 [01:53<13:52,  3.71it/s]
overall:  11%|█         | 430/4000 [01:56<15:46,  3.77it/s]

train:  12%|█▏        | 430/3515 [01:53<13:38,  3.77it/s]
overall:  11%|█         | 431/4000 [01:56<15:51,  3.75it/s]

train:  12%|█▏        | 431/3515 [01:53<13:42,  3.75it/s]
overall:  11%|█         | 432/4000 [01:56<13:44,  4.33it/s]

train:  12%|█▏        | 432/3515 [01:53<11:52,  4.33it/s]
overall:  11%|█         | 433/4000 [01:56<11:56,  4.98it/s]

train:  12%|█▏        | 433/3515 [01:54<10:18,  4.98it/s]
overall:  11%|█         | 434/4000 [01:56<13:14,  4.49it/s]

train:  12%|█▏        | 434/3515 [01:54<11:26,  4.49it/s]
overall:  11%|█         | 435/4000 [01:57<14:04,  4.22it/s]

train:  12%|█▏        | 435/3515 [01:54<12:09,  4.22it/s]
overall:  11%|█         | 436/4000 [01:57<14:16,  4.16it/s]

train:  12%|█▏        | 436/3515 [01:54<12:19,  4.16it/s]
overall:  11%|█         | 437/4000 [01:57<14:52,  3.99it/s]

train:  12%|█▏        | 437/3515 [01:55<12:50,  3.99it/s]
overall:  11%|█         | 438/4000 [01:58<14:51,  4.00it/s]

train:  12%|█▏        | 438/3515 [01:55<12:50,  4.00it/s]
overall:  11%|█         | 439/4000 [01:58<14:21,  4.14it/s]

train:  12%|█▏        | 439/3515 [01:55<12:23,  4.14it/s]train loss: 0.6406494379043579
train per: 0.18259397537277092

overall:  11%|█         | 440/4000 [01:58<15:01,  3.95it/s]

train:  13%|█▎        | 440/3515 [01:55<12:59,  3.95it/s]
overall:  11%|█         | 441/4000 [01:58<15:33,  3.81it/s]

train:  13%|█▎        | 441/3515 [01:56<13:26,  3.81it/s]
overall:  11%|█         | 442/4000 [01:59<20:05,  2.95it/s]

train:  13%|█▎        | 442/3515 [01:56<17:21,  2.95it/s]
overall:  11%|█         | 443/4000 [01:59<18:34,  3.19it/s]

train:  13%|█▎        | 443/3515 [01:56<16:02,  3.19it/s]
overall:  11%|█         | 444/4000 [01:59<17:13,  3.44it/s]

train:  13%|█▎        | 444/3515 [01:57<14:52,  3.44it/s]
overall:  11%|█         | 445/4000 [02:00<16:48,  3.53it/s]

train:  13%|█▎        | 445/3515 [01:57<14:30,  3.53it/s]
overall:  11%|█         | 446/4000 [02:00<16:29,  3.59it/s]

train:  13%|█▎        | 446/3515 [01:57<14:14,  3.59it/s]
overall:  11%|█         | 447/4000 [02:00<17:15,  3.43it/s]

train:  13%|█▎        | 447/3515 [01:58<14:54,  3.43it/s]
overall:  11%|█         | 448/4000 [02:00<16:19,  3.62it/s]

train:  13%|█▎        | 448/3515 [01:58<14:06,  3.62it/s]
overall:  11%|█         | 449/4000 [02:01<15:59,  3.70it/s]

train:  13%|█▎        | 449/3515 [01:58<13:48,  3.70it/s]
overall:  11%|█▏        | 450/4000 [02:01<16:04,  3.68it/s]

train:  13%|█▎        | 450/3515 [01:58<13:52,  3.68it/s]
overall:  11%|█▏        | 451/4000 [02:01<15:50,  3.73it/s]

train:  13%|█▎        | 451/3515 [01:59<13:40,  3.73it/s]
overall:  11%|█▏        | 452/4000 [02:01<15:56,  3.71it/s]

train:  13%|█▎        | 452/3515 [01:59<13:46,  3.71it/s]
overall:  11%|█▏        | 453/4000 [02:02<15:27,  3.82it/s]

train:  13%|█▎        | 453/3515 [01:59<13:20,  3.82it/s]
overall:  11%|█▏        | 454/4000 [02:02<15:08,  3.90it/s]

train:  13%|█▎        | 454/3515 [01:59<13:04,  3.90it/s]
overall:  11%|█▏        | 455/4000 [02:02<15:08,  3.90it/s]

train:  13%|█▎        | 455/3515 [02:00<13:04,  3.90it/s]
overall:  11%|█▏        | 456/4000 [02:02<15:17,  3.86it/s]

train:  13%|█▎        | 456/3515 [02:00<13:11,  3.86it/s]
overall:  11%|█▏        | 457/4000 [02:03<13:51,  4.26it/s]

train:  13%|█▎        | 457/3515 [02:00<11:57,  4.26it/s]
overall:  11%|█▏        | 458/4000 [02:03<13:58,  4.22it/s]

train:  13%|█▎        | 458/3515 [02:00<12:03,  4.22it/s]
overall:  11%|█▏        | 459/4000 [02:03<14:18,  4.12it/s]

train:  13%|█▎        | 459/3515 [02:01<12:21,  4.12it/s]train loss: 0.6367546319961548
train per: 0.18363066597576924

overall:  12%|█▏        | 460/4000 [02:03<14:44,  4.00it/s]

train:  13%|█▎        | 460/3515 [02:01<12:44,  4.00it/s]
overall:  12%|█▏        | 461/4000 [02:04<15:19,  3.85it/s]

train:  13%|█▎        | 461/3515 [02:01<13:13,  3.85it/s]
overall:  12%|█▏        | 462/4000 [02:04<15:37,  3.77it/s]

train:  13%|█▎        | 462/3515 [02:01<13:28,  3.77it/s]
overall:  12%|█▏        | 463/4000 [02:04<15:49,  3.72it/s]

train:  13%|█▎        | 463/3515 [02:02<13:39,  3.73it/s]
overall:  12%|█▏        | 464/4000 [02:05<15:49,  3.73it/s]

train:  13%|█▎        | 464/3515 [02:02<13:38,  3.73it/s]
overall:  12%|█▏        | 465/4000 [02:05<14:42,  4.01it/s]

train:  13%|█▎        | 465/3515 [02:02<12:41,  4.01it/s]
overall:  12%|█▏        | 466/4000 [02:05<19:20,  3.05it/s]

train:  13%|█▎        | 466/3515 [02:03<16:41,  3.05it/s]
overall:  12%|█▏        | 467/4000 [02:06<18:10,  3.24it/s]

train:  13%|█▎        | 467/3515 [02:03<15:40,  3.24it/s]
overall:  12%|█▏        | 468/4000 [02:06<17:14,  3.41it/s]

train:  13%|█▎        | 468/3515 [02:03<14:52,  3.41it/s]
overall:  12%|█▏        | 469/4000 [02:06<16:25,  3.58it/s]

train:  13%|█▎        | 469/3515 [02:03<14:10,  3.58it/s]
overall:  12%|█▏        | 470/4000 [02:06<16:11,  3.63it/s]

train:  13%|█▎        | 470/3515 [02:04<13:58,  3.63it/s]
overall:  12%|█▏        | 471/4000 [02:07<17:20,  3.39it/s]

train:  13%|█▎        | 471/3515 [02:04<14:57,  3.39it/s]
overall:  12%|█▏        | 472/4000 [02:07<16:39,  3.53it/s]

train:  13%|█▎        | 472/3515 [02:04<14:21,  3.53it/s]
overall:  12%|█▏        | 473/4000 [02:07<16:09,  3.64it/s]

train:  13%|█▎        | 473/3515 [02:05<13:56,  3.64it/s]
overall:  12%|█▏        | 474/4000 [02:07<16:16,  3.61it/s]

train:  13%|█▎        | 474/3515 [02:05<14:02,  3.61it/s]
overall:  12%|█▏        | 475/4000 [02:08<16:04,  3.65it/s]

train:  14%|█▎        | 475/3515 [02:05<13:51,  3.66it/s]
overall:  12%|█▏        | 476/4000 [02:08<15:31,  3.78it/s]

train:  14%|█▎        | 476/3515 [02:05<13:23,  3.78it/s]
overall:  12%|█▏        | 477/4000 [02:08<15:48,  3.71it/s]

train:  14%|█▎        | 477/3515 [02:06<13:38,  3.71it/s]
overall:  12%|█▏        | 478/4000 [02:08<15:39,  3.75it/s]

train:  14%|█▎        | 478/3515 [02:06<13:29,  3.75it/s]
overall:  12%|█▏        | 479/4000 [02:09<15:49,  3.71it/s]

train:  14%|█▎        | 479/3515 [02:06<13:38,  3.71it/s]train loss: 0.6208993196487427
train per: 0.18169615892261917

overall:  12%|█▏        | 480/4000 [02:09<15:27,  3.80it/s]

train:  14%|█▎        | 480/3515 [02:06<13:20,  3.79it/s]
overall:  12%|█▏        | 481/4000 [02:09<15:33,  3.77it/s]

train:  14%|█▎        | 481/3515 [02:07<13:24,  3.77it/s]
overall:  12%|█▏        | 482/4000 [02:10<15:28,  3.79it/s]

train:  14%|█▎        | 482/3515 [02:07<13:20,  3.79it/s]
overall:  12%|█▏        | 483/4000 [02:10<15:39,  3.75it/s]

train:  14%|█▎        | 483/3515 [02:07<13:29,  3.75it/s]
overall:  12%|█▏        | 484/4000 [02:10<14:01,  4.18it/s]

train:  14%|█▍        | 484/3515 [02:07<12:04,  4.18it/s]
overall:  12%|█▏        | 485/4000 [02:10<11:46,  4.97it/s]

train:  14%|█▍        | 485/3515 [02:07<10:09,  4.97it/s]
overall:  12%|█▏        | 486/4000 [02:10<13:04,  4.48it/s]

train:  14%|█▍        | 486/3515 [02:08<11:16,  4.48it/s]
overall:  12%|█▏        | 487/4000 [02:11<13:55,  4.20it/s]

train:  14%|█▍        | 487/3515 [02:08<12:00,  4.20it/s]
overall:  12%|█▏        | 488/4000 [02:11<14:23,  4.07it/s]

train:  14%|█▍        | 488/3515 [02:08<12:24,  4.07it/s]
overall:  12%|█▏        | 489/4000 [02:11<14:41,  3.99it/s]

train:  14%|█▍        | 489/3515 [02:09<12:39,  3.99it/s]
overall:  12%|█▏        | 490/4000 [02:12<19:07,  3.06it/s]

train:  14%|█▍        | 490/3515 [02:09<16:29,  3.06it/s]
overall:  12%|█▏        | 491/4000 [02:12<17:41,  3.31it/s]

train:  14%|█▍        | 491/3515 [02:09<15:14,  3.31it/s]
overall:  12%|█▏        | 492/4000 [02:12<17:00,  3.44it/s]

train:  14%|█▍        | 492/3515 [02:10<14:39,  3.44it/s]
overall:  12%|█▏        | 493/4000 [02:12<16:44,  3.49it/s]

train:  14%|█▍        | 493/3515 [02:10<14:25,  3.49it/s]
overall:  12%|█▏        | 494/4000 [02:13<15:50,  3.69it/s]

train:  14%|█▍        | 494/3515 [02:10<13:38,  3.69it/s]
overall:  12%|█▏        | 495/4000 [02:13<16:44,  3.49it/s]

train:  14%|█▍        | 495/3515 [02:10<14:25,  3.49it/s]
overall:  12%|█▏        | 496/4000 [02:13<16:08,  3.62it/s]

train:  14%|█▍        | 496/3515 [02:11<13:54,  3.62it/s]
overall:  12%|█▏        | 497/4000 [02:14<15:53,  3.68it/s]

train:  14%|█▍        | 497/3515 [02:11<13:41,  3.68it/s]
overall:  12%|█▏        | 499/4000 [02:14<13:16,  4.40it/s]

train:  14%|█▍        | 499/3515 [02:11<11:25,  4.40it/s]train loss: 0.6291483044624329
train per: 0.17530086382989196

overall:  12%|█▎        | 500/4000 [02:14<13:47,  4.23it/s]

train:  14%|█▍        | 500/3515 [02:12<11:53,  4.23it/s]
overall:  13%|█▎        | 501/4000 [02:14<14:15,  4.09it/s]

train:  14%|█▍        | 501/3515 [02:12<12:16,  4.09it/s]
overall:  13%|█▎        | 502/4000 [02:15<14:33,  4.01it/s]

train:  14%|█▍        | 502/3515 [02:12<12:31,  4.01it/s]
overall:  13%|█▎        | 503/4000 [02:15<14:30,  4.02it/s]

train:  14%|█▍        | 503/3515 [02:12<12:29,  4.02it/s]
overall:  13%|█▎        | 504/4000 [02:15<14:54,  3.91it/s]

train:  14%|█▍        | 504/3515 [02:13<12:50,  3.91it/s]
overall:  13%|█▎        | 505/4000 [02:15<15:09,  3.84it/s]

train:  14%|█▍        | 505/3515 [02:13<13:03,  3.84it/s]
overall:  13%|█▎        | 506/4000 [02:16<14:32,  4.01it/s]

train:  14%|█▍        | 506/3515 [02:13<12:31,  4.01it/s]
overall:  13%|█▎        | 507/4000 [02:16<14:51,  3.92it/s]

train:  14%|█▍        | 507/3515 [02:13<12:47,  3.92it/s]
overall:  13%|█▎        | 508/4000 [02:16<13:55,  4.18it/s]

train:  14%|█▍        | 508/3515 [02:14<11:59,  4.18it/s]
overall:  13%|█▎        | 509/4000 [02:16<14:29,  4.01it/s]

train:  14%|█▍        | 509/3515 [02:14<12:28,  4.01it/s]
overall:  13%|█▎        | 510/4000 [02:17<14:51,  3.92it/s]

train:  15%|█▍        | 510/3515 [02:14<12:47,  3.92it/s]
overall:  13%|█▎        | 511/4000 [02:17<14:54,  3.90it/s]

train:  15%|█▍        | 511/3515 [02:14<12:50,  3.90it/s]
overall:  13%|█▎        | 512/4000 [02:17<13:01,  4.46it/s]

train:  15%|█▍        | 512/3515 [02:14<11:12,  4.46it/s]
overall:  13%|█▎        | 513/4000 [02:17<13:16,  4.38it/s]

train:  15%|█▍        | 513/3515 [02:15<11:25,  4.38it/s]
overall:  13%|█▎        | 514/4000 [02:18<18:13,  3.19it/s]

train:  15%|█▍        | 514/3515 [02:15<15:41,  3.19it/s]
overall:  13%|█▎        | 515/4000 [02:18<17:00,  3.42it/s]

train:  15%|█▍        | 515/3515 [02:15<14:38,  3.42it/s]
overall:  13%|█▎        | 516/4000 [02:18<15:23,  3.77it/s]

train:  15%|█▍        | 516/3515 [02:16<13:15,  3.77it/s]
overall:  13%|█▎        | 517/4000 [02:19<14:59,  3.87it/s]

train:  15%|█▍        | 517/3515 [02:16<12:54,  3.87it/s]
overall:  13%|█▎        | 518/4000 [02:19<15:13,  3.81it/s]

train:  15%|█▍        | 518/3515 [02:16<13:06,  3.81it/s]
overall:  13%|█▎        | 519/4000 [02:19<13:40,  4.24it/s]

train:  15%|█▍        | 519/3515 [02:16<11:46,  4.24it/s]train loss: 0.6405361294746399
train per: 0.18317002575745506

overall:  13%|█▎        | 520/4000 [02:19<14:25,  4.02it/s]

train:  15%|█▍        | 520/3515 [02:17<12:25,  4.02it/s]
overall:  13%|█▎        | 521/4000 [02:20<14:34,  3.98it/s]

train:  15%|█▍        | 521/3515 [02:17<12:32,  3.98it/s]
overall:  13%|█▎        | 522/4000 [02:20<14:49,  3.91it/s]

train:  15%|█▍        | 522/3515 [02:17<12:45,  3.91it/s]
overall:  13%|█▎        | 523/4000 [02:20<12:10,  4.76it/s]

train:  15%|█▍        | 523/3515 [02:17<10:28,  4.76it/s]
overall:  13%|█▎        | 524/4000 [02:20<13:05,  4.43it/s]

train:  15%|█▍        | 524/3515 [02:18<11:15,  4.43it/s]
overall:  13%|█▎        | 525/4000 [02:20<13:45,  4.21it/s]

train:  15%|█▍        | 525/3515 [02:18<11:50,  4.21it/s]
overall:  13%|█▎        | 526/4000 [02:21<14:18,  4.05it/s]

train:  15%|█▍        | 526/3515 [02:18<12:18,  4.05it/s]
overall:  13%|█▎        | 527/4000 [02:21<14:36,  3.96it/s]

train:  15%|█▍        | 527/3515 [02:18<12:34,  3.96it/s]
overall:  13%|█▎        | 528/4000 [02:21<14:51,  3.89it/s]

train:  15%|█▌        | 528/3515 [02:19<12:46,  3.89it/s]
overall:  13%|█▎        | 529/4000 [02:21<12:53,  4.49it/s]

train:  15%|█▌        | 529/3515 [02:19<11:05,  4.49it/s]
overall:  13%|█▎        | 530/4000 [02:22<13:28,  4.29it/s]

train:  15%|█▌        | 530/3515 [02:19<11:35,  4.29it/s]
overall:  13%|█▎        | 531/4000 [02:22<13:59,  4.13it/s]

train:  15%|█▌        | 531/3515 [02:19<12:01,  4.13it/s]
overall:  13%|█▎        | 532/4000 [02:22<14:19,  4.04it/s]

train:  15%|█▌        | 532/3515 [02:20<12:18,  4.04it/s]
overall:  13%|█▎        | 533/4000 [02:22<14:40,  3.94it/s]

train:  15%|█▌        | 533/3515 [02:20<12:37,  3.94it/s]
overall:  13%|█▎        | 534/4000 [02:23<14:29,  3.98it/s]

train:  15%|█▌        | 534/3515 [02:20<12:28,  3.98it/s]
overall:  13%|█▎        | 535/4000 [02:23<14:22,  4.02it/s]

train:  15%|█▌        | 535/3515 [02:20<12:21,  4.02it/s]
overall:  13%|█▎        | 536/4000 [02:23<14:52,  3.88it/s]

train:  15%|█▌        | 536/3515 [02:21<12:47,  3.88it/s]
overall:  13%|█▎        | 537/4000 [02:23<14:56,  3.86it/s]

train:  15%|█▌        | 537/3515 [02:21<12:51,  3.86it/s]
overall:  13%|█▎        | 538/4000 [02:24<19:38,  2.94it/s]

train:  15%|█▌        | 538/3515 [02:21<16:53,  2.94it/s]
overall:  13%|█▎        | 539/4000 [02:24<18:22,  3.14it/s]

train:  15%|█▌        | 539/3515 [02:22<15:48,  3.14it/s]train loss: 0.6311535835266113
train per: 0.17985159684585145

overall:  14%|█▎        | 540/4000 [02:25<17:28,  3.30it/s]

train:  15%|█▌        | 540/3515 [02:22<15:02,  3.30it/s]
overall:  14%|█▎        | 541/4000 [02:25<17:07,  3.37it/s]

train:  15%|█▌        | 541/3515 [02:22<14:43,  3.37it/s]
overall:  14%|█▎        | 542/4000 [02:25<16:10,  3.56it/s]

train:  15%|█▌        | 542/3515 [02:22<13:54,  3.56it/s]
overall:  14%|█▎        | 543/4000 [02:25<17:12,  3.35it/s]

train:  15%|█▌        | 543/3515 [02:23<14:47,  3.35it/s]
overall:  14%|█▎        | 544/4000 [02:26<16:02,  3.59it/s]

train:  15%|█▌        | 544/3515 [02:23<13:46,  3.59it/s]
overall:  14%|█▎        | 545/4000 [02:26<15:53,  3.62it/s]

train:  16%|█▌        | 545/3515 [02:23<13:39,  3.62it/s]
overall:  14%|█▎        | 546/4000 [02:26<15:42,  3.66it/s]

train:  16%|█▌        | 546/3515 [02:24<13:30,  3.66it/s]
overall:  14%|█▎        | 547/4000 [02:26<15:32,  3.70it/s]

train:  16%|█▌        | 547/3515 [02:24<13:21,  3.70it/s]
overall:  14%|█▎        | 548/4000 [02:27<15:00,  3.83it/s]

train:  16%|█▌        | 548/3515 [02:24<12:54,  3.83it/s]
overall:  14%|█▎        | 549/4000 [02:27<15:04,  3.81it/s]

train:  16%|█▌        | 549/3515 [02:24<12:57,  3.81it/s]
overall:  14%|█▍        | 550/4000 [02:27<15:09,  3.79it/s]

train:  16%|█▌        | 550/3515 [02:25<13:01,  3.79it/s]
overall:  14%|█▍        | 551/4000 [02:27<15:01,  3.82it/s]

train:  16%|█▌        | 551/3515 [02:25<12:55,  3.82it/s]
overall:  14%|█▍        | 552/4000 [02:28<15:05,  3.81it/s]

train:  16%|█▌        | 552/3515 [02:25<12:58,  3.81it/s]
overall:  14%|█▍        | 553/4000 [02:28<14:58,  3.84it/s]

train:  16%|█▌        | 553/3515 [02:25<12:52,  3.84it/s]
overall:  14%|█▍        | 554/4000 [02:28<14:59,  3.83it/s]

train:  16%|█▌        | 554/3515 [02:26<12:53,  3.83it/s]
overall:  14%|█▍        | 555/4000 [02:28<15:02,  3.82it/s]

train:  16%|█▌        | 555/3515 [02:26<12:55,  3.82it/s]
overall:  14%|█▍        | 556/4000 [02:29<15:21,  3.74it/s]

train:  16%|█▌        | 556/3515 [02:26<13:11,  3.74it/s]
overall:  14%|█▍        | 557/4000 [02:29<14:31,  3.95it/s]

train:  16%|█▌        | 557/3515 [02:26<12:28,  3.95it/s]
overall:  14%|█▍        | 558/4000 [02:29<14:35,  3.93it/s]

train:  16%|█▌        | 558/3515 [02:27<12:32,  3.93it/s]
overall:  14%|█▍        | 559/4000 [02:30<15:03,  3.81it/s]

train:  16%|█▌        | 559/3515 [02:27<12:56,  3.81it/s]train loss: 0.5992609262466431
train per: 0.17182265046098164

overall:  14%|█▍        | 560/4000 [02:30<12:54,  4.44it/s]

train:  16%|█▌        | 560/3515 [02:27<11:06,  4.44it/s]
overall:  14%|█▍        | 561/4000 [02:30<13:32,  4.23it/s]

train:  16%|█▌        | 561/3515 [02:27<11:37,  4.24it/s]
overall:  14%|█▍        | 562/4000 [02:30<17:57,  3.19it/s]

train:  16%|█▌        | 562/3515 [02:28<15:25,  3.19it/s]
overall:  14%|█▍        | 563/4000 [02:31<17:08,  3.34it/s]

train:  16%|█▌        | 563/3515 [02:28<14:42,  3.34it/s]
overall:  14%|█▍        | 564/4000 [02:31<16:07,  3.55it/s]

train:  16%|█▌        | 564/3515 [02:28<13:50,  3.55it/s]
overall:  14%|█▍        | 565/4000 [02:31<16:18,  3.51it/s]

train:  16%|█▌        | 565/3515 [02:29<14:00,  3.51it/s]
overall:  14%|█▍        | 566/4000 [02:31<15:36,  3.67it/s]

train:  16%|█▌        | 566/3515 [02:29<13:23,  3.67it/s]
overall:  14%|█▍        | 567/4000 [02:32<16:27,  3.48it/s]

train:  16%|█▌        | 567/3515 [02:29<14:08,  3.48it/s]
overall:  14%|█▍        | 568/4000 [02:32<15:58,  3.58it/s]

train:  16%|█▌        | 568/3515 [02:29<13:43,  3.58it/s]
overall:  14%|█▍        | 569/4000 [02:32<13:23,  4.27it/s]

train:  16%|█▌        | 569/3515 [02:30<11:29,  4.27it/s]
overall:  14%|█▍        | 570/4000 [02:32<14:07,  4.05it/s]

train:  16%|█▌        | 570/3515 [02:30<12:07,  4.05it/s]
overall:  14%|█▍        | 571/4000 [02:33<14:44,  3.88it/s]

train:  16%|█▌        | 571/3515 [02:30<12:38,  3.88it/s]
overall:  14%|█▍        | 572/4000 [02:33<14:49,  3.85it/s]

train:  16%|█▋        | 572/3515 [02:30<12:43,  3.85it/s]
overall:  14%|█▍        | 573/4000 [02:33<14:32,  3.93it/s]

train:  16%|█▋        | 573/3515 [02:31<12:29,  3.93it/s]
overall:  14%|█▍        | 574/4000 [02:33<14:41,  3.89it/s]

train:  16%|█▋        | 574/3515 [02:31<12:36,  3.89it/s]
overall:  14%|█▍        | 575/4000 [02:34<14:29,  3.94it/s]

train:  16%|█▋        | 575/3515 [02:31<12:26,  3.94it/s]
overall:  14%|█▍        | 576/4000 [02:34<14:37,  3.90it/s]

train:  16%|█▋        | 576/3515 [02:31<12:33,  3.90it/s]
overall:  14%|█▍        | 577/4000 [02:34<14:40,  3.89it/s]

train:  16%|█▋        | 577/3515 [02:32<12:35,  3.89it/s]
overall:  14%|█▍        | 578/4000 [02:35<14:55,  3.82it/s]

train:  16%|█▋        | 578/3515 [02:32<12:48,  3.82it/s]
overall:  14%|█▍        | 579/4000 [02:35<13:06,  4.35it/s]

train:  16%|█▋        | 579/3515 [02:32<11:15,  4.35it/s]train loss: 0.6029238700866699
train per: 0.17278988260091666

overall:  14%|█▍        | 580/4000 [02:35<13:29,  4.22it/s]

train:  17%|█▋        | 580/3515 [02:32<11:35,  4.22it/s]
overall:  15%|█▍        | 581/4000 [02:35<13:59,  4.07it/s]

train:  17%|█▋        | 581/3515 [02:33<12:00,  4.07it/s]
overall:  15%|█▍        | 582/4000 [02:35<14:26,  3.94it/s]

train:  17%|█▋        | 582/3515 [02:33<12:23,  3.94it/s]
overall:  15%|█▍        | 583/4000 [02:36<14:36,  3.90it/s]

train:  17%|█▋        | 583/3515 [02:33<12:32,  3.90it/s]
overall:  15%|█▍        | 584/4000 [02:36<14:50,  3.84it/s]

train:  17%|█▋        | 584/3515 [02:33<12:44,  3.84it/s]
overall:  15%|█▍        | 585/4000 [02:36<14:58,  3.80it/s]

train:  17%|█▋        | 585/3515 [02:34<12:50,  3.80it/s]
overall:  15%|█▍        | 586/4000 [02:37<18:31,  3.07it/s]

train:  17%|█▋        | 586/3515 [02:34<15:53,  3.07it/s]
overall:  15%|█▍        | 587/4000 [02:37<17:18,  3.29it/s]

train:  17%|█▋        | 587/3515 [02:34<14:50,  3.29it/s]
overall:  15%|█▍        | 589/4000 [02:37<13:23,  4.25it/s]

train:  17%|█▋        | 589/3515 [02:35<11:29,  4.25it/s]
overall:  15%|█▍        | 590/4000 [02:38<14:09,  4.01it/s]

train:  17%|█▋        | 590/3515 [02:35<12:08,  4.01it/s]
overall:  15%|█▍        | 591/4000 [02:38<15:36,  3.64it/s]

train:  17%|█▋        | 591/3515 [02:35<13:22,  3.64it/s]
overall:  15%|█▍        | 592/4000 [02:38<15:13,  3.73it/s]

train:  17%|█▋        | 592/3515 [02:36<13:03,  3.73it/s]
overall:  15%|█▍        | 593/4000 [02:38<14:46,  3.84it/s]

train:  17%|█▋        | 593/3515 [02:36<12:40,  3.84it/s]
overall:  15%|█▍        | 594/4000 [02:39<14:51,  3.82it/s]

train:  17%|█▋        | 594/3515 [02:36<12:44,  3.82it/s]
overall:  15%|█▍        | 595/4000 [02:39<14:44,  3.85it/s]

train:  17%|█▋        | 595/3515 [02:36<12:38,  3.85it/s]
overall:  15%|█▍        | 596/4000 [02:39<14:55,  3.80it/s]

train:  17%|█▋        | 596/3515 [02:37<12:47,  3.80it/s]
overall:  15%|█▍        | 597/4000 [02:39<14:50,  3.82it/s]

train:  17%|█▋        | 597/3515 [02:37<12:43,  3.82it/s]
overall:  15%|█▍        | 598/4000 [02:40<14:56,  3.79it/s]

train:  17%|█▋        | 598/3515 [02:37<12:48,  3.79it/s]train loss: 0.6219621300697327
train per: 0.17780863418328705

overall:  15%|█▌        | 600/4000 [02:40<12:35,  4.50it/s]

train:  17%|█▋        | 600/3515 [02:37<10:48,  4.50it/s]
overall:  15%|█▌        | 601/4000 [02:40<11:17,  5.02it/s]

train:  17%|█▋        | 601/3515 [02:38<09:40,  5.02it/s]
overall:  15%|█▌        | 602/4000 [02:41<12:15,  4.62it/s]

train:  17%|█▋        | 602/3515 [02:38<10:29,  4.62it/s]
overall:  15%|█▌        | 603/4000 [02:41<11:03,  5.12it/s]

train:  17%|█▋        | 603/3515 [02:38<09:28,  5.12it/s]
overall:  15%|█▌        | 604/4000 [02:41<12:01,  4.70it/s]

train:  17%|█▋        | 604/3515 [02:38<10:18,  4.70it/s]
overall:  15%|█▌        | 605/4000 [02:41<12:45,  4.43it/s]

train:  17%|█▋        | 605/3515 [02:39<10:56,  4.43it/s]
overall:  15%|█▌        | 606/4000 [02:41<13:31,  4.18it/s]

train:  17%|█▋        | 606/3515 [02:39<11:35,  4.18it/s]
overall:  15%|█▌        | 607/4000 [02:42<14:03,  4.02it/s]

train:  17%|█▋        | 607/3515 [02:39<12:02,  4.02it/s]
overall:  15%|█▌        | 608/4000 [02:42<13:59,  4.04it/s]

train:  17%|█▋        | 608/3515 [02:39<11:59,  4.04it/s]
overall:  15%|█▌        | 609/4000 [02:42<14:20,  3.94it/s]

train:  17%|█▋        | 609/3515 [02:40<12:17,  3.94it/s]
overall:  15%|█▌        | 610/4000 [02:43<17:41,  3.19it/s]

train:  17%|█▋        | 610/3515 [02:40<15:09,  3.19it/s]
overall:  15%|█▌        | 611/4000 [02:43<16:29,  3.43it/s]

train:  17%|█▋        | 611/3515 [02:40<14:07,  3.43it/s]
overall:  15%|█▌        | 612/4000 [02:43<16:09,  3.49it/s]

train:  17%|█▋        | 612/3515 [02:41<13:50,  3.49it/s]
overall:  15%|█▌        | 613/4000 [02:43<16:03,  3.52it/s]

train:  17%|█▋        | 613/3515 [02:41<13:45,  3.52it/s]
overall:  15%|█▌        | 614/4000 [02:44<14:13,  3.97it/s]

train:  17%|█▋        | 614/3515 [02:41<12:10,  3.97it/s]
overall:  15%|█▌        | 615/4000 [02:44<15:38,  3.61it/s]

train:  17%|█▋        | 615/3515 [02:41<13:23,  3.61it/s]
overall:  15%|█▌        | 616/4000 [02:44<14:52,  3.79it/s]

train:  18%|█▊        | 616/3515 [02:42<12:44,  3.79it/s]
overall:  15%|█▌        | 617/4000 [02:44<14:59,  3.76it/s]

train:  18%|█▊        | 617/3515 [02:42<12:50,  3.76it/s]
overall:  15%|█▌        | 618/4000 [02:45<14:59,  3.76it/s]

train:  18%|█▊        | 618/3515 [02:42<12:50,  3.76it/s]
overall:  15%|█▌        | 619/4000 [02:45<15:01,  3.75it/s]

train:  18%|█▊        | 619/3515 [02:42<12:52,  3.75it/s]train loss: 0.6125777363777161
train per: 0.17799109915131442

overall:  16%|█▌        | 620/4000 [02:45<14:42,  3.83it/s]

train:  18%|█▊        | 620/3515 [02:43<12:36,  3.83it/s]
overall:  16%|█▌        | 621/4000 [02:45<14:04,  4.00it/s]

train:  18%|█▊        | 621/3515 [02:43<12:03,  4.00it/s]
overall:  16%|█▌        | 622/4000 [02:46<14:05,  4.00it/s]

train:  18%|█▊        | 622/3515 [02:43<12:03,  4.00it/s]
overall:  16%|█▌        | 623/4000 [02:46<14:25,  3.90it/s]

train:  18%|█▊        | 623/3515 [02:43<12:21,  3.90it/s]
overall:  16%|█▌        | 624/4000 [02:46<14:15,  3.95it/s]

train:  18%|█▊        | 624/3515 [02:44<12:12,  3.95it/s]
overall:  16%|█▌        | 625/4000 [02:47<14:36,  3.85it/s]

train:  18%|█▊        | 625/3515 [02:44<12:30,  3.85it/s]
overall:  16%|█▌        | 626/4000 [02:47<14:20,  3.92it/s]

train:  18%|█▊        | 626/3515 [02:44<12:17,  3.92it/s]
overall:  16%|█▌        | 627/4000 [02:47<14:38,  3.84it/s]

train:  18%|█▊        | 627/3515 [02:44<12:32,  3.84it/s]
overall:  16%|█▌        | 628/4000 [02:47<14:35,  3.85it/s]

train:  18%|█▊        | 628/3515 [02:45<12:29,  3.85it/s]
overall:  16%|█▌        | 629/4000 [02:48<14:49,  3.79it/s]

train:  18%|█▊        | 629/3515 [02:45<12:41,  3.79it/s]
overall:  16%|█▌        | 630/4000 [02:48<14:52,  3.78it/s]

train:  18%|█▊        | 630/3515 [02:45<12:44,  3.78it/s]
overall:  16%|█▌        | 631/4000 [02:48<14:49,  3.79it/s]

train:  18%|█▊        | 631/3515 [02:45<12:41,  3.79it/s]
overall:  16%|█▌        | 632/4000 [02:48<14:18,  3.92it/s]

train:  18%|█▊        | 632/3515 [02:46<12:14,  3.92it/s]
overall:  16%|█▌        | 633/4000 [02:49<14:35,  3.84it/s]

train:  18%|█▊        | 633/3515 [02:46<12:29,  3.84it/s]
overall:  16%|█▌        | 634/4000 [02:49<18:26,  3.04it/s]

train:  18%|█▊        | 634/3515 [02:46<15:46,  3.04it/s]
overall:  16%|█▌        | 635/4000 [02:49<17:29,  3.21it/s]

train:  18%|█▊        | 635/3515 [02:47<14:58,  3.21it/s]
overall:  16%|█▌        | 636/4000 [02:50<16:36,  3.38it/s]

train:  18%|█▊        | 636/3515 [02:47<14:13,  3.38it/s]
overall:  16%|█▌        | 637/4000 [02:50<16:27,  3.40it/s]

train:  18%|█▊        | 637/3515 [02:47<14:05,  3.40it/s]
overall:  16%|█▌        | 638/4000 [02:50<14:51,  3.77it/s]

train:  18%|█▊        | 638/3515 [02:47<12:42,  3.77it/s]
overall:  16%|█▌        | 639/4000 [02:50<16:03,  3.49it/s]

train:  18%|█▊        | 639/3515 [02:48<13:44,  3.49it/s]train loss: 0.6016343832015991
train per: 0.17305413780532264

overall:  16%|█▌        | 640/4000 [02:51<15:59,  3.50it/s]

train:  18%|█▊        | 640/3515 [02:48<13:41,  3.50it/s]
overall:  16%|█▌        | 641/4000 [02:51<13:53,  4.03it/s]

train:  18%|█▊        | 641/3515 [02:48<11:52,  4.03it/s]
overall:  16%|█▌        | 642/4000 [02:51<14:27,  3.87it/s]

train:  18%|█▊        | 642/3515 [02:49<12:22,  3.87it/s]
overall:  16%|█▌        | 643/4000 [02:51<14:33,  3.84it/s]

train:  18%|█▊        | 643/3515 [02:49<12:27,  3.84it/s]
overall:  16%|█▌        | 644/4000 [02:52<14:28,  3.87it/s]

train:  18%|█▊        | 644/3515 [02:49<12:22,  3.87it/s]
overall:  16%|█▌        | 645/4000 [02:52<14:11,  3.94it/s]

train:  18%|█▊        | 645/3515 [02:49<12:08,  3.94it/s]
overall:  16%|█▌        | 646/4000 [02:52<12:35,  4.44it/s]

train:  18%|█▊        | 646/3515 [02:49<10:45,  4.44it/s]
overall:  16%|█▌        | 647/4000 [02:52<13:21,  4.18it/s]

train:  18%|█▊        | 647/3515 [02:50<11:25,  4.18it/s]
overall:  16%|█▌        | 649/4000 [02:53<11:57,  4.67it/s]

train:  18%|█▊        | 649/3515 [02:50<10:13,  4.67it/s]
overall:  16%|█▋        | 650/4000 [02:53<12:17,  4.55it/s]

train:  18%|█▊        | 650/3515 [02:50<10:30,  4.55it/s]
overall:  16%|█▋        | 651/4000 [02:53<13:08,  4.25it/s]

train:  19%|█▊        | 651/3515 [02:51<11:14,  4.25it/s]
overall:  16%|█▋        | 652/4000 [02:54<13:30,  4.13it/s]

train:  19%|█▊        | 652/3515 [02:51<11:32,  4.13it/s]
overall:  16%|█▋        | 654/4000 [02:54<11:34,  4.82it/s]

train:  19%|█▊        | 654/3515 [02:51<09:54,  4.82it/s]
overall:  16%|█▋        | 655/4000 [02:54<12:23,  4.50it/s]

train:  19%|█▊        | 655/3515 [02:51<10:35,  4.50it/s]
overall:  16%|█▋        | 656/4000 [02:54<12:33,  4.44it/s]

train:  19%|█▊        | 656/3515 [02:52<10:44,  4.44it/s]
overall:  16%|█▋        | 657/4000 [02:55<13:13,  4.21it/s]

train:  19%|█▊        | 657/3515 [02:52<11:18,  4.21it/s]
overall:  16%|█▋        | 658/4000 [02:55<17:27,  3.19it/s]

train:  19%|█▊        | 658/3515 [02:53<14:55,  3.19it/s]
overall:  16%|█▋        | 659/4000 [02:55<16:32,  3.37it/s]

train:  19%|█▊        | 659/3515 [02:53<14:08,  3.37it/s]train loss: 0.6424077153205872
train per: 0.1819485381980403

overall:  16%|█▋        | 660/4000 [02:56<16:07,  3.45it/s]

train:  19%|█▉        | 660/3515 [02:53<13:48,  3.45it/s]
overall:  17%|█▋        | 661/4000 [02:56<15:48,  3.52it/s]

train:  19%|█▉        | 661/3515 [02:53<13:30,  3.52it/s]
overall:  17%|█▋        | 662/4000 [02:56<15:30,  3.59it/s]

train:  19%|█▉        | 662/3515 [02:54<13:15,  3.59it/s]
overall:  17%|█▋        | 663/4000 [02:56<15:30,  3.59it/s]

train:  19%|█▉        | 663/3515 [02:54<13:14,  3.59it/s]
overall:  17%|█▋        | 664/4000 [02:57<14:57,  3.72it/s]

train:  19%|█▉        | 664/3515 [02:54<12:47,  3.72it/s]
overall:  17%|█▋        | 665/4000 [02:57<13:47,  4.03it/s]

train:  19%|█▉        | 665/3515 [02:54<11:47,  4.03it/s]
overall:  17%|█▋        | 666/4000 [02:57<14:10,  3.92it/s]

train:  19%|█▉        | 666/3515 [02:55<12:06,  3.92it/s]
overall:  17%|█▋        | 667/4000 [02:57<12:09,  4.57it/s]

train:  19%|█▉        | 667/3515 [02:55<10:23,  4.57it/s]
overall:  17%|█▋        | 668/4000 [02:58<12:58,  4.28it/s]

train:  19%|█▉        | 668/3515 [02:55<11:05,  4.28it/s]
overall:  17%|█▋        | 669/4000 [02:58<13:23,  4.15it/s]

train:  19%|█▉        | 669/3515 [02:55<11:26,  4.15it/s]
overall:  17%|█▋        | 671/4000 [02:58<11:34,  4.79it/s]

train:  19%|█▉        | 671/3515 [02:56<09:53,  4.79it/s]
overall:  17%|█▋        | 672/4000 [02:58<12:12,  4.54it/s]

train:  19%|█▉        | 672/3515 [02:56<10:26,  4.54it/s]
overall:  17%|█▋        | 673/4000 [02:59<12:50,  4.32it/s]

train:  19%|█▉        | 673/3515 [02:56<10:57,  4.32it/s]
overall:  17%|█▋        | 674/4000 [02:59<13:16,  4.18it/s]

train:  19%|█▉        | 674/3515 [02:56<11:20,  4.18it/s]
overall:  17%|█▋        | 675/4000 [02:59<13:35,  4.08it/s]

train:  19%|█▉        | 675/3515 [02:57<11:36,  4.08it/s]
overall:  17%|█▋        | 676/4000 [02:59<13:42,  4.04it/s]

train:  19%|█▉        | 676/3515 [02:57<11:42,  4.04it/s]
overall:  17%|█▋        | 677/4000 [03:00<13:37,  4.06it/s]

train:  19%|█▉        | 677/3515 [02:57<11:38,  4.06it/s]
overall:  17%|█▋        | 678/4000 [03:00<12:56,  4.28it/s]

train:  19%|█▉        | 678/3515 [02:57<11:03,  4.28it/s]
overall:  17%|█▋        | 679/4000 [03:00<13:21,  4.14it/s]

train:  19%|█▉        | 679/3515 [02:58<11:24,  4.14it/s]train loss: 0.6030438542366028
train per: 0.17165606887625906

overall:  17%|█▋        | 681/4000 [03:00<10:54,  5.07it/s]

train:  19%|█▉        | 681/3515 [02:58<09:19,  5.07it/s]
overall:  17%|█▋        | 682/4000 [03:01<14:27,  3.82it/s]

train:  19%|█▉        | 682/3515 [02:58<12:20,  3.82it/s]
overall:  17%|█▋        | 683/4000 [03:01<12:08,  4.55it/s]

train:  19%|█▉        | 683/3515 [02:58<10:21,  4.55it/s]
overall:  17%|█▋        | 684/4000 [03:01<12:51,  4.30it/s]

train:  19%|█▉        | 684/3515 [02:59<10:58,  4.30it/s]
overall:  17%|█▋        | 685/4000 [03:02<13:20,  4.14it/s]

train:  19%|█▉        | 685/3515 [02:59<11:23,  4.14it/s]
overall:  17%|█▋        | 686/4000 [03:02<13:49,  4.00it/s]

train:  20%|█▉        | 686/3515 [02:59<11:47,  4.00it/s]
overall:  17%|█▋        | 687/4000 [03:02<15:25,  3.58it/s]

train:  20%|█▉        | 687/3515 [03:00<13:09,  3.58it/s]
overall:  17%|█▋        | 688/4000 [03:02<15:15,  3.62it/s]

train:  20%|█▉        | 688/3515 [03:00<13:01,  3.62it/s]
overall:  17%|█▋        | 689/4000 [03:03<14:52,  3.71it/s]

train:  20%|█▉        | 689/3515 [03:00<12:41,  3.71it/s]
overall:  17%|█▋        | 690/4000 [03:03<13:27,  4.10it/s]

train:  20%|█▉        | 690/3515 [03:00<11:29,  4.10it/s]
overall:  17%|█▋        | 691/4000 [03:03<13:38,  4.04it/s]

train:  20%|█▉        | 691/3515 [03:01<11:38,  4.04it/s]
overall:  17%|█▋        | 692/4000 [03:03<14:07,  3.91it/s]

train:  20%|█▉        | 692/3515 [03:01<12:02,  3.91it/s]
overall:  17%|█▋        | 693/4000 [03:04<13:47,  3.99it/s]

train:  20%|█▉        | 693/3515 [03:01<11:46,  4.00it/s]
overall:  17%|█▋        | 694/4000 [03:04<14:11,  3.88it/s]

train:  20%|█▉        | 694/3515 [03:01<12:06,  3.88it/s]
overall:  17%|█▋        | 695/4000 [03:04<11:36,  4.75it/s]

train:  20%|█▉        | 695/3515 [03:01<09:54,  4.75it/s]
overall:  17%|█▋        | 696/4000 [03:04<12:25,  4.43it/s]

train:  20%|█▉        | 696/3515 [03:02<10:35,  4.43it/s]
overall:  17%|█▋        | 697/4000 [03:05<13:14,  4.16it/s]

train:  20%|█▉        | 697/3515 [03:02<11:17,  4.16it/s]
overall:  17%|█▋        | 698/4000 [03:05<12:53,  4.27it/s]

train:  20%|█▉        | 698/3515 [03:02<10:59,  4.27it/s]
overall:  17%|█▋        | 699/4000 [03:05<13:38,  4.03it/s]

train:  20%|█▉        | 699/3515 [03:02<11:38,  4.03it/s]train loss: 0.6511558294296265
train per: 0.18167852936658302

overall:  18%|█▊        | 700/4000 [03:05<13:44,  4.00it/s]

train:  20%|█▉        | 700/3515 [03:03<11:43,  4.00it/s]
overall:  18%|█▊        | 701/4000 [03:06<13:56,  3.94it/s]

train:  20%|█▉        | 701/3515 [03:03<11:53,  3.95it/s]
overall:  18%|█▊        | 702/4000 [03:06<13:57,  3.94it/s]

train:  20%|█▉        | 702/3515 [03:03<11:54,  3.94it/s]
overall:  18%|█▊        | 703/4000 [03:06<14:05,  3.90it/s]

train:  20%|██        | 703/3515 [03:03<12:01,  3.90it/s]
overall:  18%|█▊        | 704/4000 [03:06<12:01,  4.57it/s]

train:  20%|██        | 704/3515 [03:04<10:15,  4.57it/s]
overall:  18%|█▊        | 705/4000 [03:07<12:44,  4.31it/s]

train:  20%|██        | 705/3515 [03:04<10:52,  4.31it/s]
overall:  18%|█▊        | 706/4000 [03:07<17:22,  3.16it/s]

train:  20%|██        | 706/3515 [03:04<14:48,  3.16it/s]
overall:  18%|█▊        | 707/4000 [03:07<16:19,  3.36it/s]

train:  20%|██        | 707/3515 [03:05<13:55,  3.36it/s]
overall:  18%|█▊        | 708/4000 [03:08<15:25,  3.56it/s]

train:  20%|██        | 708/3515 [03:05<13:09,  3.56it/s]
overall:  18%|█▊        | 709/4000 [03:08<15:28,  3.54it/s]

train:  20%|██        | 709/3515 [03:05<13:11,  3.54it/s]
overall:  18%|█▊        | 710/4000 [03:08<15:19,  3.58it/s]

train:  20%|██        | 710/3515 [03:05<13:04,  3.58it/s]
overall:  18%|█▊        | 711/4000 [03:08<15:53,  3.45it/s]

train:  20%|██        | 711/3515 [03:06<13:32,  3.45it/s]
overall:  18%|█▊        | 712/4000 [03:09<15:30,  3.53it/s]

train:  20%|██        | 712/3515 [03:06<13:13,  3.53it/s]
overall:  18%|█▊        | 714/4000 [03:09<12:22,  4.43it/s]

train:  20%|██        | 714/3515 [03:06<10:32,  4.43it/s]
overall:  18%|█▊        | 715/4000 [03:09<12:59,  4.21it/s]

train:  20%|██        | 715/3515 [03:07<11:04,  4.21it/s]
overall:  18%|█▊        | 716/4000 [03:09<11:49,  4.63it/s]

train:  20%|██        | 716/3515 [03:07<10:04,  4.63it/s]
overall:  18%|█▊        | 717/4000 [03:10<12:14,  4.47it/s]

train:  20%|██        | 717/3515 [03:07<10:25,  4.47it/s]
overall:  18%|█▊        | 718/4000 [03:10<12:42,  4.30it/s]

train:  20%|██        | 718/3515 [03:07<10:50,  4.30it/s]
overall:  18%|█▊        | 719/4000 [03:10<13:16,  4.12it/s]

train:  20%|██        | 719/3515 [03:08<11:18,  4.12it/s]train loss: 0.637615442276001
train per: 0.17810551684392517

overall:  18%|█▊        | 720/4000 [03:10<13:53,  3.94it/s]

train:  20%|██        | 720/3515 [03:08<11:50,  3.93it/s]
overall:  18%|█▊        | 722/4000 [03:11<12:08,  4.50it/s]

train:  21%|██        | 722/3515 [03:08<10:20,  4.50it/s]
overall:  18%|█▊        | 723/4000 [03:11<12:36,  4.33it/s]

train:  21%|██        | 723/3515 [03:08<10:44,  4.34it/s]
overall:  18%|█▊        | 724/4000 [03:11<13:02,  4.19it/s]

train:  21%|██        | 724/3515 [03:09<11:06,  4.19it/s]
overall:  18%|█▊        | 725/4000 [03:12<13:07,  4.16it/s]

train:  21%|██        | 725/3515 [03:09<11:11,  4.16it/s]
overall:  18%|█▊        | 726/4000 [03:12<13:32,  4.03it/s]

train:  21%|██        | 726/3515 [03:09<11:32,  4.03it/s]
overall:  18%|█▊        | 727/4000 [03:12<13:39,  4.00it/s]

train:  21%|██        | 727/3515 [03:09<11:37,  4.00it/s]
overall:  18%|█▊        | 728/4000 [03:12<13:52,  3.93it/s]

train:  21%|██        | 728/3515 [03:10<11:49,  3.93it/s]
overall:  18%|█▊        | 729/4000 [03:13<13:45,  3.96it/s]

train:  21%|██        | 729/3515 [03:10<11:42,  3.96it/s]
overall:  18%|█▊        | 730/4000 [03:13<17:48,  3.06it/s]

train:  21%|██        | 730/3515 [03:11<15:10,  3.06it/s]
overall:  18%|█▊        | 731/4000 [03:13<14:11,  3.84it/s]

train:  21%|██        | 731/3515 [03:11<12:05,  3.84it/s]
overall:  18%|█▊        | 732/4000 [03:14<14:19,  3.80it/s]

train:  21%|██        | 732/3515 [03:11<12:11,  3.80it/s]
overall:  18%|█▊        | 733/4000 [03:14<12:00,  4.54it/s]

train:  21%|██        | 733/3515 [03:11<10:13,  4.54it/s]
overall:  18%|█▊        | 734/4000 [03:14<12:45,  4.26it/s]

train:  21%|██        | 734/3515 [03:11<10:52,  4.26it/s]
overall:  18%|█▊        | 735/4000 [03:14<14:27,  3.76it/s]

train:  21%|██        | 735/3515 [03:12<12:18,  3.76it/s]
overall:  18%|█▊        | 736/4000 [03:14<14:27,  3.76it/s]

train:  21%|██        | 736/3515 [03:12<12:18,  3.76it/s]
overall:  18%|█▊        | 738/4000 [03:15<10:40,  5.09it/s]

train:  21%|██        | 738/3515 [03:12<09:05,  5.09it/s]
overall:  18%|█▊        | 739/4000 [03:15<11:41,  4.65it/s]

train:  21%|██        | 739/3515 [03:12<09:57,  4.65it/s]train loss: 0.6010352373123169
train per: 0.17298719618055555

overall:  18%|█▊        | 740/4000 [03:15<12:28,  4.36it/s]

train:  21%|██        | 740/3515 [03:13<10:37,  4.35it/s]
overall:  19%|█▊        | 741/4000 [03:16<12:54,  4.21it/s]

train:  21%|██        | 741/3515 [03:13<10:59,  4.21it/s]
overall:  19%|█▊        | 742/4000 [03:16<13:12,  4.11it/s]

train:  21%|██        | 742/3515 [03:13<11:14,  4.11it/s]
overall:  19%|█▊        | 743/4000 [03:16<11:50,  4.58it/s]

train:  21%|██        | 743/3515 [03:13<10:04,  4.58it/s]
overall:  19%|█▊        | 744/4000 [03:16<12:48,  4.23it/s]

train:  21%|██        | 744/3515 [03:14<10:54,  4.23it/s]
overall:  19%|█▊        | 745/4000 [03:16<12:55,  4.20it/s]

train:  21%|██        | 745/3515 [03:14<10:59,  4.20it/s]
overall:  19%|█▊        | 746/4000 [03:17<13:15,  4.09it/s]

train:  21%|██        | 746/3515 [03:14<11:17,  4.09it/s]
overall:  19%|█▊        | 747/4000 [03:17<11:46,  4.61it/s]

train:  21%|██▏       | 747/3515 [03:14<10:00,  4.61it/s]
overall:  19%|█▊        | 748/4000 [03:17<11:56,  4.54it/s]

train:  21%|██▏       | 748/3515 [03:14<10:09,  4.54it/s]
overall:  19%|█▊        | 749/4000 [03:17<12:14,  4.43it/s]

train:  21%|██▏       | 749/3515 [03:15<10:24,  4.43it/s]
overall:  19%|█▉        | 750/4000 [03:18<12:43,  4.26it/s]

train:  21%|██▏       | 750/3515 [03:15<10:49,  4.26it/s]
overall:  19%|█▉        | 751/4000 [03:18<11:08,  4.86it/s]

train:  21%|██▏       | 751/3515 [03:15<09:28,  4.86it/s]
overall:  19%|█▉        | 752/4000 [03:18<10:50,  4.99it/s]

train:  21%|██▏       | 752/3515 [03:15<09:13,  4.99it/s]
overall:  19%|█▉        | 753/4000 [03:18<12:05,  4.47it/s]

train:  21%|██▏       | 753/3515 [03:16<10:17,  4.47it/s]
overall:  19%|█▉        | 754/4000 [03:19<16:07,  3.35it/s]

train:  21%|██▏       | 754/3515 [03:16<13:43,  3.35it/s]
overall:  19%|█▉        | 755/4000 [03:19<15:26,  3.50it/s]

train:  21%|██▏       | 755/3515 [03:16<13:07,  3.50it/s]
overall:  19%|█▉        | 756/4000 [03:19<14:58,  3.61it/s]

train:  22%|██▏       | 756/3515 [03:17<12:44,  3.61it/s]
overall:  19%|█▉        | 757/4000 [03:19<14:55,  3.62it/s]

train:  22%|██▏       | 757/3515 [03:17<12:41,  3.62it/s]
overall:  19%|█▉        | 758/4000 [03:20<12:39,  4.27it/s]

train:  22%|██▏       | 758/3515 [03:17<10:45,  4.27it/s]
overall:  19%|█▉        | 759/4000 [03:20<13:46,  3.92it/s]

train:  22%|██▏       | 759/3515 [03:17<11:42,  3.92it/s]train loss: 0.6738370060920715
train per: 0.1878603459320948

overall:  19%|█▉        | 760/4000 [03:20<14:01,  3.85it/s]

train:  22%|██▏       | 760/3515 [03:18<11:55,  3.85it/s]
overall:  19%|█▉        | 761/4000 [03:20<14:08,  3.82it/s]

train:  22%|██▏       | 761/3515 [03:18<12:01,  3.82it/s]
overall:  19%|█▉        | 762/4000 [03:21<13:25,  4.02it/s]

train:  22%|██▏       | 762/3515 [03:18<11:24,  4.02it/s]
overall:  19%|█▉        | 763/4000 [03:21<13:49,  3.90it/s]

train:  22%|██▏       | 763/3515 [03:18<11:45,  3.90it/s]
overall:  19%|█▉        | 764/4000 [03:21<13:36,  3.96it/s]

train:  22%|██▏       | 764/3515 [03:19<11:34,  3.96it/s]
overall:  19%|█▉        | 765/4000 [03:21<13:41,  3.94it/s]

train:  22%|██▏       | 765/3515 [03:19<11:38,  3.94it/s]
overall:  19%|█▉        | 766/4000 [03:22<14:06,  3.82it/s]

train:  22%|██▏       | 766/3515 [03:19<11:59,  3.82it/s]
overall:  19%|█▉        | 767/4000 [03:22<13:19,  4.05it/s]

train:  22%|██▏       | 767/3515 [03:19<11:19,  4.05it/s]
overall:  19%|█▉        | 769/4000 [03:22<11:22,  4.73it/s]

train:  22%|██▏       | 769/3515 [03:20<09:40,  4.73it/s]
overall:  19%|█▉        | 770/4000 [03:22<10:20,  5.20it/s]

train:  22%|██▏       | 770/3515 [03:20<08:47,  5.20it/s]
overall:  19%|█▉        | 771/4000 [03:23<11:13,  4.79it/s]

train:  22%|██▏       | 771/3515 [03:20<09:32,  4.79it/s]
overall:  19%|█▉        | 772/4000 [03:23<12:11,  4.41it/s]

train:  22%|██▏       | 772/3515 [03:20<10:21,  4.41it/s]
overall:  19%|█▉        | 773/4000 [03:23<12:27,  4.31it/s]

train:  22%|██▏       | 773/3515 [03:21<10:35,  4.31it/s]
overall:  19%|█▉        | 774/4000 [03:23<13:12,  4.07it/s]

train:  22%|██▏       | 774/3515 [03:21<11:13,  4.07it/s]
overall:  19%|█▉        | 775/4000 [03:24<12:43,  4.23it/s]

train:  22%|██▏       | 775/3515 [03:21<10:48,  4.23it/s]
overall:  19%|█▉        | 776/4000 [03:24<12:50,  4.19it/s]

train:  22%|██▏       | 776/3515 [03:21<10:54,  4.19it/s]
overall:  19%|█▉        | 777/4000 [03:24<13:01,  4.13it/s]

train:  22%|██▏       | 777/3515 [03:22<11:03,  4.13it/s]
overall:  19%|█▉        | 778/4000 [03:25<16:33,  3.24it/s]

train:  22%|██▏       | 778/3515 [03:22<14:04,  3.24it/s]
overall:  19%|█▉        | 779/4000 [03:25<16:00,  3.35it/s]

train:  22%|██▏       | 779/3515 [03:22<13:35,  3.35it/s]train loss: 0.6115749478340149
train per: 0.17555652107240524

overall:  20%|█▉        | 780/4000 [03:25<15:47,  3.40it/s]

train:  22%|██▏       | 780/3515 [03:23<13:25,  3.40it/s]
overall:  20%|█▉        | 781/4000 [03:25<15:33,  3.45it/s]

train:  22%|██▏       | 781/3515 [03:23<13:12,  3.45it/s]
overall:  20%|█▉        | 782/4000 [03:26<15:07,  3.54it/s]

train:  22%|██▏       | 782/3515 [03:23<12:50,  3.54it/s]
overall:  20%|█▉        | 783/4000 [03:26<15:34,  3.44it/s]

train:  22%|██▏       | 783/3515 [03:23<13:13,  3.44it/s]
overall:  20%|█▉        | 784/4000 [03:26<15:10,  3.53it/s]

train:  22%|██▏       | 784/3515 [03:24<12:52,  3.53it/s]
overall:  20%|█▉        | 785/4000 [03:27<14:53,  3.60it/s]

train:  22%|██▏       | 785/3515 [03:24<12:38,  3.60it/s]
overall:  20%|█▉        | 786/4000 [03:27<15:00,  3.57it/s]

train:  22%|██▏       | 786/3515 [03:24<12:44,  3.57it/s]
overall:  20%|█▉        | 787/4000 [03:27<13:43,  3.90it/s]

train:  22%|██▏       | 787/3515 [03:24<11:38,  3.90it/s]
overall:  20%|█▉        | 788/4000 [03:27<11:45,  4.55it/s]

train:  22%|██▏       | 788/3515 [03:25<09:59,  4.55it/s]
overall:  20%|█▉        | 789/4000 [03:27<12:33,  4.26it/s]

train:  22%|██▏       | 789/3515 [03:25<10:39,  4.26it/s]
overall:  20%|█▉        | 790/4000 [03:28<13:14,  4.04it/s]

train:  22%|██▏       | 790/3515 [03:25<11:14,  4.04it/s]
overall:  20%|█▉        | 791/4000 [03:28<13:13,  4.04it/s]

train:  23%|██▎       | 791/3515 [03:25<11:13,  4.04it/s]
overall:  20%|█▉        | 792/4000 [03:28<11:48,  4.53it/s]

train:  23%|██▎       | 792/3515 [03:26<10:01,  4.53it/s]
overall:  20%|█▉        | 793/4000 [03:28<12:32,  4.26it/s]

train:  23%|██▎       | 793/3515 [03:26<10:38,  4.26it/s]
overall:  20%|█▉        | 794/4000 [03:29<13:09,  4.06it/s]

train:  23%|██▎       | 794/3515 [03:26<11:10,  4.06it/s]
overall:  20%|█▉        | 795/4000 [03:29<13:22,  3.99it/s]

train:  23%|██▎       | 795/3515 [03:26<11:21,  3.99it/s]
overall:  20%|█▉        | 796/4000 [03:29<13:37,  3.92it/s]

train:  23%|██▎       | 796/3515 [03:27<11:33,  3.92it/s]
overall:  20%|█▉        | 797/4000 [03:29<13:39,  3.91it/s]

train:  23%|██▎       | 797/3515 [03:27<11:35,  3.91it/s]
overall:  20%|█▉        | 798/4000 [03:30<13:45,  3.88it/s]

train:  23%|██▎       | 798/3515 [03:27<11:40,  3.88it/s]
overall:  20%|█▉        | 799/4000 [03:30<13:55,  3.83it/s]

train:  23%|██▎       | 799/3515 [03:27<11:49,  3.83it/s]train loss: 0.5813694000244141
train per: 0.17087238644556596

overall:  20%|██        | 800/4000 [03:30<13:50,  3.85it/s]

train:  23%|██▎       | 800/3515 [03:28<11:44,  3.85it/s]
overall:  20%|██        | 801/4000 [03:31<14:14,  3.75it/s]

train:  23%|██▎       | 801/3515 [03:28<12:04,  3.75it/s]
overall:  20%|██        | 802/4000 [03:31<16:47,  3.17it/s]

train:  23%|██▎       | 802/3515 [03:28<14:14,  3.17it/s]
overall:  20%|██        | 803/4000 [03:31<15:04,  3.54it/s]

train:  23%|██▎       | 803/3515 [03:29<12:46,  3.54it/s]
overall:  20%|██        | 805/4000 [03:31<11:52,  4.49it/s]

train:  23%|██▎       | 805/3515 [03:29<10:04,  4.49it/s]
overall:  20%|██        | 806/4000 [03:32<10:19,  5.15it/s]

train:  23%|██▎       | 806/3515 [03:29<08:45,  5.16it/s]
overall:  20%|██        | 807/4000 [03:32<12:25,  4.28it/s]

train:  23%|██▎       | 807/3515 [03:29<10:32,  4.28it/s]
overall:  20%|██        | 808/4000 [03:32<12:41,  4.19it/s]

train:  23%|██▎       | 808/3515 [03:30<10:45,  4.19it/s]
overall:  20%|██        | 809/4000 [03:32<13:11,  4.03it/s]

train:  23%|██▎       | 809/3515 [03:30<11:11,  4.03it/s]
overall:  20%|██        | 810/4000 [03:33<13:25,  3.96it/s]

train:  23%|██▎       | 810/3515 [03:30<11:23,  3.96it/s]
overall:  20%|██        | 811/4000 [03:33<13:37,  3.90it/s]

train:  23%|██▎       | 811/3515 [03:30<11:33,  3.90it/s]
overall:  20%|██        | 812/4000 [03:33<13:38,  3.90it/s]

train:  23%|██▎       | 812/3515 [03:31<11:33,  3.90it/s]
overall:  20%|██        | 813/4000 [03:33<13:26,  3.95it/s]

train:  23%|██▎       | 813/3515 [03:31<11:23,  3.95it/s]
overall:  20%|██        | 814/4000 [03:34<13:47,  3.85it/s]

train:  23%|██▎       | 814/3515 [03:31<11:41,  3.85it/s]
overall:  20%|██        | 816/4000 [03:34<11:35,  4.58it/s]

train:  23%|██▎       | 816/3515 [03:31<09:49,  4.58it/s]
overall:  20%|██        | 817/4000 [03:34<11:50,  4.48it/s]

train:  23%|██▎       | 817/3515 [03:32<10:02,  4.48it/s]
overall:  20%|██        | 818/4000 [03:35<12:35,  4.21it/s]

train:  23%|██▎       | 818/3515 [03:32<10:40,  4.21it/s]
overall:  20%|██        | 819/4000 [03:35<13:03,  4.06it/s]

train:  23%|██▎       | 819/3515 [03:32<11:03,  4.06it/s]train loss: 0.6349223852157593
train per: 0.18175717530456328

overall:  20%|██        | 820/4000 [03:35<13:26,  3.94it/s]

train:  23%|██▎       | 820/3515 [03:33<11:23,  3.94it/s]
overall:  21%|██        | 822/4000 [03:36<11:49,  4.48it/s]

train:  23%|██▎       | 822/3515 [03:33<10:00,  4.48it/s]
overall:  21%|██        | 823/4000 [03:36<12:28,  4.24it/s]

train:  23%|██▎       | 823/3515 [03:33<10:34,  4.24it/s]
overall:  21%|██        | 824/4000 [03:36<12:34,  4.21it/s]

train:  23%|██▎       | 824/3515 [03:33<10:39,  4.21it/s]
overall:  21%|██        | 825/4000 [03:36<12:58,  4.08it/s]

train:  23%|██▎       | 825/3515 [03:34<10:59,  4.08it/s]
overall:  21%|██        | 826/4000 [03:37<16:54,  3.13it/s]

train:  23%|██▎       | 826/3515 [03:34<14:19,  3.13it/s]
overall:  21%|██        | 827/4000 [03:37<15:24,  3.43it/s]

train:  24%|██▎       | 827/3515 [03:34<13:02,  3.43it/s]
overall:  21%|██        | 828/4000 [03:37<14:55,  3.54it/s]

train:  24%|██▎       | 828/3515 [03:35<12:38,  3.54it/s]
overall:  21%|██        | 829/4000 [03:38<14:45,  3.58it/s]

train:  24%|██▎       | 829/3515 [03:35<12:29,  3.58it/s]
overall:  21%|██        | 830/4000 [03:38<14:25,  3.66it/s]

train:  24%|██▎       | 830/3515 [03:35<12:12,  3.66it/s]
overall:  21%|██        | 831/4000 [03:38<15:01,  3.51it/s]

train:  24%|██▎       | 831/3515 [03:36<12:43,  3.51it/s]
overall:  21%|██        | 832/4000 [03:38<14:40,  3.60it/s]

train:  24%|██▎       | 832/3515 [03:36<12:25,  3.60it/s]
overall:  21%|██        | 833/4000 [03:39<14:24,  3.66it/s]

train:  24%|██▎       | 833/3515 [03:36<12:11,  3.66it/s]
overall:  21%|██        | 834/4000 [03:39<14:29,  3.64it/s]

train:  24%|██▎       | 834/3515 [03:36<12:16,  3.64it/s]
overall:  21%|██        | 835/4000 [03:39<13:53,  3.80it/s]

train:  24%|██▍       | 835/3515 [03:37<11:45,  3.80it/s]
overall:  21%|██        | 836/4000 [03:39<13:35,  3.88it/s]

train:  24%|██▍       | 836/3515 [03:37<11:30,  3.88it/s]
overall:  21%|██        | 837/4000 [03:40<13:47,  3.82it/s]

train:  24%|██▍       | 837/3515 [03:37<11:40,  3.82it/s]
overall:  21%|██        | 838/4000 [03:40<13:53,  3.80it/s]

train:  24%|██▍       | 838/3515 [03:37<11:45,  3.80it/s]
overall:  21%|██        | 839/4000 [03:40<11:21,  4.63it/s]

train:  24%|██▍       | 839/3515 [03:37<09:37,  4.63it/s]train loss: 0.6289446353912354
train per: 0.17767852808388757

overall:  21%|██        | 840/4000 [03:40<12:22,  4.26it/s]

train:  24%|██▍       | 840/3515 [03:38<10:28,  4.25it/s]
overall:  21%|██        | 841/4000 [03:41<12:42,  4.15it/s]

train:  24%|██▍       | 841/3515 [03:38<10:44,  4.15it/s]
overall:  21%|██        | 842/4000 [03:41<12:57,  4.06it/s]

train:  24%|██▍       | 842/3515 [03:38<10:58,  4.06it/s]
overall:  21%|██        | 843/4000 [03:41<13:31,  3.89it/s]

train:  24%|██▍       | 843/3515 [03:39<11:26,  3.89it/s]
overall:  21%|██        | 844/4000 [03:41<13:35,  3.87it/s]

train:  24%|██▍       | 844/3515 [03:39<11:29,  3.87it/s]
overall:  21%|██        | 845/4000 [03:42<13:32,  3.88it/s]

train:  24%|██▍       | 845/3515 [03:39<11:27,  3.88it/s]
overall:  21%|██        | 846/4000 [03:42<13:29,  3.90it/s]

train:  24%|██▍       | 846/3515 [03:39<11:24,  3.90it/s]
overall:  21%|██        | 847/4000 [03:42<13:40,  3.84it/s]

train:  24%|██▍       | 847/3515 [03:40<11:34,  3.84it/s]
overall:  21%|██        | 848/4000 [03:42<13:45,  3.82it/s]

train:  24%|██▍       | 848/3515 [03:40<11:38,  3.82it/s]
overall:  21%|██        | 849/4000 [03:43<13:42,  3.83it/s]

train:  24%|██▍       | 849/3515 [03:40<11:35,  3.83it/s]
overall:  21%|██▏       | 850/4000 [03:43<16:47,  3.13it/s]

train:  24%|██▍       | 850/3515 [03:41<14:12,  3.13it/s]
overall:  21%|██▏       | 851/4000 [03:43<15:05,  3.48it/s]

train:  24%|██▍       | 851/3515 [03:41<12:45,  3.48it/s]
overall:  21%|██▏       | 852/4000 [03:44<14:38,  3.59it/s]

train:  24%|██▍       | 852/3515 [03:41<12:22,  3.59it/s]
overall:  21%|██▏       | 853/4000 [03:44<14:27,  3.63it/s]

train:  24%|██▍       | 853/3515 [03:41<12:13,  3.63it/s]
overall:  21%|██▏       | 854/4000 [03:44<14:10,  3.70it/s]

train:  24%|██▍       | 854/3515 [03:42<11:59,  3.70it/s]
overall:  21%|██▏       | 855/4000 [03:45<15:22,  3.41it/s]

train:  24%|██▍       | 855/3515 [03:42<13:00,  3.41it/s]
overall:  21%|██▏       | 856/4000 [03:45<14:47,  3.54it/s]

train:  24%|██▍       | 856/3515 [03:42<12:30,  3.54it/s]
overall:  21%|██▏       | 857/4000 [03:45<14:17,  3.66it/s]

train:  24%|██▍       | 857/3515 [03:42<12:05,  3.66it/s]
overall:  21%|██▏       | 858/4000 [03:45<14:25,  3.63it/s]

train:  24%|██▍       | 858/3515 [03:43<12:12,  3.63it/s]
overall:  21%|██▏       | 859/4000 [03:46<14:08,  3.70it/s]

train:  24%|██▍       | 859/3515 [03:43<11:57,  3.70it/s]train loss: 0.5792397856712341
train per: 0.1668216523101018

overall:  22%|██▏       | 860/4000 [03:46<13:11,  3.97it/s]

train:  24%|██▍       | 860/3515 [03:43<11:09,  3.96it/s]
overall:  22%|██▏       | 861/4000 [03:46<13:24,  3.90it/s]

train:  24%|██▍       | 861/3515 [03:43<11:19,  3.90it/s]
overall:  22%|██▏       | 862/4000 [03:46<13:18,  3.93it/s]

train:  25%|██▍       | 862/3515 [03:44<11:14,  3.93it/s]
overall:  22%|██▏       | 863/4000 [03:47<13:41,  3.82it/s]

train:  25%|██▍       | 863/3515 [03:44<11:34,  3.82it/s]
overall:  22%|██▏       | 864/4000 [03:47<13:44,  3.80it/s]

train:  25%|██▍       | 864/3515 [03:44<11:37,  3.80it/s]
overall:  22%|██▏       | 865/4000 [03:47<13:39,  3.83it/s]

train:  25%|██▍       | 865/3515 [03:44<11:32,  3.83it/s]
overall:  22%|██▏       | 866/4000 [03:47<13:56,  3.75it/s]

train:  25%|██▍       | 866/3515 [03:45<11:47,  3.75it/s]
overall:  22%|██▏       | 867/4000 [03:48<13:50,  3.77it/s]

train:  25%|██▍       | 867/3515 [03:45<11:41,  3.77it/s]
overall:  22%|██▏       | 868/4000 [03:48<13:58,  3.73it/s]

train:  25%|██▍       | 868/3515 [03:45<11:48,  3.73it/s]
overall:  22%|██▏       | 869/4000 [03:48<13:54,  3.75it/s]

train:  25%|██▍       | 869/3515 [03:46<11:44,  3.75it/s]
overall:  22%|██▏       | 870/4000 [03:48<13:46,  3.79it/s]

train:  25%|██▍       | 870/3515 [03:46<11:38,  3.79it/s]
overall:  22%|██▏       | 871/4000 [03:49<13:14,  3.94it/s]

train:  25%|██▍       | 871/3515 [03:46<11:11,  3.94it/s]
overall:  22%|██▏       | 872/4000 [03:49<13:07,  3.97it/s]

train:  25%|██▍       | 872/3515 [03:46<11:05,  3.97it/s]
overall:  22%|██▏       | 873/4000 [03:49<13:14,  3.93it/s]

train:  25%|██▍       | 873/3515 [03:47<11:11,  3.93it/s]
overall:  22%|██▏       | 874/4000 [03:50<16:52,  3.09it/s]

train:  25%|██▍       | 874/3515 [03:47<14:15,  3.09it/s]
overall:  22%|██▏       | 875/4000 [03:50<15:32,  3.35it/s]

train:  25%|██▍       | 875/3515 [03:47<13:07,  3.35it/s]
overall:  22%|██▏       | 876/4000 [03:50<14:50,  3.51it/s]

train:  25%|██▍       | 876/3515 [03:48<12:32,  3.51it/s]
overall:  22%|██▏       | 877/4000 [03:50<14:42,  3.54it/s]

train:  25%|██▍       | 877/3515 [03:48<12:25,  3.54it/s]
overall:  22%|██▏       | 878/4000 [03:51<13:58,  3.72it/s]

train:  25%|██▍       | 878/3515 [03:48<11:48,  3.72it/s]
overall:  22%|██▏       | 879/4000 [03:51<11:56,  4.36it/s]

train:  25%|██▌       | 879/3515 [03:48<10:04,  4.36it/s]train loss: 0.5775921940803528
train per: 0.16574783968819715

overall:  22%|██▏       | 880/4000 [03:51<12:21,  4.21it/s]

train:  25%|██▌       | 880/3515 [03:48<10:26,  4.21it/s]
overall:  22%|██▏       | 881/4000 [03:51<12:23,  4.20it/s]

train:  25%|██▌       | 881/3515 [03:49<10:27,  4.20it/s]
overall:  22%|██▏       | 882/4000 [03:52<12:37,  4.12it/s]

train:  25%|██▌       | 882/3515 [03:49<10:39,  4.12it/s]
overall:  22%|██▏       | 883/4000 [03:52<13:02,  3.98it/s]

train:  25%|██▌       | 883/3515 [03:49<11:00,  3.99it/s]
overall:  22%|██▏       | 884/4000 [03:52<13:07,  3.96it/s]

train:  25%|██▌       | 884/3515 [03:49<11:04,  3.96it/s]
overall:  22%|██▏       | 886/4000 [03:52<11:04,  4.69it/s]

train:  25%|██▌       | 886/3515 [03:50<09:20,  4.69it/s]
overall:  22%|██▏       | 887/4000 [03:53<11:46,  4.40it/s]

train:  25%|██▌       | 887/3515 [03:50<09:56,  4.40it/s]
overall:  22%|██▏       | 888/4000 [03:53<11:07,  4.66it/s]

train:  25%|██▌       | 888/3515 [03:50<09:23,  4.66it/s]
overall:  22%|██▏       | 889/4000 [03:53<09:36,  5.39it/s]

train:  25%|██▌       | 889/3515 [03:50<08:07,  5.39it/s]
overall:  22%|██▏       | 890/4000 [03:53<09:53,  5.24it/s]

train:  25%|██▌       | 890/3515 [03:51<08:20,  5.24it/s]
overall:  22%|██▏       | 891/4000 [03:53<10:43,  4.83it/s]

train:  25%|██▌       | 891/3515 [03:51<09:03,  4.83it/s]
overall:  22%|██▏       | 892/4000 [03:54<11:23,  4.54it/s]

train:  25%|██▌       | 892/3515 [03:51<09:37,  4.54it/s]
overall:  22%|██▏       | 893/4000 [03:54<12:05,  4.28it/s]

train:  25%|██▌       | 893/3515 [03:51<10:12,  4.28it/s]
overall:  22%|██▏       | 894/4000 [03:54<12:45,  4.06it/s]

train:  25%|██▌       | 894/3515 [03:52<10:46,  4.06it/s]
overall:  22%|██▏       | 895/4000 [03:54<13:03,  3.97it/s]

train:  25%|██▌       | 895/3515 [03:52<11:00,  3.97it/s]
overall:  22%|██▏       | 896/4000 [03:55<12:21,  4.18it/s]

train:  25%|██▌       | 896/3515 [03:52<10:26,  4.18it/s]
overall:  22%|██▏       | 897/4000 [03:55<12:44,  4.06it/s]

train:  26%|██▌       | 897/3515 [03:52<10:44,  4.06it/s]
overall:  22%|██▏       | 898/4000 [03:55<16:54,  3.06it/s]

train:  26%|██▌       | 898/3515 [03:53<14:15,  3.06it/s]
overall:  22%|██▏       | 899/4000 [03:56<15:23,  3.36it/s]

train:  26%|██▌       | 899/3515 [03:53<12:58,  3.36it/s]train loss: 0.5804818868637085
train per: 0.17087211491902943

overall:  22%|██▎       | 900/4000 [03:56<14:57,  3.45it/s]

train:  26%|██▌       | 900/3515 [03:53<12:37,  3.45it/s]
overall:  23%|██▎       | 901/4000 [03:56<14:45,  3.50it/s]

train:  26%|██▌       | 901/3515 [03:54<12:27,  3.50it/s]
overall:  23%|██▎       | 902/4000 [03:57<14:41,  3.52it/s]

train:  26%|██▌       | 902/3515 [03:54<12:23,  3.52it/s]
overall:  23%|██▎       | 903/4000 [03:57<15:03,  3.43it/s]

train:  26%|██▌       | 903/3515 [03:54<12:41,  3.43it/s]
overall:  23%|██▎       | 904/4000 [03:57<14:30,  3.56it/s]

train:  26%|██▌       | 904/3515 [03:54<12:14,  3.56it/s]
overall:  23%|██▎       | 905/4000 [03:57<14:21,  3.59it/s]

train:  26%|██▌       | 905/3515 [03:55<12:06,  3.59it/s]
overall:  23%|██▎       | 906/4000 [03:58<14:19,  3.60it/s]

train:  26%|██▌       | 906/3515 [03:55<12:04,  3.60it/s]
overall:  23%|██▎       | 907/4000 [03:58<14:15,  3.62it/s]

train:  26%|██▌       | 907/3515 [03:55<12:00,  3.62it/s]
overall:  23%|██▎       | 908/4000 [03:58<13:39,  3.77it/s]

train:  26%|██▌       | 908/3515 [03:56<11:30,  3.77it/s]
overall:  23%|██▎       | 909/4000 [03:58<13:48,  3.73it/s]

train:  26%|██▌       | 909/3515 [03:56<11:38,  3.73it/s]
overall:  23%|██▎       | 910/4000 [03:59<12:38,  4.07it/s]

train:  26%|██▌       | 910/3515 [03:56<10:39,  4.07it/s]
overall:  23%|██▎       | 911/4000 [03:59<12:56,  3.98it/s]

train:  26%|██▌       | 911/3515 [03:56<10:54,  3.98it/s]
overall:  23%|██▎       | 912/4000 [03:59<13:05,  3.93it/s]

train:  26%|██▌       | 912/3515 [03:57<11:02,  3.93it/s]
overall:  23%|██▎       | 914/4000 [03:59<11:12,  4.59it/s]

train:  26%|██▌       | 914/3515 [03:57<09:26,  4.59it/s]
overall:  23%|██▎       | 915/4000 [04:00<11:29,  4.47it/s]

train:  26%|██▌       | 915/3515 [03:57<09:41,  4.47it/s]
overall:  23%|██▎       | 916/4000 [04:00<11:44,  4.38it/s]

train:  26%|██▌       | 916/3515 [03:57<09:53,  4.38it/s]
overall:  23%|██▎       | 917/4000 [04:00<12:23,  4.14it/s]

train:  26%|██▌       | 917/3515 [03:58<10:26,  4.14it/s]
overall:  23%|██▎       | 918/4000 [04:01<12:35,  4.08it/s]

train:  26%|██▌       | 918/3515 [03:58<10:36,  4.08it/s]
overall:  23%|██▎       | 919/4000 [04:01<13:01,  3.94it/s]

train:  26%|██▌       | 919/3515 [03:58<10:58,  3.94it/s]train loss: 0.6089640259742737
train per: 0.1736020325102982

overall:  23%|██▎       | 920/4000 [04:01<13:13,  3.88it/s]

train:  26%|██▌       | 920/3515 [03:58<11:09,  3.88it/s]
overall:  23%|██▎       | 921/4000 [04:01<13:12,  3.88it/s]

train:  26%|██▌       | 921/3515 [03:59<11:07,  3.88it/s]
overall:  23%|██▎       | 922/4000 [04:02<17:45,  2.89it/s]

train:  26%|██▌       | 922/3515 [03:59<14:57,  2.89it/s]
overall:  23%|██▎       | 923/4000 [04:02<16:31,  3.10it/s]

train:  26%|██▋       | 923/3515 [04:00<13:54,  3.10it/s]
overall:  23%|██▎       | 924/4000 [04:02<15:47,  3.25it/s]

train:  26%|██▋       | 924/3515 [04:00<13:17,  3.25it/s]
overall:  23%|██▎       | 925/4000 [04:03<15:21,  3.34it/s]

train:  26%|██▋       | 925/3515 [04:00<12:56,  3.34it/s]
overall:  23%|██▎       | 926/4000 [04:03<15:00,  3.41it/s]

train:  26%|██▋       | 926/3515 [04:00<12:38,  3.41it/s]
overall:  23%|██▎       | 927/4000 [04:03<15:51,  3.23it/s]

train:  26%|██▋       | 927/3515 [04:01<13:21,  3.23it/s]
overall:  23%|██▎       | 928/4000 [04:04<15:15,  3.35it/s]

train:  26%|██▋       | 928/3515 [04:01<12:51,  3.35it/s]
overall:  23%|██▎       | 929/4000 [04:04<14:22,  3.56it/s]

train:  26%|██▋       | 929/3515 [04:01<12:06,  3.56it/s]
overall:  23%|██▎       | 930/4000 [04:04<14:15,  3.59it/s]

train:  26%|██▋       | 930/3515 [04:01<12:00,  3.59it/s]
overall:  23%|██▎       | 931/4000 [04:04<14:04,  3.63it/s]

train:  26%|██▋       | 931/3515 [04:02<11:51,  3.63it/s]
overall:  23%|██▎       | 932/4000 [04:05<13:38,  3.75it/s]

train:  27%|██▋       | 932/3515 [04:02<11:28,  3.75it/s]
overall:  23%|██▎       | 933/4000 [04:05<13:25,  3.81it/s]

train:  27%|██▋       | 933/3515 [04:02<11:17,  3.81it/s]
overall:  23%|██▎       | 934/4000 [04:05<13:38,  3.74it/s]

train:  27%|██▋       | 934/3515 [04:03<11:29,  3.74it/s]
overall:  23%|██▎       | 935/4000 [04:05<13:18,  3.84it/s]

train:  27%|██▋       | 935/3515 [04:03<11:12,  3.84it/s]
overall:  23%|██▎       | 936/4000 [04:06<12:37,  4.05it/s]

train:  27%|██▋       | 936/3515 [04:03<10:37,  4.05it/s]
overall:  23%|██▎       | 937/4000 [04:06<12:57,  3.94it/s]

train:  27%|██▋       | 937/3515 [04:03<10:54,  3.94it/s]
overall:  23%|██▎       | 938/4000 [04:06<13:11,  3.87it/s]

train:  27%|██▋       | 938/3515 [04:04<11:06,  3.87it/s]
overall:  23%|██▎       | 939/4000 [04:06<13:12,  3.86it/s]

train:  27%|██▋       | 939/3515 [04:04<11:07,  3.86it/s]train loss: 0.5949804186820984
train per: 0.17257892467261296

overall:  24%|██▎       | 940/4000 [04:07<13:02,  3.91it/s]

train:  27%|██▋       | 940/3515 [04:04<10:58,  3.91it/s]
overall:  24%|██▎       | 941/4000 [04:07<13:13,  3.85it/s]

train:  27%|██▋       | 941/3515 [04:04<11:07,  3.85it/s]
overall:  24%|██▎       | 942/4000 [04:07<13:15,  3.85it/s]

train:  27%|██▋       | 942/3515 [04:05<11:08,  3.85it/s]
overall:  24%|██▎       | 943/4000 [04:07<13:27,  3.78it/s]

train:  27%|██▋       | 943/3515 [04:05<11:19,  3.79it/s]
overall:  24%|██▎       | 944/4000 [04:08<13:34,  3.75it/s]

train:  27%|██▋       | 944/3515 [04:05<11:25,  3.75it/s]
overall:  24%|██▎       | 945/4000 [04:08<13:16,  3.84it/s]

train:  27%|██▋       | 945/3515 [04:05<11:09,  3.84it/s]
overall:  24%|██▎       | 946/4000 [04:08<16:24,  3.10it/s]

train:  27%|██▋       | 946/3515 [04:06<13:48,  3.10it/s]
overall:  24%|██▎       | 947/4000 [04:09<15:41,  3.24it/s]

train:  27%|██▋       | 947/3515 [04:06<13:11,  3.24it/s]
overall:  24%|██▎       | 948/4000 [04:09<15:00,  3.39it/s]

train:  27%|██▋       | 948/3515 [04:06<12:37,  3.39it/s]
overall:  24%|██▎       | 949/4000 [04:09<14:39,  3.47it/s]

train:  27%|██▋       | 949/3515 [04:07<12:19,  3.47it/s]
overall:  24%|██▍       | 950/4000 [04:10<14:16,  3.56it/s]

train:  27%|██▋       | 950/3515 [04:07<12:00,  3.56it/s]
overall:  24%|██▍       | 951/4000 [04:10<15:13,  3.34it/s]

train:  27%|██▋       | 951/3515 [04:07<12:47,  3.34it/s]
overall:  24%|██▍       | 952/4000 [04:10<12:27,  4.08it/s]

train:  27%|██▋       | 952/3515 [04:07<10:28,  4.08it/s]
overall:  24%|██▍       | 953/4000 [04:10<12:27,  4.08it/s]

train:  27%|██▋       | 953/3515 [04:08<10:28,  4.08it/s]
overall:  24%|██▍       | 954/4000 [04:10<12:25,  4.09it/s]

train:  27%|██▋       | 954/3515 [04:08<10:26,  4.09it/s]
overall:  24%|██▍       | 955/4000 [04:11<12:50,  3.95it/s]

train:  27%|██▋       | 955/3515 [04:08<10:47,  3.95it/s]
overall:  24%|██▍       | 956/4000 [04:11<13:06,  3.87it/s]

train:  27%|██▋       | 956/3515 [04:08<11:01,  3.87it/s]
overall:  24%|██▍       | 957/4000 [04:11<13:08,  3.86it/s]

train:  27%|██▋       | 957/3515 [04:09<11:02,  3.86it/s]
overall:  24%|██▍       | 958/4000 [04:12<12:50,  3.95it/s]

train:  27%|██▋       | 958/3515 [04:09<10:47,  3.95it/s]
overall:  24%|██▍       | 959/4000 [04:12<12:56,  3.92it/s]

train:  27%|██▋       | 959/3515 [04:09<10:52,  3.92it/s]train loss: 0.607603907585144
train per: 0.169238522642732

overall:  24%|██▍       | 960/4000 [04:12<12:59,  3.90it/s]

train:  27%|██▋       | 960/3515 [04:09<10:55,  3.90it/s]
overall:  24%|██▍       | 961/4000 [04:12<12:42,  3.99it/s]

train:  27%|██▋       | 961/3515 [04:10<10:40,  3.99it/s]
overall:  24%|██▍       | 963/4000 [04:13<10:49,  4.68it/s]

train:  27%|██▋       | 963/3515 [04:10<09:05,  4.68it/s]
overall:  24%|██▍       | 964/4000 [04:13<11:21,  4.45it/s]

train:  27%|██▋       | 964/3515 [04:10<09:32,  4.45it/s]
overall:  24%|██▍       | 965/4000 [04:13<12:04,  4.19it/s]

train:  27%|██▋       | 965/3515 [04:11<10:09,  4.19it/s]
overall:  24%|██▍       | 966/4000 [04:13<12:19,  4.10it/s]

train:  27%|██▋       | 966/3515 [04:11<10:21,  4.10it/s]
overall:  24%|██▍       | 967/4000 [04:14<12:41,  3.98it/s]

train:  28%|██▊       | 967/3515 [04:11<10:39,  3.98it/s]
overall:  24%|██▍       | 968/4000 [04:14<12:29,  4.05it/s]

train:  28%|██▊       | 968/3515 [04:11<10:29,  4.05it/s]
overall:  24%|██▍       | 969/4000 [04:14<12:54,  3.91it/s]

train:  28%|██▊       | 969/3515 [04:12<10:50,  3.91it/s]
overall:  24%|██▍       | 970/4000 [04:15<14:19,  3.52it/s]

train:  28%|██▊       | 970/3515 [04:12<12:02,  3.52it/s]
overall:  24%|██▍       | 971/4000 [04:15<12:23,  4.08it/s]

train:  28%|██▊       | 971/3515 [04:12<10:24,  4.08it/s]
overall:  24%|██▍       | 972/4000 [04:15<12:25,  4.06it/s]

train:  28%|██▊       | 972/3515 [04:12<10:26,  4.06it/s]
overall:  24%|██▍       | 973/4000 [04:15<12:35,  4.01it/s]

train:  28%|██▊       | 973/3515 [04:13<10:34,  4.01it/s]
overall:  24%|██▍       | 974/4000 [04:15<12:52,  3.92it/s]

train:  28%|██▊       | 974/3515 [04:13<10:49,  3.92it/s]
overall:  24%|██▍       | 975/4000 [04:16<14:17,  3.53it/s]

train:  28%|██▊       | 975/3515 [04:13<12:00,  3.53it/s]
overall:  24%|██▍       | 976/4000 [04:16<13:56,  3.61it/s]

train:  28%|██▊       | 976/3515 [04:13<11:42,  3.61it/s]
overall:  24%|██▍       | 977/4000 [04:16<11:18,  4.46it/s]

train:  28%|██▊       | 977/3515 [04:14<09:29,  4.46it/s]
overall:  24%|██▍       | 978/4000 [04:16<11:35,  4.35it/s]

train:  28%|██▊       | 978/3515 [04:14<09:43,  4.35it/s]
overall:  24%|██▍       | 979/4000 [04:17<11:57,  4.21it/s]

train:  28%|██▊       | 979/3515 [04:14<10:01,  4.21it/s]train loss: 0.6004529595375061
train per: 0.17006894687035284

overall:  24%|██▍       | 980/4000 [04:17<12:32,  4.01it/s]

train:  28%|██▊       | 980/3515 [04:14<10:32,  4.01it/s]
overall:  25%|██▍       | 981/4000 [04:17<12:58,  3.88it/s]

train:  28%|██▊       | 981/3515 [04:15<10:53,  3.88it/s]
overall:  25%|██▍       | 982/4000 [04:17<13:10,  3.82it/s]

train:  28%|██▊       | 982/3515 [04:15<11:03,  3.82it/s]
overall:  25%|██▍       | 983/4000 [04:18<10:47,  4.66it/s]

train:  28%|██▊       | 983/3515 [04:15<09:03,  4.66it/s]
overall:  25%|██▍       | 984/4000 [04:18<11:14,  4.47it/s]

train:  28%|██▊       | 984/3515 [04:15<09:25,  4.47it/s]
overall:  25%|██▍       | 985/4000 [04:18<11:49,  4.25it/s]

train:  28%|██▊       | 985/3515 [04:15<09:55,  4.25it/s]
overall:  25%|██▍       | 986/4000 [04:18<12:19,  4.07it/s]

train:  28%|██▊       | 986/3515 [04:16<10:20,  4.07it/s]
overall:  25%|██▍       | 987/4000 [04:19<12:17,  4.09it/s]

train:  28%|██▊       | 987/3515 [04:16<10:18,  4.09it/s]
overall:  25%|██▍       | 988/4000 [04:19<12:44,  3.94it/s]

train:  28%|██▊       | 988/3515 [04:16<10:41,  3.94it/s]
overall:  25%|██▍       | 989/4000 [04:19<13:03,  3.84it/s]

train:  28%|██▊       | 989/3515 [04:17<10:57,  3.84it/s]
overall:  25%|██▍       | 990/4000 [04:19<13:11,  3.80it/s]

train:  28%|██▊       | 990/3515 [04:17<11:04,  3.80it/s]
overall:  25%|██▍       | 991/4000 [04:20<13:23,  3.75it/s]

train:  28%|██▊       | 991/3515 [04:17<11:13,  3.75it/s]
overall:  25%|██▍       | 992/4000 [04:20<13:21,  3.75it/s]

train:  28%|██▊       | 992/3515 [04:17<11:12,  3.75it/s]
overall:  25%|██▍       | 993/4000 [04:20<13:29,  3.72it/s]

train:  28%|██▊       | 993/3515 [04:18<11:18,  3.72it/s]
overall:  25%|██▍       | 994/4000 [04:21<16:30,  3.03it/s]

train:  28%|██▊       | 994/3515 [04:18<13:50,  3.03it/s]
overall:  25%|██▍       | 995/4000 [04:21<15:37,  3.20it/s]

train:  28%|██▊       | 995/3515 [04:18<13:06,  3.20it/s]
overall:  25%|██▍       | 996/4000 [04:21<14:34,  3.43it/s]

train:  28%|██▊       | 996/3515 [04:19<12:13,  3.43it/s]
overall:  25%|██▍       | 997/4000 [04:22<14:30,  3.45it/s]

train:  28%|██▊       | 997/3515 [04:19<12:09,  3.45it/s]
overall:  25%|██▍       | 998/4000 [04:22<14:04,  3.55it/s]

train:  28%|██▊       | 998/3515 [04:19<11:48,  3.55it/s]
overall:  25%|██▍       | 999/4000 [04:22<13:43,  3.64it/s]

train:  28%|██▊       | 999/3515 [04:19<11:30,  3.64it/s]train loss: 0.6399202346801758
train per: 0.17385126673730855
[LibriPhone] - Taking the first phoneme sequences for a deterministic behavior.



word -> phonemes:   0%|          | 0/2703 [00:00<?, ?it/s]


word -> phonemes:  12%|█▏        | 335/2703 [00:00<00:00, 3340.13it/s]


word -> phonemes:  25%|██▍       | 670/2703 [00:00<00:00, 3147.07it/s]


word -> phonemes:  37%|███▋      | 1004/2703 [00:00<00:00, 3228.01it/s]


word -> phonemes:  49%|████▉     | 1336/2703 [00:00<00:00, 3262.68it/s]


word -> phonemes:  62%|██████▏   | 1663/2703 [00:00<00:00, 3101.60it/s]


word -> phonemes:  74%|███████▎  | 1993/2703 [00:00<00:00, 3166.07it/s]


word -> phonemes:  86%|████████▌ | 2323/2703 [00:00<00:00, 3208.05it/s]


word -> phonemes:  98%|█████████▊| 2645/2703 [00:00<00:00, 3189.01it/s]
word -> phonemes: 100%|██████████| 2703/2703 [00:00<00:00, 3188.09it/s]



dev:   0%|          | 0/2703 [00:00<?, ?it/s]


dev:   0%|          | 1/2703 [00:00<32:15,  1.40it/s]


dev:   0%|          | 4/2703 [00:00<07:50,  5.74it/s]


dev:   0%|          | 6/2703 [00:00<05:24,  8.30it/s]


dev:   0%|          | 9/2703 [00:01<03:46, 11.87it/s]


dev:   0%|          | 12/2703 [00:01<03:03, 14.65it/s]


dev:   1%|          | 15/2703 [00:01<02:35, 17.29it/s]


dev:   1%|          | 18/2703 [00:01<02:22, 18.83it/s]


dev:   1%|          | 21/2703 [00:01<02:11, 20.42it/s]


dev:   1%|          | 24/2703 [00:01<02:05, 21.33it/s]


dev:   1%|          | 27/2703 [00:01<02:00, 22.22it/s]


dev:   1%|          | 30/2703 [00:01<01:59, 22.41it/s]


dev:   1%|          | 33/2703 [00:02<01:51, 23.85it/s]


dev:   1%|▏         | 36/2703 [00:02<01:49, 24.45it/s]


dev:   1%|▏         | 39/2703 [00:02<01:47, 24.90it/s]


dev:   2%|▏         | 42/2703 [00:02<01:46, 24.95it/s]


dev:   2%|▏         | 45/2703 [00:02<01:43, 25.78it/s]


dev:   2%|▏         | 48/2703 [00:02<01:44, 25.49it/s]


dev:   2%|▏         | 51/2703 [00:02<01:41, 26.05it/s]


dev:   2%|▏         | 54/2703 [00:02<01:42, 25.83it/s]


dev:   2%|▏         | 58/2703 [00:03<01:36, 27.38it/s]


dev:   2%|▏         | 61/2703 [00:03<01:35, 27.57it/s]


dev:   2%|▏         | 65/2703 [00:03<01:32, 28.66it/s]


dev:   3%|▎         | 68/2703 [00:03<01:36, 27.22it/s]


dev:   3%|▎         | 72/2703 [00:03<01:34, 27.71it/s]


dev:   3%|▎         | 75/2703 [00:03<01:34, 27.87it/s]


dev:   3%|▎         | 78/2703 [00:03<01:33, 27.93it/s]


dev:   3%|▎         | 82/2703 [00:03<01:29, 29.28it/s]


dev:   3%|▎         | 85/2703 [00:03<01:31, 28.60it/s]


dev:   3%|▎         | 89/2703 [00:04<01:29, 29.23it/s]


dev:   3%|▎         | 92/2703 [00:04<01:31, 28.47it/s]


dev:   4%|▎         | 96/2703 [00:04<01:30, 28.66it/s]


dev:   4%|▎         | 100/2703 [00:04<01:28, 29.35it/s]


dev:   4%|▍         | 103/2703 [00:04<01:30, 28.69it/s]


dev:   4%|▍         | 107/2703 [00:04<01:27, 29.68it/s]


dev:   4%|▍         | 111/2703 [00:04<01:27, 29.55it/s]


dev:   4%|▍         | 114/2703 [00:04<01:28, 29.13it/s]


dev:   4%|▍         | 117/2703 [00:05<01:28, 29.15it/s]


dev:   4%|▍         | 121/2703 [00:05<01:26, 29.84it/s]


dev:   5%|▍         | 124/2703 [00:05<01:27, 29.40it/s]


dev:   5%|▍         | 128/2703 [00:05<01:26, 29.78it/s]


dev:   5%|▍         | 132/2703 [00:05<01:24, 30.26it/s]


dev:   5%|▌         | 136/2703 [00:05<01:24, 30.31it/s]


dev:   5%|▌         | 140/2703 [00:05<01:22, 30.97it/s]


dev:   5%|▌         | 144/2703 [00:05<01:23, 30.47it/s]


dev:   5%|▌         | 148/2703 [00:06<01:22, 31.03it/s]


dev:   6%|▌         | 152/2703 [00:06<01:23, 30.46it/s]


dev:   6%|▌         | 156/2703 [00:06<01:22, 30.90it/s]


dev:   6%|▌         | 160/2703 [00:06<01:21, 31.03it/s]


dev:   6%|▌         | 164/2703 [00:06<01:21, 31.11it/s]


dev:   6%|▌         | 168/2703 [00:06<01:22, 30.76it/s]


dev:   6%|▋         | 172/2703 [00:06<01:18, 32.17it/s]


dev:   7%|▋         | 176/2703 [00:06<01:17, 32.40it/s]


dev:   7%|▋         | 180/2703 [00:07<01:16, 32.78it/s]


dev:   7%|▋         | 184/2703 [00:07<01:18, 32.19it/s]


dev:   7%|▋         | 188/2703 [00:07<01:18, 32.16it/s]


dev:   7%|▋         | 192/2703 [00:07<01:16, 32.74it/s]


dev:   7%|▋         | 196/2703 [00:07<01:15, 33.41it/s]


dev:   7%|▋         | 200/2703 [00:07<01:17, 32.19it/s]


dev:   8%|▊         | 204/2703 [00:07<01:16, 32.78it/s]


dev:   8%|▊         | 208/2703 [00:07<01:16, 32.50it/s]


dev:   8%|▊         | 212/2703 [00:08<01:17, 31.99it/s]


dev:   8%|▊         | 216/2703 [00:08<01:17, 32.15it/s]


dev:   8%|▊         | 220/2703 [00:08<01:14, 33.34it/s]


dev:   8%|▊         | 224/2703 [00:08<01:13, 33.82it/s]


dev:   8%|▊         | 228/2703 [00:08<01:10, 35.00it/s]


dev:   9%|▊         | 232/2703 [00:08<01:10, 34.86it/s]


dev:   9%|▊         | 236/2703 [00:08<01:11, 34.27it/s]


dev:   9%|▉         | 240/2703 [00:08<01:12, 34.08it/s]


dev:   9%|▉         | 244/2703 [00:09<01:12, 33.80it/s]


dev:   9%|▉         | 248/2703 [00:09<01:12, 33.80it/s]


dev:   9%|▉         | 252/2703 [00:09<01:12, 33.99it/s]


dev:   9%|▉         | 256/2703 [00:09<01:12, 33.70it/s]


dev:  10%|▉         | 260/2703 [00:09<01:13, 33.40it/s]


dev:  10%|▉         | 264/2703 [00:09<01:12, 33.42it/s]


dev:  10%|▉         | 268/2703 [00:09<01:12, 33.70it/s]


dev:  10%|█         | 272/2703 [00:09<01:11, 33.83it/s]


dev:  10%|█         | 276/2703 [00:09<01:10, 34.59it/s]


dev:  10%|█         | 280/2703 [00:10<01:09, 34.89it/s]


dev:  11%|█         | 284/2703 [00:10<01:10, 34.26it/s]


dev:  11%|█         | 288/2703 [00:10<01:09, 34.80it/s]


dev:  11%|█         | 292/2703 [00:10<01:11, 33.87it/s]


dev:  11%|█         | 296/2703 [00:10<01:10, 33.99it/s]


dev:  11%|█         | 300/2703 [00:10<01:08, 34.95it/s]


dev:  11%|█         | 304/2703 [00:10<01:07, 35.59it/s]


dev:  11%|█▏        | 308/2703 [00:10<01:08, 35.04it/s]


dev:  12%|█▏        | 312/2703 [00:10<01:09, 34.61it/s]


dev:  12%|█▏        | 316/2703 [00:11<01:08, 34.86it/s]


dev:  12%|█▏        | 320/2703 [00:11<01:09, 34.26it/s]


dev:  12%|█▏        | 324/2703 [00:11<01:11, 33.49it/s]


dev:  12%|█▏        | 328/2703 [00:11<01:08, 34.87it/s]


dev:  12%|█▏        | 333/2703 [00:11<01:05, 36.46it/s]


dev:  12%|█▏        | 337/2703 [00:11<01:05, 36.24it/s]


dev:  13%|█▎        | 341/2703 [00:11<01:04, 36.77it/s]


dev:  13%|█▎        | 345/2703 [00:11<01:04, 36.51it/s]


dev:  13%|█▎        | 349/2703 [00:12<01:06, 35.46it/s]


dev:  13%|█▎        | 353/2703 [00:12<01:07, 34.59it/s]


dev:  13%|█▎        | 357/2703 [00:12<01:08, 34.45it/s]


dev:  13%|█▎        | 361/2703 [00:12<01:07, 34.67it/s]


dev:  14%|█▎        | 366/2703 [00:12<01:04, 36.50it/s]


dev:  14%|█▎        | 370/2703 [00:12<01:04, 36.40it/s]


dev:  14%|█▍        | 374/2703 [00:12<01:03, 36.54it/s]


dev:  14%|█▍        | 378/2703 [00:12<01:06, 35.22it/s]


dev:  14%|█▍        | 382/2703 [00:12<01:07, 34.36it/s]


dev:  14%|█▍        | 386/2703 [00:13<01:05, 35.24it/s]


dev:  14%|█▍        | 390/2703 [00:13<01:06, 35.03it/s]


dev:  15%|█▍        | 394/2703 [00:13<01:07, 34.45it/s]


dev:  15%|█▍        | 398/2703 [00:13<01:06, 34.53it/s]


dev:  15%|█▍        | 402/2703 [00:13<01:07, 33.96it/s]


dev:  15%|█▌        | 406/2703 [00:13<01:07, 33.84it/s]


dev:  15%|█▌        | 410/2703 [00:13<01:07, 33.97it/s]


dev:  15%|█▌        | 414/2703 [00:13<01:06, 34.27it/s]


dev:  15%|█▌        | 418/2703 [00:14<01:05, 35.03it/s]


dev:  16%|█▌        | 422/2703 [00:14<01:05, 34.83it/s]


dev:  16%|█▌        | 426/2703 [00:14<01:04, 35.37it/s]


dev:  16%|█▌        | 431/2703 [00:14<01:02, 36.47it/s]


dev:  16%|█▌        | 435/2703 [00:14<01:00, 37.39it/s]


dev:  16%|█▋        | 440/2703 [00:14<00:59, 38.13it/s]


dev:  16%|█▋        | 444/2703 [00:14<00:59, 37.92it/s]


dev:  17%|█▋        | 449/2703 [00:14<00:58, 38.70it/s]


dev:  17%|█▋        | 454/2703 [00:14<00:56, 39.69it/s]


dev:  17%|█▋        | 458/2703 [00:15<00:58, 38.24it/s]


dev:  17%|█▋        | 463/2703 [00:15<00:56, 39.56it/s]


dev:  17%|█▋        | 468/2703 [00:15<00:55, 40.14it/s]


dev:  17%|█▋        | 473/2703 [00:15<00:56, 39.55it/s]


dev:  18%|█▊        | 477/2703 [00:15<00:56, 39.42it/s]


dev:  18%|█▊        | 481/2703 [00:15<00:56, 39.17it/s]


dev:  18%|█▊        | 486/2703 [00:15<00:55, 39.73it/s]


dev:  18%|█▊        | 490/2703 [00:15<00:58, 37.96it/s]


dev:  18%|█▊        | 494/2703 [00:15<00:59, 37.19it/s]


dev:  18%|█▊        | 499/2703 [00:16<00:56, 39.08it/s]


dev:  19%|█▊        | 503/2703 [00:16<00:56, 39.10it/s]


dev:  19%|█▉        | 507/2703 [00:16<00:56, 39.07it/s]


dev:  19%|█▉        | 511/2703 [00:16<00:57, 38.26it/s]


dev:  19%|█▉        | 515/2703 [00:16<00:57, 38.02it/s]


dev:  19%|█▉        | 520/2703 [00:16<00:55, 39.57it/s]


dev:  19%|█▉        | 525/2703 [00:16<00:52, 41.44it/s]


dev:  20%|█▉        | 530/2703 [00:16<00:53, 40.88it/s]


dev:  20%|█▉        | 535/2703 [00:16<00:52, 40.93it/s]


dev:  20%|█▉        | 540/2703 [00:17<00:53, 40.29it/s]


dev:  20%|██        | 545/2703 [00:17<00:50, 42.76it/s]


dev:  20%|██        | 550/2703 [00:17<00:49, 43.39it/s]


dev:  21%|██        | 555/2703 [00:17<00:48, 43.87it/s]


dev:  21%|██        | 560/2703 [00:17<00:49, 43.32it/s]


dev:  21%|██        | 565/2703 [00:17<00:52, 40.65it/s]


dev:  21%|██        | 570/2703 [00:17<00:52, 40.45it/s]


dev:  21%|██▏       | 575/2703 [00:17<00:53, 39.73it/s]


dev:  21%|██▏       | 581/2703 [00:18<00:49, 43.18it/s]


dev:  22%|██▏       | 586/2703 [00:18<00:48, 43.32it/s]


dev:  22%|██▏       | 591/2703 [00:18<00:47, 44.27it/s]


dev:  22%|██▏       | 596/2703 [00:18<00:48, 43.23it/s]


dev:  22%|██▏       | 601/2703 [00:18<00:47, 44.07it/s]


dev:  22%|██▏       | 606/2703 [00:18<00:46, 45.57it/s]


dev:  23%|██▎       | 611/2703 [00:18<00:47, 44.28it/s]


dev:  23%|██▎       | 616/2703 [00:18<00:45, 45.52it/s]


dev:  23%|██▎       | 621/2703 [00:18<00:44, 46.28it/s]


dev:  23%|██▎       | 626/2703 [00:19<00:47, 44.05it/s]


dev:  23%|██▎       | 632/2703 [00:19<00:43, 47.32it/s]


dev:  24%|██▎       | 637/2703 [00:19<00:45, 45.56it/s]


dev:  24%|██▍       | 642/2703 [00:19<00:46, 44.78it/s]


dev:  24%|██▍       | 648/2703 [00:19<00:43, 46.97it/s]


dev:  24%|██▍       | 653/2703 [00:19<00:44, 45.83it/s]


dev:  24%|██▍       | 658/2703 [00:19<00:46, 44.06it/s]


dev:  25%|██▍       | 663/2703 [00:19<00:45, 44.80it/s]


dev:  25%|██▍       | 668/2703 [00:19<00:44, 45.70it/s]


dev:  25%|██▍       | 673/2703 [00:20<00:45, 44.16it/s]


dev:  25%|██▌       | 678/2703 [00:20<00:46, 43.42it/s]


dev:  25%|██▌       | 683/2703 [00:20<00:46, 43.57it/s]


dev:  25%|██▌       | 688/2703 [00:20<00:47, 42.24it/s]


dev:  26%|██▌       | 693/2703 [00:20<00:47, 42.65it/s]


dev:  26%|██▌       | 698/2703 [00:20<00:48, 41.74it/s]


dev:  26%|██▌       | 703/2703 [00:20<00:45, 43.52it/s]


dev:  26%|██▌       | 708/2703 [00:20<00:44, 44.78it/s]


dev:  26%|██▋       | 713/2703 [00:21<00:46, 42.72it/s]


dev:  27%|██▋       | 718/2703 [00:21<00:46, 43.04it/s]


dev:  27%|██▋       | 723/2703 [00:21<00:45, 43.49it/s]


dev:  27%|██▋       | 728/2703 [00:21<00:43, 45.15it/s]


dev:  27%|██▋       | 733/2703 [00:21<00:44, 44.44it/s]


dev:  27%|██▋       | 738/2703 [00:21<00:44, 44.44it/s]


dev:  28%|██▊       | 744/2703 [00:21<00:42, 46.27it/s]


dev:  28%|██▊       | 749/2703 [00:21<00:41, 46.68it/s]


dev:  28%|██▊       | 754/2703 [00:21<00:44, 43.90it/s]


dev:  28%|██▊       | 759/2703 [00:22<00:45, 43.14it/s]


dev:  28%|██▊       | 764/2703 [00:22<00:44, 43.78it/s]


dev:  28%|██▊       | 769/2703 [00:22<00:45, 42.70it/s]


dev:  29%|██▊       | 774/2703 [00:22<00:44, 43.25it/s]


dev:  29%|██▉       | 779/2703 [00:22<00:45, 42.12it/s]


dev:  29%|██▉       | 784/2703 [00:22<00:44, 42.92it/s]


dev:  29%|██▉       | 789/2703 [00:22<00:44, 43.30it/s]


dev:  29%|██▉       | 794/2703 [00:22<00:45, 42.00it/s]


dev:  30%|██▉       | 799/2703 [00:23<00:44, 43.25it/s]


dev:  30%|██▉       | 804/2703 [00:23<00:45, 42.16it/s]


dev:  30%|██▉       | 809/2703 [00:23<00:43, 43.82it/s]


dev:  30%|███       | 815/2703 [00:23<00:41, 45.97it/s]


dev:  30%|███       | 820/2703 [00:23<00:41, 45.36it/s]


dev:  31%|███       | 826/2703 [00:23<00:39, 47.80it/s]


dev:  31%|███       | 831/2703 [00:23<00:39, 46.85it/s]


dev:  31%|███       | 836/2703 [00:23<00:40, 46.01it/s]


dev:  31%|███       | 842/2703 [00:23<00:39, 46.96it/s]


dev:  31%|███▏      | 847/2703 [00:24<00:41, 45.04it/s]


dev:  32%|███▏      | 852/2703 [00:24<00:41, 44.50it/s]


dev:  32%|███▏      | 857/2703 [00:24<00:40, 45.45it/s]


dev:  32%|███▏      | 862/2703 [00:24<00:40, 45.77it/s]


dev:  32%|███▏      | 867/2703 [00:24<00:42, 43.63it/s]


dev:  32%|███▏      | 872/2703 [00:24<00:40, 44.86it/s]


dev:  32%|███▏      | 877/2703 [00:24<00:41, 44.38it/s]


dev:  33%|███▎      | 883/2703 [00:24<00:38, 46.71it/s]


dev:  33%|███▎      | 888/2703 [00:24<00:40, 44.75it/s]


dev:  33%|███▎      | 893/2703 [00:25<00:40, 44.31it/s]


dev:  33%|███▎      | 898/2703 [00:25<00:39, 45.21it/s]


dev:  33%|███▎      | 903/2703 [00:25<00:40, 44.17it/s]


dev:  34%|███▎      | 908/2703 [00:25<00:40, 44.10it/s]


dev:  34%|███▍      | 913/2703 [00:25<00:40, 44.20it/s]


dev:  34%|███▍      | 919/2703 [00:25<00:37, 48.05it/s]


dev:  34%|███▍      | 924/2703 [00:25<00:37, 47.16it/s]


dev:  34%|███▍      | 929/2703 [00:25<00:37, 47.48it/s]


dev:  35%|███▍      | 935/2703 [00:25<00:35, 49.98it/s]


dev:  35%|███▍      | 941/2703 [00:26<00:34, 51.62it/s]


dev:  35%|███▌      | 947/2703 [00:26<00:34, 50.32it/s]


dev:  35%|███▌      | 953/2703 [00:26<00:34, 50.87it/s]


dev:  35%|███▌      | 959/2703 [00:26<00:34, 50.43it/s]


dev:  36%|███▌      | 965/2703 [00:26<00:33, 51.73it/s]


dev:  36%|███▌      | 971/2703 [00:26<00:34, 50.35it/s]


dev:  36%|███▌      | 977/2703 [00:26<00:37, 46.24it/s]


dev:  36%|███▋      | 983/2703 [00:26<00:36, 47.70it/s]


dev:  37%|███▋      | 989/2703 [00:27<00:34, 50.39it/s]


dev:  37%|███▋      | 995/2703 [00:27<00:33, 50.70it/s]


dev:  37%|███▋      | 1001/2703 [00:27<00:35, 48.61it/s]


dev:  37%|███▋      | 1007/2703 [00:27<00:34, 49.48it/s]


dev:  37%|███▋      | 1012/2703 [00:27<00:34, 49.52it/s]


dev:  38%|███▊      | 1018/2703 [00:27<00:33, 50.69it/s]


dev:  38%|███▊      | 1024/2703 [00:27<00:32, 52.13it/s]


dev:  38%|███▊      | 1030/2703 [00:27<00:32, 50.93it/s]


dev:  38%|███▊      | 1036/2703 [00:27<00:33, 50.21it/s]


dev:  39%|███▊      | 1042/2703 [00:28<00:33, 49.27it/s]


dev:  39%|███▉      | 1048/2703 [00:28<00:33, 49.66it/s]


dev:  39%|███▉      | 1054/2703 [00:28<00:32, 50.34it/s]


dev:  39%|███▉      | 1060/2703 [00:28<00:31, 51.43it/s]


dev:  39%|███▉      | 1066/2703 [00:28<00:32, 51.05it/s]


dev:  40%|███▉      | 1072/2703 [00:28<00:32, 50.22it/s]


dev:  40%|███▉      | 1078/2703 [00:28<00:33, 49.10it/s]


dev:  40%|████      | 1084/2703 [00:28<00:31, 51.05it/s]


dev:  40%|████      | 1090/2703 [00:29<00:30, 53.36it/s]


dev:  41%|████      | 1096/2703 [00:29<00:30, 52.71it/s]


dev:  41%|████      | 1102/2703 [00:29<00:29, 53.89it/s]


dev:  41%|████      | 1108/2703 [00:29<00:31, 51.06it/s]


dev:  41%|████      | 1114/2703 [00:29<00:31, 51.12it/s]


dev:  41%|████▏     | 1120/2703 [00:29<00:31, 49.60it/s]


dev:  42%|████▏     | 1125/2703 [00:29<00:32, 49.19it/s]


dev:  42%|████▏     | 1130/2703 [00:29<00:31, 49.30it/s]


dev:  42%|████▏     | 1135/2703 [00:29<00:32, 48.85it/s]


dev:  42%|████▏     | 1140/2703 [00:30<00:32, 47.87it/s]


dev:  42%|████▏     | 1145/2703 [00:30<00:32, 48.18it/s]


dev:  43%|████▎     | 1150/2703 [00:30<00:32, 48.28it/s]


dev:  43%|████▎     | 1155/2703 [00:30<00:32, 48.33it/s]


dev:  43%|████▎     | 1161/2703 [00:30<00:30, 50.50it/s]


dev:  43%|████▎     | 1167/2703 [00:30<00:31, 49.54it/s]


dev:  43%|████▎     | 1173/2703 [00:30<00:30, 50.47it/s]


dev:  44%|████▎     | 1179/2703 [00:30<00:31, 48.59it/s]


dev:  44%|████▍     | 1185/2703 [00:30<00:29, 50.80it/s]


dev:  44%|████▍     | 1191/2703 [00:31<00:30, 48.85it/s]


dev:  44%|████▍     | 1197/2703 [00:31<00:30, 49.68it/s]


dev:  45%|████▍     | 1203/2703 [00:31<00:30, 49.31it/s]


dev:  45%|████▍     | 1209/2703 [00:31<00:28, 51.61it/s]


dev:  45%|████▍     | 1215/2703 [00:31<00:29, 51.17it/s]


dev:  45%|████▌     | 1221/2703 [00:31<00:28, 51.72it/s]


dev:  45%|████▌     | 1227/2703 [00:31<00:27, 53.09it/s]


dev:  46%|████▌     | 1233/2703 [00:31<00:29, 49.33it/s]


dev:  46%|████▌     | 1239/2703 [00:32<00:29, 49.93it/s]


dev:  46%|████▌     | 1245/2703 [00:32<00:28, 50.89it/s]


dev:  46%|████▋     | 1251/2703 [00:32<00:29, 49.87it/s]


dev:  47%|████▋     | 1257/2703 [00:32<00:28, 50.36it/s]


dev:  47%|████▋     | 1263/2703 [00:32<00:28, 50.07it/s]


dev:  47%|████▋     | 1269/2703 [00:32<00:27, 52.59it/s]


dev:  47%|████▋     | 1275/2703 [00:32<00:28, 49.72it/s]


dev:  47%|████▋     | 1281/2703 [00:32<00:28, 49.28it/s]


dev:  48%|████▊     | 1287/2703 [00:32<00:28, 50.53it/s]


dev:  48%|████▊     | 1293/2703 [00:33<00:28, 49.70it/s]


dev:  48%|████▊     | 1299/2703 [00:33<00:28, 48.78it/s]


dev:  48%|████▊     | 1305/2703 [00:33<00:28, 49.06it/s]


dev:  49%|████▊     | 1311/2703 [00:33<00:27, 50.63it/s]


dev:  49%|████▊     | 1317/2703 [00:33<00:26, 52.61it/s]


dev:  49%|████▉     | 1323/2703 [00:33<00:26, 53.03it/s]


dev:  49%|████▉     | 1329/2703 [00:33<00:27, 50.54it/s]


dev:  49%|████▉     | 1335/2703 [00:33<00:26, 51.47it/s]


dev:  50%|████▉     | 1342/2703 [00:34<00:24, 55.15it/s]


dev:  50%|████▉     | 1348/2703 [00:34<00:24, 56.11it/s]


dev:  50%|█████     | 1354/2703 [00:34<00:24, 54.20it/s]


dev:  50%|█████     | 1360/2703 [00:34<00:25, 53.52it/s]


dev:  51%|█████     | 1366/2703 [00:34<00:24, 54.99it/s]


dev:  51%|█████     | 1372/2703 [00:34<00:25, 52.03it/s]


dev:  51%|█████     | 1378/2703 [00:34<00:25, 52.58it/s]


dev:  51%|█████▏    | 1386/2703 [00:34<00:23, 56.59it/s]


dev:  51%|█████▏    | 1392/2703 [00:34<00:24, 53.39it/s]


dev:  52%|█████▏    | 1398/2703 [00:35<00:23, 55.07it/s]


dev:  52%|█████▏    | 1404/2703 [00:35<00:23, 55.17it/s]


dev:  52%|█████▏    | 1410/2703 [00:35<00:23, 54.96it/s]


dev:  52%|█████▏    | 1417/2703 [00:35<00:22, 56.57it/s]


dev:  53%|█████▎    | 1424/2703 [00:35<00:22, 57.97it/s]


dev:  53%|█████▎    | 1431/2703 [00:35<00:20, 60.89it/s]


dev:  53%|█████▎    | 1438/2703 [00:35<00:22, 56.76it/s]


dev:  53%|█████▎    | 1444/2703 [00:35<00:22, 55.00it/s]


dev:  54%|█████▎    | 1450/2703 [00:35<00:22, 55.50it/s]


dev:  54%|█████▍    | 1456/2703 [00:36<00:22, 56.18it/s]


dev:  54%|█████▍    | 1462/2703 [00:36<00:23, 53.76it/s]


dev:  54%|█████▍    | 1468/2703 [00:36<00:22, 53.73it/s]


dev:  55%|█████▍    | 1474/2703 [00:36<00:23, 53.39it/s]


dev:  55%|█████▍    | 1480/2703 [00:36<00:23, 52.88it/s]


dev:  55%|█████▍    | 1486/2703 [00:36<00:23, 51.70it/s]


dev:  55%|█████▌    | 1492/2703 [00:36<00:22, 53.49it/s]


dev:  55%|█████▌    | 1498/2703 [00:36<00:22, 53.72it/s]


dev:  56%|█████▌    | 1505/2703 [00:36<00:21, 56.60it/s]


dev:  56%|█████▌    | 1512/2703 [00:37<00:19, 59.98it/s]


dev:  56%|█████▌    | 1519/2703 [00:37<00:21, 56.14it/s]


dev:  56%|█████▋    | 1526/2703 [00:37<00:19, 58.86it/s]


dev:  57%|█████▋    | 1532/2703 [00:37<00:19, 59.08it/s]


dev:  57%|█████▋    | 1539/2703 [00:37<00:19, 58.48it/s]


dev:  57%|█████▋    | 1545/2703 [00:37<00:19, 58.62it/s]


dev:  57%|█████▋    | 1551/2703 [00:37<00:20, 56.49it/s]


dev:  58%|█████▊    | 1558/2703 [00:37<00:19, 59.42it/s]


dev:  58%|█████▊    | 1565/2703 [00:37<00:18, 61.46it/s]


dev:  58%|█████▊    | 1572/2703 [00:38<00:18, 60.65it/s]


dev:  58%|█████▊    | 1579/2703 [00:38<00:18, 59.68it/s]


dev:  59%|█████▊    | 1585/2703 [00:38<00:18, 59.57it/s]


dev:  59%|█████▉    | 1591/2703 [00:38<00:18, 59.30it/s]


dev:  59%|█████▉    | 1597/2703 [00:38<00:19, 56.46it/s]


dev:  59%|█████▉    | 1604/2703 [00:38<00:18, 59.33it/s]


dev:  60%|█████▉    | 1610/2703 [00:38<00:19, 56.51it/s]


dev:  60%|█████▉    | 1616/2703 [00:38<00:19, 56.35it/s]


dev:  60%|██████    | 1623/2703 [00:38<00:18, 57.50it/s]


dev:  60%|██████    | 1630/2703 [00:39<00:18, 59.04it/s]


dev:  61%|██████    | 1638/2703 [00:39<00:16, 62.82it/s]


dev:  61%|██████    | 1645/2703 [00:39<00:16, 62.78it/s]


dev:  61%|██████    | 1652/2703 [00:39<00:16, 63.12it/s]


dev:  61%|██████▏   | 1659/2703 [00:39<00:16, 61.57it/s]


dev:  62%|██████▏   | 1666/2703 [00:39<00:16, 61.81it/s]


dev:  62%|██████▏   | 1673/2703 [00:39<00:16, 61.00it/s]


dev:  62%|██████▏   | 1680/2703 [00:39<00:17, 59.14it/s]


dev:  62%|██████▏   | 1686/2703 [00:40<00:17, 58.44it/s]


dev:  63%|██████▎   | 1693/2703 [00:40<00:16, 59.53it/s]


dev:  63%|██████▎   | 1699/2703 [00:40<00:17, 56.43it/s]


dev:  63%|██████▎   | 1705/2703 [00:40<00:17, 56.87it/s]


dev:  63%|██████▎   | 1712/2703 [00:40<00:16, 58.59it/s]


dev:  64%|██████▎   | 1719/2703 [00:40<00:16, 60.42it/s]


dev:  64%|██████▍   | 1726/2703 [00:40<00:15, 61.64it/s]


dev:  64%|██████▍   | 1733/2703 [00:40<00:15, 63.22it/s]


dev:  64%|██████▍   | 1740/2703 [00:40<00:14, 64.57it/s]


dev:  65%|██████▍   | 1747/2703 [00:41<00:14, 64.00it/s]


dev:  65%|██████▍   | 1754/2703 [00:41<00:14, 63.50it/s]


dev:  65%|██████▌   | 1761/2703 [00:41<00:14, 64.49it/s]


dev:  65%|██████▌   | 1768/2703 [00:41<00:15, 60.91it/s]


dev:  66%|██████▌   | 1775/2703 [00:41<00:15, 61.50it/s]


dev:  66%|██████▌   | 1782/2703 [00:41<00:14, 62.95it/s]


dev:  66%|██████▌   | 1789/2703 [00:41<00:14, 62.14it/s]


dev:  66%|██████▋   | 1796/2703 [00:41<00:14, 60.47it/s]


dev:  67%|██████▋   | 1803/2703 [00:41<00:14, 61.81it/s]


dev:  67%|██████▋   | 1811/2703 [00:42<00:13, 64.92it/s]


dev:  67%|██████▋   | 1818/2703 [00:42<00:13, 64.71it/s]


dev:  68%|██████▊   | 1825/2703 [00:42<00:14, 60.72it/s]


dev:  68%|██████▊   | 1832/2703 [00:42<00:14, 60.95it/s]


dev:  68%|██████▊   | 1839/2703 [00:42<00:14, 60.19it/s]


dev:  68%|██████▊   | 1846/2703 [00:42<00:14, 60.25it/s]


dev:  69%|██████▊   | 1853/2703 [00:42<00:14, 59.86it/s]


dev:  69%|██████▉   | 1860/2703 [00:42<00:14, 59.19it/s]


dev:  69%|██████▉   | 1867/2703 [00:42<00:13, 60.07it/s]


dev:  69%|██████▉   | 1874/2703 [00:43<00:13, 59.66it/s]


dev:  70%|██████▉   | 1880/2703 [00:43<00:13, 59.75it/s]


dev:  70%|██████▉   | 1887/2703 [00:43<00:13, 58.94it/s]


dev:  70%|███████   | 1893/2703 [00:43<00:13, 58.13it/s]


dev:  70%|███████   | 1900/2703 [00:43<00:13, 59.47it/s]


dev:  71%|███████   | 1907/2703 [00:43<00:12, 62.09it/s]


dev:  71%|███████   | 1914/2703 [00:43<00:12, 62.01it/s]


dev:  71%|███████   | 1921/2703 [00:43<00:12, 63.78it/s]


dev:  71%|███████▏  | 1928/2703 [00:43<00:12, 63.33it/s]


dev:  72%|███████▏  | 1935/2703 [00:44<00:12, 62.08it/s]


dev:  72%|███████▏  | 1942/2703 [00:44<00:12, 59.20it/s]


dev:  72%|███████▏  | 1950/2703 [00:44<00:11, 62.81it/s]


dev:  72%|███████▏  | 1957/2703 [00:44<00:11, 62.19it/s]


dev:  73%|███████▎  | 1964/2703 [00:44<00:12, 61.17it/s]


dev:  73%|███████▎  | 1972/2703 [00:44<00:11, 65.04it/s]


dev:  73%|███████▎  | 1979/2703 [00:44<00:11, 64.90it/s]


dev:  73%|███████▎  | 1986/2703 [00:44<00:11, 63.55it/s]


dev:  74%|███████▎  | 1993/2703 [00:45<00:11, 60.99it/s]


dev:  74%|███████▍  | 2000/2703 [00:45<00:11, 60.62it/s]


dev:  74%|███████▍  | 2008/2703 [00:45<00:10, 64.03it/s]


dev:  75%|███████▍  | 2016/2703 [00:45<00:10, 65.39it/s]


dev:  75%|███████▍  | 2023/2703 [00:45<00:10, 65.90it/s]


dev:  75%|███████▌  | 2031/2703 [00:45<00:09, 68.40it/s]


dev:  75%|███████▌  | 2038/2703 [00:45<00:10, 62.96it/s]


dev:  76%|███████▌  | 2045/2703 [00:45<00:10, 63.25it/s]


dev:  76%|███████▌  | 2052/2703 [00:45<00:10, 63.55it/s]


dev:  76%|███████▌  | 2059/2703 [00:46<00:10, 59.94it/s]


dev:  76%|███████▋  | 2066/2703 [00:46<00:10, 60.63it/s]


dev:  77%|███████▋  | 2073/2703 [00:46<00:10, 61.70it/s]


dev:  77%|███████▋  | 2080/2703 [00:46<00:10, 61.41it/s]


dev:  77%|███████▋  | 2087/2703 [00:46<00:09, 62.43it/s]


dev:  77%|███████▋  | 2094/2703 [00:46<00:10, 60.89it/s]


dev:  78%|███████▊  | 2101/2703 [00:46<00:09, 60.50it/s]


dev:  78%|███████▊  | 2109/2703 [00:46<00:09, 64.37it/s]


dev:  78%|███████▊  | 2116/2703 [00:46<00:09, 64.06it/s]


dev:  79%|███████▊  | 2123/2703 [00:47<00:09, 61.20it/s]


dev:  79%|███████▉  | 2130/2703 [00:47<00:09, 62.54it/s]


dev:  79%|███████▉  | 2137/2703 [00:47<00:09, 61.99it/s]


dev:  79%|███████▉  | 2144/2703 [00:47<00:09, 60.70it/s]


dev:  80%|███████▉  | 2151/2703 [00:47<00:08, 61.65it/s]


dev:  80%|███████▉  | 2158/2703 [00:47<00:08, 61.92it/s]


dev:  80%|████████  | 2165/2703 [00:47<00:08, 63.83it/s]


dev:  80%|████████  | 2172/2703 [00:47<00:09, 58.31it/s]


dev:  81%|████████  | 2180/2703 [00:47<00:08, 62.26it/s]


dev:  81%|████████  | 2187/2703 [00:48<00:08, 63.39it/s]


dev:  81%|████████  | 2194/2703 [00:48<00:08, 62.73it/s]


dev:  81%|████████▏ | 2201/2703 [00:48<00:08, 61.62it/s]


dev:  82%|████████▏ | 2208/2703 [00:48<00:08, 59.91it/s]


dev:  82%|████████▏ | 2216/2703 [00:48<00:07, 61.86it/s]


dev:  82%|████████▏ | 2223/2703 [00:48<00:07, 61.42it/s]


dev:  83%|████████▎ | 2230/2703 [00:48<00:07, 62.18it/s]


dev:  83%|████████▎ | 2237/2703 [00:48<00:07, 62.35it/s]


dev:  83%|████████▎ | 2244/2703 [00:49<00:07, 59.85it/s]


dev:  83%|████████▎ | 2251/2703 [00:49<00:07, 61.47it/s]


dev:  84%|████████▎ | 2258/2703 [00:49<00:07, 60.63it/s]


dev:  84%|████████▍ | 2265/2703 [00:49<00:07, 59.65it/s]


dev:  84%|████████▍ | 2272/2703 [00:49<00:06, 62.22it/s]


dev:  84%|████████▍ | 2279/2703 [00:49<00:06, 63.80it/s]


dev:  85%|████████▍ | 2286/2703 [00:49<00:06, 62.26it/s]


dev:  85%|████████▍ | 2293/2703 [00:49<00:06, 62.05it/s]


dev:  85%|████████▌ | 2300/2703 [00:49<00:06, 61.90it/s]


dev:  85%|████████▌ | 2307/2703 [00:50<00:06, 62.73it/s]


dev:  86%|████████▌ | 2314/2703 [00:50<00:06, 63.60it/s]


dev:  86%|████████▌ | 2321/2703 [00:50<00:06, 62.31it/s]


dev:  86%|████████▌ | 2328/2703 [00:50<00:05, 63.34it/s]


dev:  86%|████████▋ | 2335/2703 [00:50<00:06, 61.04it/s]


dev:  87%|████████▋ | 2343/2703 [00:50<00:05, 64.73it/s]


dev:  87%|████████▋ | 2350/2703 [00:50<00:05, 65.68it/s]


dev:  87%|████████▋ | 2357/2703 [00:50<00:05, 66.42it/s]


dev:  87%|████████▋ | 2365/2703 [00:50<00:05, 67.47it/s]


dev:  88%|████████▊ | 2372/2703 [00:51<00:04, 67.02it/s]


dev:  88%|████████▊ | 2380/2703 [00:51<00:04, 68.60it/s]


dev:  88%|████████▊ | 2387/2703 [00:51<00:04, 67.18it/s]


dev:  89%|████████▊ | 2394/2703 [00:51<00:04, 67.05it/s]


dev:  89%|████████▉ | 2401/2703 [00:51<00:04, 67.42it/s]


dev:  89%|████████▉ | 2408/2703 [00:51<00:04, 66.96it/s]


dev:  89%|████████▉ | 2415/2703 [00:51<00:04, 64.72it/s]


dev:  90%|████████▉ | 2422/2703 [00:51<00:04, 66.17it/s]


dev:  90%|████████▉ | 2430/2703 [00:51<00:04, 67.22it/s]


dev:  90%|█████████ | 2437/2703 [00:52<00:04, 64.26it/s]


dev:  90%|█████████ | 2445/2703 [00:52<00:03, 66.90it/s]


dev:  91%|█████████ | 2452/2703 [00:52<00:03, 65.87it/s]


dev:  91%|█████████ | 2459/2703 [00:52<00:03, 63.04it/s]


dev:  91%|█████████ | 2466/2703 [00:52<00:03, 62.43it/s]


dev:  91%|█████████▏| 2473/2703 [00:52<00:03, 63.56it/s]


dev:  92%|█████████▏| 2480/2703 [00:52<00:03, 63.34it/s]


dev:  92%|█████████▏| 2487/2703 [00:52<00:03, 64.71it/s]


dev:  92%|█████████▏| 2494/2703 [00:52<00:03, 65.97it/s]


dev:  93%|█████████▎| 2501/2703 [00:53<00:03, 67.11it/s]


dev:  93%|█████████▎| 2508/2703 [00:53<00:02, 67.15it/s]


dev:  93%|█████████▎| 2515/2703 [00:53<00:02, 63.28it/s]


dev:  93%|█████████▎| 2523/2703 [00:53<00:02, 66.44it/s]


dev:  94%|█████████▎| 2530/2703 [00:53<00:02, 66.42it/s]


dev:  94%|█████████▍| 2537/2703 [00:53<00:02, 65.91it/s]


dev:  94%|█████████▍| 2544/2703 [00:53<00:02, 66.36it/s]


dev:  94%|█████████▍| 2551/2703 [00:53<00:02, 66.99it/s]


dev:  95%|█████████▍| 2559/2703 [00:53<00:02, 69.86it/s]


dev:  95%|█████████▍| 2566/2703 [00:53<00:01, 68.83it/s]


dev:  95%|█████████▌| 2574/2703 [00:54<00:01, 71.23it/s]


dev:  96%|█████████▌| 2582/2703 [00:54<00:01, 71.94it/s]


dev:  96%|█████████▌| 2590/2703 [00:54<00:01, 73.62it/s]


dev:  96%|█████████▌| 2598/2703 [00:54<00:01, 73.97it/s]


dev:  96%|█████████▋| 2606/2703 [00:54<00:01, 69.98it/s]


dev:  97%|█████████▋| 2614/2703 [00:54<00:01, 68.86it/s]


dev:  97%|█████████▋| 2621/2703 [00:54<00:01, 67.93it/s]


dev:  97%|█████████▋| 2629/2703 [00:54<00:01, 69.93it/s]


dev:  98%|█████████▊| 2637/2703 [00:54<00:00, 70.59it/s]


dev:  98%|█████████▊| 2645/2703 [00:55<00:00, 71.24it/s]


dev:  98%|█████████▊| 2653/2703 [00:55<00:00, 69.08it/s]


dev:  98%|█████████▊| 2661/2703 [00:55<00:00, 70.26it/s]


dev:  99%|█████████▊| 2669/2703 [00:55<00:00, 69.96it/s]


dev:  99%|█████████▉| 2678/2703 [00:55<00:00, 72.94it/s]


dev:  99%|█████████▉| 2686/2703 [00:55<00:00, 73.46it/s]


dev: 100%|█████████▉| 2695/2703 [00:55<00:00, 76.55it/s]


dev: 100%|██████████| 2703/2703 [00:55<00:00, 75.84it/s]
dev: 100%|██████████| 2703/2703 [00:55<00:00, 48.28it/s]
dev loss: 0.7841302156448364
dev per: 0.20060874089490113

                                                           

                                                         
[Runner] - Save the checkpoint to:

overall:  25%|██▍       | 999/4000 [05:20<13:43,  3.64it/s]

train:  28%|██▊       | 999/3515 [05:18<11:30,  3.64it/s]
                                                           

                                                         
1. result/downstream/grid/dev-best.ckpt

overall:  25%|██▍       | 999/4000 [05:20<13:43,  3.64it/s]

train:  28%|██▊       | 999/3515 [05:18<11:30,  3.64it/s]
overall:  25%|██▌       | 1000/4000 [05:21<14:48:45, 17.78s/it]

train:  28%|██▊       | 1000/3515 [05:18<12:25:04, 17.78s/it]
overall:  25%|██▌       | 1002/4000 [05:21<8:00:39,  9.62s/it] 

train:  29%|██▊       | 1002/3515 [05:18<6:42:54,  9.62s/it] 
overall:  25%|██▌       | 1003/4000 [05:21<6:03:56,  7.29s/it]

train:  29%|██▊       | 1003/3515 [05:18<5:05:02,  7.29s/it]
overall:  25%|██▌       | 1004/4000 [05:21<4:31:37,  5.44s/it]

train:  29%|██▊       | 1004/3515 [05:19<3:47:38,  5.44s/it]
overall:  25%|██▌       | 1005/4000 [05:21<3:19:37,  4.00s/it]

train:  29%|██▊       | 1005/3515 [05:19<2:47:17,  4.00s/it]
overall:  25%|██▌       | 1006/4000 [05:22<2:27:35,  2.96s/it]

train:  29%|██▊       | 1006/3515 [05:19<2:03:41,  2.96s/it]
overall:  25%|██▌       | 1007/4000 [05:22<1:49:14,  2.19s/it]

train:  29%|██▊       | 1007/3515 [05:19<1:31:32,  2.19s/it]
overall:  25%|██▌       | 1008/4000 [05:22<1:21:31,  1.63s/it]

train:  29%|██▊       | 1008/3515 [05:20<1:08:18,  1.63s/it]
overall:  25%|██▌       | 1010/4000 [05:23<48:06,  1.04it/s]  

train:  29%|██▊       | 1010/3515 [05:20<40:18,  1.04it/s]  
overall:  25%|██▌       | 1011/4000 [05:23<39:29,  1.26it/s]

train:  29%|██▉       | 1011/3515 [05:20<33:05,  1.26it/s]
overall:  25%|██▌       | 1012/4000 [05:23<32:31,  1.53it/s]

train:  29%|██▉       | 1012/3515 [05:20<27:15,  1.53it/s]
overall:  25%|██▌       | 1013/4000 [05:23<26:42,  1.86it/s]

train:  29%|██▉       | 1013/3515 [05:21<22:22,  1.86it/s]
overall:  25%|██▌       | 1014/4000 [05:24<22:51,  2.18it/s]

train:  29%|██▉       | 1014/3515 [05:21<19:09,  2.18it/s]
overall:  25%|██▌       | 1015/4000 [05:24<20:04,  2.48it/s]

train:  29%|██▉       | 1015/3515 [05:21<16:48,  2.48it/s]
overall:  25%|██▌       | 1016/4000 [05:24<18:05,  2.75it/s]

train:  29%|██▉       | 1016/3515 [05:21<15:08,  2.75it/s]
overall:  25%|██▌       | 1017/4000 [05:24<16:30,  3.01it/s]

train:  29%|██▉       | 1017/3515 [05:22<13:49,  3.01it/s]
overall:  25%|██▌       | 1018/4000 [05:25<19:42,  2.52it/s]

train:  29%|██▉       | 1018/3515 [05:22<16:29,  2.52it/s]
overall:  25%|██▌       | 1019/4000 [05:25<17:41,  2.81it/s]

train:  29%|██▉       | 1019/3515 [05:23<14:48,  2.81it/s]train loss: 0.6162915825843811
train per: 0.17628275710195992

overall:  26%|██▌       | 1020/4000 [05:25<16:03,  3.09it/s]

train:  29%|██▉       | 1020/3515 [05:23<13:27,  3.09it/s]
overall:  26%|██▌       | 1021/4000 [05:26<15:17,  3.25it/s]

train:  29%|██▉       | 1021/3515 [05:23<12:48,  3.25it/s]
overall:  26%|██▌       | 1022/4000 [05:26<14:30,  3.42it/s]

train:  29%|██▉       | 1022/3515 [05:23<12:08,  3.42it/s]
overall:  26%|██▌       | 1023/4000 [05:26<15:03,  3.29it/s]

train:  29%|██▉       | 1023/3515 [05:24<12:36,  3.29it/s]
overall:  26%|██▌       | 1024/4000 [05:26<12:06,  4.10it/s]

train:  29%|██▉       | 1024/3515 [05:24<10:08,  4.10it/s]
overall:  26%|██▌       | 1025/4000 [05:27<12:20,  4.02it/s]

train:  29%|██▉       | 1025/3515 [05:24<10:19,  4.02it/s]
overall:  26%|██▌       | 1026/4000 [05:27<12:32,  3.95it/s]

train:  29%|██▉       | 1026/3515 [05:24<10:29,  3.95it/s]
overall:  26%|██▌       | 1027/4000 [05:27<12:52,  3.85it/s]

train:  29%|██▉       | 1027/3515 [05:25<10:46,  3.85it/s]
overall:  26%|██▌       | 1028/4000 [05:27<12:52,  3.85it/s]

train:  29%|██▉       | 1028/3515 [05:25<10:46,  3.85it/s]
overall:  26%|██▌       | 1029/4000 [05:28<12:57,  3.82it/s]

train:  29%|██▉       | 1029/3515 [05:25<10:50,  3.82it/s]
overall:  26%|██▌       | 1030/4000 [05:28<11:28,  4.32it/s]

train:  29%|██▉       | 1030/3515 [05:25<09:35,  4.32it/s]
overall:  26%|██▌       | 1031/4000 [05:28<12:07,  4.08it/s]

train:  29%|██▉       | 1031/3515 [05:26<10:08,  4.08it/s]
overall:  26%|██▌       | 1032/4000 [05:28<12:33,  3.94it/s]

train:  29%|██▉       | 1032/3515 [05:26<10:30,  3.94it/s]
overall:  26%|██▌       | 1033/4000 [05:29<12:51,  3.84it/s]

train:  29%|██▉       | 1033/3515 [05:26<10:45,  3.84it/s]
overall:  26%|██▌       | 1034/4000 [05:29<13:00,  3.80it/s]

train:  29%|██▉       | 1034/3515 [05:26<10:53,  3.80it/s]
overall:  26%|██▌       | 1035/4000 [05:29<13:20,  3.70it/s]

train:  29%|██▉       | 1035/3515 [05:27<11:09,  3.70it/s]
overall:  26%|██▌       | 1036/4000 [05:30<13:19,  3.71it/s]

train:  29%|██▉       | 1036/3515 [05:27<11:08,  3.71it/s]
overall:  26%|██▌       | 1037/4000 [05:30<12:55,  3.82it/s]

train:  30%|██▉       | 1037/3515 [05:27<10:48,  3.82it/s]
overall:  26%|██▌       | 1038/4000 [05:30<12:18,  4.01it/s]

train:  30%|██▉       | 1038/3515 [05:27<10:17,  4.01it/s]
overall:  26%|██▌       | 1039/4000 [05:30<12:36,  3.91it/s]

train:  30%|██▉       | 1039/3515 [05:28<10:32,  3.91it/s]train loss: 0.6384461522102356
train per: 0.18153805227338224

overall:  26%|██▌       | 1040/4000 [05:31<12:56,  3.81it/s]

train:  30%|██▉       | 1040/3515 [05:28<10:49,  3.81it/s]
overall:  26%|██▌       | 1041/4000 [05:31<12:41,  3.88it/s]

train:  30%|██▉       | 1041/3515 [05:28<10:36,  3.88it/s]
overall:  26%|██▌       | 1042/4000 [05:31<16:00,  3.08it/s]

train:  30%|██▉       | 1042/3515 [05:29<13:22,  3.08it/s]
overall:  26%|██▌       | 1043/4000 [05:31<14:39,  3.36it/s]

train:  30%|██▉       | 1043/3515 [05:29<12:15,  3.36it/s]
overall:  26%|██▌       | 1044/4000 [05:32<14:11,  3.47it/s]

train:  30%|██▉       | 1044/3515 [05:29<11:51,  3.47it/s]
overall:  26%|██▌       | 1045/4000 [05:32<12:21,  3.98it/s]

train:  30%|██▉       | 1045/3515 [05:29<10:20,  3.98it/s]
overall:  26%|██▌       | 1046/4000 [05:32<12:31,  3.93it/s]

train:  30%|██▉       | 1046/3515 [05:30<10:28,  3.93it/s]
overall:  26%|██▌       | 1047/4000 [05:32<13:19,  3.69it/s]

train:  30%|██▉       | 1047/3515 [05:30<11:08,  3.69it/s]
overall:  26%|██▌       | 1048/4000 [05:33<12:59,  3.79it/s]

train:  30%|██▉       | 1048/3515 [05:30<10:51,  3.79it/s]
overall:  26%|██▌       | 1049/4000 [05:33<12:45,  3.85it/s]

train:  30%|██▉       | 1049/3515 [05:30<10:39,  3.85it/s]
overall:  26%|██▋       | 1050/4000 [05:33<12:39,  3.88it/s]

train:  30%|██▉       | 1050/3515 [05:31<10:34,  3.88it/s]
overall:  26%|██▋       | 1052/4000 [05:34<10:34,  4.64it/s]

train:  30%|██▉       | 1052/3515 [05:31<08:50,  4.64it/s]
overall:  26%|██▋       | 1053/4000 [05:34<09:35,  5.12it/s]

train:  30%|██▉       | 1053/3515 [05:31<08:00,  5.12it/s]
overall:  26%|██▋       | 1055/4000 [05:34<08:53,  5.52it/s]

train:  30%|███       | 1055/3515 [05:31<07:25,  5.52it/s]
overall:  26%|██▋       | 1057/4000 [05:34<08:50,  5.55it/s]

train:  30%|███       | 1057/3515 [05:32<07:23,  5.55it/s]
overall:  26%|██▋       | 1058/4000 [05:35<08:07,  6.03it/s]

train:  30%|███       | 1058/3515 [05:32<06:47,  6.03it/s]
overall:  26%|██▋       | 1059/4000 [05:35<09:15,  5.29it/s]

train:  30%|███       | 1059/3515 [05:32<07:44,  5.29it/s]train loss: 0.6042833924293518
train per: 0.174116632266412

overall:  26%|██▋       | 1060/4000 [05:35<10:09,  4.82it/s]

train:  30%|███       | 1060/3515 [05:32<08:29,  4.82it/s]
overall:  27%|██▋       | 1061/4000 [05:35<10:59,  4.46it/s]

train:  30%|███       | 1061/3515 [05:33<09:10,  4.46it/s]
overall:  27%|██▋       | 1062/4000 [05:36<11:30,  4.26it/s]

train:  30%|███       | 1062/3515 [05:33<09:36,  4.26it/s]
overall:  27%|██▋       | 1063/4000 [05:36<11:55,  4.11it/s]

train:  30%|███       | 1063/3515 [05:33<09:57,  4.11it/s]
overall:  27%|██▋       | 1064/4000 [05:36<12:14,  4.00it/s]

train:  30%|███       | 1064/3515 [05:33<10:13,  4.00it/s]
overall:  27%|██▋       | 1065/4000 [05:36<12:06,  4.04it/s]

train:  30%|███       | 1065/3515 [05:34<10:06,  4.04it/s]
overall:  27%|██▋       | 1066/4000 [05:37<15:17,  3.20it/s]

train:  30%|███       | 1066/3515 [05:34<12:45,  3.20it/s]
overall:  27%|██▋       | 1067/4000 [05:37<12:42,  3.85it/s]

train:  30%|███       | 1067/3515 [05:34<10:36,  3.85it/s]
overall:  27%|██▋       | 1069/4000 [05:37<11:01,  4.43it/s]

train:  30%|███       | 1069/3515 [05:35<09:12,  4.43it/s]
overall:  27%|██▋       | 1070/4000 [05:38<11:34,  4.22it/s]

train:  30%|███       | 1070/3515 [05:35<09:39,  4.22it/s]
overall:  27%|██▋       | 1071/4000 [05:38<12:35,  3.88it/s]

train:  30%|███       | 1071/3515 [05:35<10:30,  3.88it/s]
overall:  27%|██▋       | 1073/4000 [05:38<10:45,  4.54it/s]

train:  31%|███       | 1073/3515 [05:36<08:58,  4.54it/s]
overall:  27%|██▋       | 1074/4000 [05:38<09:45,  5.00it/s]

train:  31%|███       | 1074/3515 [05:36<08:08,  5.00it/s]
overall:  27%|██▋       | 1075/4000 [05:39<10:48,  4.51it/s]

train:  31%|███       | 1075/3515 [05:36<09:00,  4.51it/s]
overall:  27%|██▋       | 1076/4000 [05:39<11:19,  4.30it/s]

train:  31%|███       | 1076/3515 [05:36<09:26,  4.31it/s]
overall:  27%|██▋       | 1077/4000 [05:39<10:50,  4.49it/s]

train:  31%|███       | 1077/3515 [05:36<09:02,  4.49it/s]
overall:  27%|██▋       | 1078/4000 [05:39<11:33,  4.21it/s]

train:  31%|███       | 1078/3515 [05:37<09:38,  4.21it/s]
overall:  27%|██▋       | 1079/4000 [05:40<12:05,  4.03it/s]

train:  31%|███       | 1079/3515 [05:37<10:05,  4.03it/s]train loss: 0.6321051716804504
train per: 0.1842459484197491

overall:  27%|██▋       | 1081/4000 [05:40<10:19,  4.71it/s]

train:  31%|███       | 1081/3515 [05:37<08:36,  4.71it/s]
overall:  27%|██▋       | 1082/4000 [05:40<09:28,  5.13it/s]

train:  31%|███       | 1082/3515 [05:38<07:54,  5.13it/s]
overall:  27%|██▋       | 1083/4000 [05:40<10:23,  4.68it/s]

train:  31%|███       | 1083/3515 [05:38<08:40,  4.68it/s]
overall:  27%|██▋       | 1084/4000 [05:41<10:34,  4.59it/s]

train:  31%|███       | 1084/3515 [05:38<08:49,  4.59it/s]
overall:  27%|██▋       | 1085/4000 [05:41<11:17,  4.31it/s]

train:  31%|███       | 1085/3515 [05:38<09:24,  4.31it/s]
overall:  27%|██▋       | 1086/4000 [05:41<11:42,  4.15it/s]

train:  31%|███       | 1086/3515 [05:39<09:45,  4.15it/s]
overall:  27%|██▋       | 1087/4000 [05:41<10:06,  4.80it/s]

train:  31%|███       | 1087/3515 [05:39<08:25,  4.80it/s]
overall:  27%|██▋       | 1088/4000 [05:41<09:39,  5.03it/s]

train:  31%|███       | 1088/3515 [05:39<08:02,  5.03it/s]
overall:  27%|██▋       | 1089/4000 [05:42<10:36,  4.57it/s]

train:  31%|███       | 1089/3515 [05:39<08:50,  4.57it/s]
overall:  27%|██▋       | 1090/4000 [05:42<13:28,  3.60it/s]

train:  31%|███       | 1090/3515 [05:40<11:13,  3.60it/s]
overall:  27%|██▋       | 1091/4000 [05:42<13:13,  3.67it/s]

train:  31%|███       | 1091/3515 [05:40<11:00,  3.67it/s]
overall:  27%|██▋       | 1092/4000 [05:43<13:00,  3.72it/s]

train:  31%|███       | 1092/3515 [05:40<10:50,  3.72it/s]
overall:  27%|██▋       | 1093/4000 [05:43<13:01,  3.72it/s]

train:  31%|███       | 1093/3515 [05:40<10:51,  3.72it/s]
overall:  27%|██▋       | 1094/4000 [05:43<11:37,  4.17it/s]

train:  31%|███       | 1094/3515 [05:41<09:41,  4.17it/s]
overall:  27%|██▋       | 1095/4000 [05:43<13:08,  3.68it/s]

train:  31%|███       | 1095/3515 [05:41<10:57,  3.68it/s]
overall:  27%|██▋       | 1096/4000 [05:44<13:01,  3.71it/s]

train:  31%|███       | 1096/3515 [05:41<10:51,  3.71it/s]
overall:  27%|██▋       | 1097/4000 [05:44<13:04,  3.70it/s]

train:  31%|███       | 1097/3515 [05:41<10:53,  3.70it/s]
overall:  27%|██▋       | 1099/4000 [05:44<10:45,  4.49it/s]

train:  31%|███▏      | 1099/3515 [05:42<08:57,  4.49it/s]train loss: 0.5961498022079468
train per: 0.17392150863627515

overall:  28%|██▊       | 1100/4000 [05:45<11:19,  4.27it/s]

train:  31%|███▏      | 1100/3515 [05:42<09:26,  4.26it/s]
overall:  28%|██▊       | 1101/4000 [05:45<11:32,  4.19it/s]

train:  31%|███▏      | 1101/3515 [05:42<09:36,  4.19it/s]
overall:  28%|██▊       | 1102/4000 [05:45<11:36,  4.16it/s]

train:  31%|███▏      | 1102/3515 [05:42<09:40,  4.16it/s]
overall:  28%|██▊       | 1103/4000 [05:45<11:52,  4.07it/s]

train:  31%|███▏      | 1103/3515 [05:43<09:52,  4.07it/s]
overall:  28%|██▊       | 1104/4000 [05:46<12:08,  3.97it/s]

train:  31%|███▏      | 1104/3515 [05:43<10:06,  3.97it/s]
overall:  28%|██▊       | 1105/4000 [05:46<10:43,  4.50it/s]

train:  31%|███▏      | 1105/3515 [05:43<08:55,  4.50it/s]
overall:  28%|██▊       | 1106/4000 [05:46<11:13,  4.30it/s]

train:  31%|███▏      | 1106/3515 [05:43<09:20,  4.30it/s]
overall:  28%|██▊       | 1107/4000 [05:46<11:16,  4.28it/s]

train:  31%|███▏      | 1107/3515 [05:44<09:23,  4.28it/s]
overall:  28%|██▊       | 1108/4000 [05:47<11:44,  4.10it/s]

train:  32%|███▏      | 1108/3515 [05:44<09:46,  4.10it/s]
overall:  28%|██▊       | 1109/4000 [05:47<12:04,  3.99it/s]

train:  32%|███▏      | 1109/3515 [05:44<10:03,  3.99it/s]
overall:  28%|██▊       | 1110/4000 [05:47<12:19,  3.91it/s]

train:  32%|███▏      | 1110/3515 [05:44<10:15,  3.91it/s]
overall:  28%|██▊       | 1111/4000 [05:47<12:26,  3.87it/s]

train:  32%|███▏      | 1111/3515 [05:45<10:21,  3.87it/s]
overall:  28%|██▊       | 1112/4000 [05:48<12:07,  3.97it/s]

train:  32%|███▏      | 1112/3515 [05:45<10:05,  3.97it/s]
overall:  28%|██▊       | 1113/4000 [05:48<12:24,  3.88it/s]

train:  32%|███▏      | 1113/3515 [05:45<10:19,  3.88it/s]
overall:  28%|██▊       | 1114/4000 [05:48<15:58,  3.01it/s]

train:  32%|███▏      | 1114/3515 [05:46<13:17,  3.01it/s]
overall:  28%|██▊       | 1115/4000 [05:49<15:06,  3.18it/s]

train:  32%|███▏      | 1115/3515 [05:46<12:33,  3.18it/s]
overall:  28%|██▊       | 1116/4000 [05:49<14:18,  3.36it/s]

train:  32%|███▏      | 1116/3515 [05:46<11:54,  3.36it/s]
overall:  28%|██▊       | 1117/4000 [05:49<14:06,  3.40it/s]

train:  32%|███▏      | 1117/3515 [05:47<11:44,  3.40it/s]
overall:  28%|██▊       | 1118/4000 [05:49<13:50,  3.47it/s]

train:  32%|███▏      | 1118/3515 [05:47<11:31,  3.47it/s]
overall:  28%|██▊       | 1119/4000 [05:50<11:32,  4.16it/s]

train:  32%|███▏      | 1119/3515 [05:47<09:36,  4.16it/s]train loss: 0.6029472351074219
train per: 0.1724962092981419

overall:  28%|██▊       | 1120/4000 [05:50<10:00,  4.80it/s]

train:  32%|███▏      | 1120/3515 [05:47<08:19,  4.79it/s]
overall:  28%|██▊       | 1121/4000 [05:50<10:51,  4.42it/s]

train:  32%|███▏      | 1121/3515 [05:47<09:01,  4.42it/s]
overall:  28%|██▊       | 1122/4000 [05:50<11:21,  4.22it/s]

train:  32%|███▏      | 1122/3515 [05:48<09:26,  4.22it/s]
overall:  28%|██▊       | 1123/4000 [05:51<11:52,  4.04it/s]

train:  32%|███▏      | 1123/3515 [05:48<09:52,  4.04it/s]
overall:  28%|██▊       | 1124/4000 [05:51<12:14,  3.91it/s]

train:  32%|███▏      | 1124/3515 [05:48<10:10,  3.91it/s]
overall:  28%|██▊       | 1125/4000 [05:51<12:23,  3.87it/s]

train:  32%|███▏      | 1125/3515 [05:48<10:18,  3.87it/s]
overall:  28%|██▊       | 1126/4000 [05:51<12:35,  3.80it/s]

train:  32%|███▏      | 1126/3515 [05:49<10:28,  3.80it/s]
overall:  28%|██▊       | 1127/4000 [05:52<11:58,  4.00it/s]

train:  32%|███▏      | 1127/3515 [05:49<09:57,  4.00it/s]
overall:  28%|██▊       | 1128/4000 [05:52<12:15,  3.91it/s]

train:  32%|███▏      | 1128/3515 [05:49<10:11,  3.91it/s]
overall:  28%|██▊       | 1129/4000 [05:52<12:05,  3.95it/s]

train:  32%|███▏      | 1129/3515 [05:49<10:03,  3.95it/s]
overall:  28%|██▊       | 1130/4000 [05:52<12:15,  3.90it/s]

train:  32%|███▏      | 1130/3515 [05:50<10:11,  3.90it/s]
overall:  28%|██▊       | 1131/4000 [05:52<10:39,  4.48it/s]

train:  32%|███▏      | 1131/3515 [05:50<08:51,  4.48it/s]
overall:  28%|██▊       | 1132/4000 [05:53<10:27,  4.57it/s]

train:  32%|███▏      | 1132/3515 [05:50<08:41,  4.57it/s]
overall:  28%|██▊       | 1133/4000 [05:53<10:48,  4.42it/s]

train:  32%|███▏      | 1133/3515 [05:50<08:58,  4.42it/s]
overall:  28%|██▊       | 1134/4000 [05:53<11:37,  4.11it/s]

train:  32%|███▏      | 1134/3515 [05:51<09:39,  4.11it/s]
overall:  28%|██▊       | 1135/4000 [05:53<11:03,  4.32it/s]

train:  32%|███▏      | 1135/3515 [05:51<09:10,  4.32it/s]
overall:  28%|██▊       | 1136/4000 [05:54<11:14,  4.25it/s]

train:  32%|███▏      | 1136/3515 [05:51<09:20,  4.25it/s]
overall:  28%|██▊       | 1137/4000 [05:54<11:13,  4.25it/s]

train:  32%|███▏      | 1137/3515 [05:51<09:19,  4.25it/s]
overall:  28%|██▊       | 1138/4000 [05:54<15:13,  3.13it/s]

train:  32%|███▏      | 1138/3515 [05:52<12:38,  3.13it/s]
overall:  28%|██▊       | 1139/4000 [05:55<14:03,  3.39it/s]

train:  32%|███▏      | 1139/3515 [05:52<11:40,  3.39it/s]train loss: 0.6024952530860901
train per: 0.17200641262794425

overall:  28%|██▊       | 1140/4000 [05:55<13:56,  3.42it/s]

train:  32%|███▏      | 1140/3515 [05:52<11:35,  3.41it/s]
overall:  29%|██▊       | 1141/4000 [05:55<13:59,  3.41it/s]

train:  32%|███▏      | 1141/3515 [05:53<11:36,  3.41it/s]
overall:  29%|██▊       | 1142/4000 [05:56<13:43,  3.47it/s]

train:  32%|███▏      | 1142/3515 [05:53<11:23,  3.47it/s]
overall:  29%|██▊       | 1143/4000 [05:56<11:38,  4.09it/s]

train:  33%|███▎      | 1143/3515 [05:53<09:40,  4.09it/s]
overall:  29%|██▊       | 1144/4000 [05:56<11:10,  4.26it/s]

train:  33%|███▎      | 1144/3515 [05:53<09:16,  4.26it/s]
overall:  29%|██▊       | 1146/4000 [05:56<08:53,  5.35it/s]

train:  33%|███▎      | 1146/3515 [05:53<07:22,  5.35it/s]
overall:  29%|██▊       | 1147/4000 [05:56<09:46,  4.87it/s]

train:  33%|███▎      | 1147/3515 [05:54<08:06,  4.87it/s]
overall:  29%|██▊       | 1148/4000 [05:57<10:08,  4.69it/s]

train:  33%|███▎      | 1148/3515 [05:54<08:25,  4.69it/s]
overall:  29%|██▊       | 1149/4000 [05:57<10:43,  4.43it/s]

train:  33%|███▎      | 1149/3515 [05:54<08:54,  4.43it/s]
overall:  29%|██▉       | 1150/4000 [05:57<11:13,  4.23it/s]

train:  33%|███▎      | 1150/3515 [05:55<09:18,  4.23it/s]
overall:  29%|██▉       | 1151/4000 [05:57<11:36,  4.09it/s]

train:  33%|███▎      | 1151/3515 [05:55<09:37,  4.09it/s]
overall:  29%|██▉       | 1152/4000 [05:58<11:58,  3.96it/s]

train:  33%|███▎      | 1152/3515 [05:55<09:56,  3.96it/s]
overall:  29%|██▉       | 1153/4000 [05:58<12:13,  3.88it/s]

train:  33%|███▎      | 1153/3515 [05:55<10:08,  3.88it/s]
overall:  29%|██▉       | 1154/4000 [05:58<11:48,  4.02it/s]

train:  33%|███▎      | 1154/3515 [05:56<09:47,  4.02it/s]
overall:  29%|██▉       | 1155/4000 [05:58<12:03,  3.93it/s]

train:  33%|███▎      | 1155/3515 [05:56<10:00,  3.93it/s]
overall:  29%|██▉       | 1156/4000 [05:59<11:53,  3.99it/s]

train:  33%|███▎      | 1156/3515 [05:56<09:51,  3.99it/s]
overall:  29%|██▉       | 1157/4000 [05:59<11:57,  3.96it/s]

train:  33%|███▎      | 1157/3515 [05:56<09:55,  3.96it/s]
overall:  29%|██▉       | 1158/4000 [05:59<12:15,  3.87it/s]

train:  33%|███▎      | 1158/3515 [05:57<10:09,  3.87it/s]
overall:  29%|██▉       | 1159/4000 [05:59<12:30,  3.79it/s]

train:  33%|███▎      | 1159/3515 [05:57<10:22,  3.79it/s]train loss: 0.5994102954864502
train per: 0.17598887905977506

overall:  29%|██▉       | 1160/4000 [06:00<12:39,  3.74it/s]

train:  33%|███▎      | 1160/3515 [05:57<10:30,  3.73it/s]
overall:  29%|██▉       | 1161/4000 [06:00<10:18,  4.59it/s]

train:  33%|███▎      | 1161/3515 [05:57<08:32,  4.59it/s]
overall:  29%|██▉       | 1162/4000 [06:00<12:04,  3.92it/s]

train:  33%|███▎      | 1162/3515 [05:58<10:00,  3.92it/s]
overall:  29%|██▉       | 1163/4000 [06:00<11:18,  4.18it/s]

train:  33%|███▎      | 1163/3515 [05:58<09:22,  4.18it/s]
overall:  29%|██▉       | 1164/4000 [06:01<11:21,  4.16it/s]

train:  33%|███▎      | 1164/3515 [05:58<09:25,  4.16it/s]
overall:  29%|██▉       | 1165/4000 [06:01<11:40,  4.05it/s]

train:  33%|███▎      | 1165/3515 [05:58<09:40,  4.05it/s]
overall:  29%|██▉       | 1166/4000 [06:01<12:10,  3.88it/s]

train:  33%|███▎      | 1166/3515 [05:59<10:05,  3.88it/s]
overall:  29%|██▉       | 1167/4000 [06:02<13:12,  3.58it/s]

train:  33%|███▎      | 1167/3515 [05:59<10:56,  3.58it/s]
overall:  29%|██▉       | 1168/4000 [06:02<13:05,  3.60it/s]

train:  33%|███▎      | 1168/3515 [05:59<10:51,  3.60it/s]
overall:  29%|██▉       | 1169/4000 [06:02<12:54,  3.65it/s]

train:  33%|███▎      | 1169/3515 [05:59<10:41,  3.65it/s]
overall:  29%|██▉       | 1170/4000 [06:02<12:51,  3.67it/s]

train:  33%|███▎      | 1170/3515 [06:00<10:39,  3.67it/s]
overall:  29%|██▉       | 1171/4000 [06:03<12:26,  3.79it/s]

train:  33%|███▎      | 1171/3515 [06:00<10:18,  3.79it/s]
overall:  29%|██▉       | 1172/4000 [06:03<12:24,  3.80it/s]

train:  33%|███▎      | 1172/3515 [06:00<10:17,  3.80it/s]
overall:  29%|██▉       | 1173/4000 [06:03<11:50,  3.98it/s]

train:  33%|███▎      | 1173/3515 [06:00<09:48,  3.98it/s]
overall:  29%|██▉       | 1174/4000 [06:03<12:06,  3.89it/s]

train:  33%|███▎      | 1174/3515 [06:01<10:01,  3.89it/s]
overall:  29%|██▉       | 1175/4000 [06:04<11:54,  3.95it/s]

train:  33%|███▎      | 1175/3515 [06:01<09:52,  3.95it/s]
overall:  29%|██▉       | 1176/4000 [06:04<12:12,  3.85it/s]

train:  33%|███▎      | 1176/3515 [06:01<10:06,  3.85it/s]
overall:  29%|██▉       | 1177/4000 [06:04<12:18,  3.82it/s]

train:  33%|███▎      | 1177/3515 [06:01<10:11,  3.82it/s]
overall:  29%|██▉       | 1178/4000 [06:04<12:28,  3.77it/s]

train:  34%|███▎      | 1178/3515 [06:02<10:19,  3.77it/s]
overall:  29%|██▉       | 1179/4000 [06:05<12:34,  3.74it/s]

train:  34%|███▎      | 1179/3515 [06:02<10:25,  3.74it/s]train loss: 0.6301527619361877
train per: 0.17987099914810759

overall:  30%|██▉       | 1180/4000 [06:05<12:35,  3.73it/s]

train:  34%|███▎      | 1180/3515 [06:02<10:25,  3.73it/s]
overall:  30%|██▉       | 1181/4000 [06:05<12:32,  3.75it/s]

train:  34%|███▎      | 1181/3515 [06:03<10:23,  3.75it/s]
overall:  30%|██▉       | 1182/4000 [06:05<12:34,  3.74it/s]

train:  34%|███▎      | 1182/3515 [06:03<10:24,  3.74it/s]
overall:  30%|██▉       | 1183/4000 [06:06<10:14,  4.58it/s]

train:  34%|███▎      | 1183/3515 [06:03<08:28,  4.58it/s]
overall:  30%|██▉       | 1184/4000 [06:06<10:46,  4.35it/s]

train:  34%|███▎      | 1184/3515 [06:03<08:55,  4.35it/s]
overall:  30%|██▉       | 1185/4000 [06:06<11:15,  4.17it/s]

train:  34%|███▎      | 1185/3515 [06:03<09:19,  4.17it/s]
overall:  30%|██▉       | 1186/4000 [06:07<14:59,  3.13it/s]

train:  34%|███▎      | 1186/3515 [06:04<12:24,  3.13it/s]
overall:  30%|██▉       | 1187/4000 [06:07<14:15,  3.29it/s]

train:  34%|███▍      | 1187/3515 [06:04<11:47,  3.29it/s]
overall:  30%|██▉       | 1188/4000 [06:07<11:32,  4.06it/s]

train:  34%|███▍      | 1188/3515 [06:04<09:32,  4.06it/s]
overall:  30%|██▉       | 1189/4000 [06:07<12:03,  3.89it/s]

train:  34%|███▍      | 1189/3515 [06:05<09:58,  3.89it/s]
overall:  30%|██▉       | 1190/4000 [06:08<12:22,  3.79it/s]

train:  34%|███▍      | 1190/3515 [06:05<10:14,  3.79it/s]
overall:  30%|██▉       | 1191/4000 [06:08<13:25,  3.49it/s]

train:  34%|███▍      | 1191/3515 [06:05<11:06,  3.49it/s]
overall:  30%|██▉       | 1192/4000 [06:08<12:37,  3.71it/s]

train:  34%|███▍      | 1192/3515 [06:05<10:26,  3.71it/s]
overall:  30%|██▉       | 1193/4000 [06:08<12:40,  3.69it/s]

train:  34%|███▍      | 1193/3515 [06:06<10:28,  3.69it/s]
overall:  30%|██▉       | 1194/4000 [06:09<12:50,  3.64it/s]

train:  34%|███▍      | 1194/3515 [06:06<10:37,  3.64it/s]
overall:  30%|██▉       | 1195/4000 [06:09<12:55,  3.62it/s]

train:  34%|███▍      | 1195/3515 [06:06<10:41,  3.62it/s]
overall:  30%|██▉       | 1196/4000 [06:09<12:48,  3.65it/s]

train:  34%|███▍      | 1196/3515 [06:07<10:35,  3.65it/s]
overall:  30%|██▉       | 1197/4000 [06:09<12:41,  3.68it/s]

train:  34%|███▍      | 1197/3515 [06:07<10:29,  3.68it/s]
overall:  30%|██▉       | 1198/4000 [06:10<12:28,  3.74it/s]

train:  34%|███▍      | 1198/3515 [06:07<10:19,  3.74it/s]
overall:  30%|██▉       | 1199/4000 [06:10<12:25,  3.76it/s]

train:  34%|███▍      | 1199/3515 [06:07<10:16,  3.76it/s]train loss: 0.6017759442329407
train per: 0.17196035296734805

overall:  30%|███       | 1201/4000 [06:10<10:29,  4.45it/s]

train:  34%|███▍      | 1201/3515 [06:08<08:40,  4.45it/s]
overall:  30%|███       | 1202/4000 [06:11<11:04,  4.21it/s]

train:  34%|███▍      | 1202/3515 [06:08<09:09,  4.21it/s]
overall:  30%|███       | 1203/4000 [06:11<11:09,  4.18it/s]

train:  34%|███▍      | 1203/3515 [06:08<09:13,  4.18it/s]
overall:  30%|███       | 1204/4000 [06:11<11:26,  4.07it/s]

train:  34%|███▍      | 1204/3515 [06:09<09:27,  4.07it/s]
overall:  30%|███       | 1205/4000 [06:11<09:33,  4.88it/s]

train:  34%|███▍      | 1205/3515 [06:09<07:53,  4.88it/s]
overall:  30%|███       | 1206/4000 [06:12<10:29,  4.44it/s]

train:  34%|███▍      | 1206/3515 [06:09<08:40,  4.44it/s]
overall:  30%|███       | 1207/4000 [06:12<11:13,  4.15it/s]

train:  34%|███▍      | 1207/3515 [06:09<09:16,  4.15it/s]
overall:  30%|███       | 1208/4000 [06:12<11:29,  4.05it/s]

train:  34%|███▍      | 1208/3515 [06:09<09:30,  4.05it/s]
overall:  30%|███       | 1209/4000 [06:12<11:42,  3.98it/s]

train:  34%|███▍      | 1209/3515 [06:10<09:40,  3.98it/s]
overall:  30%|███       | 1210/4000 [06:13<15:22,  3.03it/s]

train:  34%|███▍      | 1210/3515 [06:10<12:41,  3.03it/s]
overall:  30%|███       | 1211/4000 [06:13<14:38,  3.18it/s]

train:  34%|███▍      | 1211/3515 [06:10<12:05,  3.18it/s]
overall:  30%|███       | 1212/4000 [06:13<13:52,  3.35it/s]

train:  34%|███▍      | 1212/3515 [06:11<11:27,  3.35it/s]
overall:  30%|███       | 1213/4000 [06:14<13:09,  3.53it/s]

train:  35%|███▍      | 1213/3515 [06:11<10:52,  3.53it/s]
overall:  30%|███       | 1214/4000 [06:14<12:56,  3.59it/s]

train:  35%|███▍      | 1214/3515 [06:11<10:41,  3.59it/s]
overall:  30%|███       | 1215/4000 [06:14<13:39,  3.40it/s]

train:  35%|███▍      | 1215/3515 [06:12<11:16,  3.40it/s]
overall:  30%|███       | 1216/4000 [06:14<12:51,  3.61it/s]

train:  35%|███▍      | 1216/3515 [06:12<10:37,  3.61it/s]
overall:  30%|███       | 1217/4000 [06:15<10:24,  4.46it/s]

train:  35%|███▍      | 1217/3515 [06:12<08:35,  4.46it/s]
overall:  30%|███       | 1218/4000 [06:15<11:10,  4.15it/s]

train:  35%|███▍      | 1218/3515 [06:12<09:13,  4.15it/s]
overall:  30%|███       | 1219/4000 [06:15<11:42,  3.96it/s]

train:  35%|███▍      | 1219/3515 [06:12<09:39,  3.96it/s]train loss: 0.5727323293685913
train per: 0.16778096897217384

overall:  30%|███       | 1220/4000 [06:15<13:11,  3.51it/s]

train:  35%|███▍      | 1220/3515 [06:13<10:53,  3.51it/s]
overall:  31%|███       | 1221/4000 [06:16<12:52,  3.60it/s]

train:  35%|███▍      | 1221/3515 [06:13<10:37,  3.60it/s]
overall:  31%|███       | 1222/4000 [06:16<12:48,  3.61it/s]

train:  35%|███▍      | 1222/3515 [06:13<10:34,  3.61it/s]
overall:  31%|███       | 1223/4000 [06:16<12:39,  3.66it/s]

train:  35%|███▍      | 1223/3515 [06:14<10:26,  3.66it/s]
overall:  31%|███       | 1224/4000 [06:17<12:36,  3.67it/s]

train:  35%|███▍      | 1224/3515 [06:14<10:24,  3.67it/s]
overall:  31%|███       | 1225/4000 [06:17<12:28,  3.71it/s]

train:  35%|███▍      | 1225/3515 [06:14<10:17,  3.71it/s]
overall:  31%|███       | 1226/4000 [06:17<12:28,  3.71it/s]

train:  35%|███▍      | 1226/3515 [06:14<10:17,  3.71it/s]
overall:  31%|███       | 1227/4000 [06:17<12:30,  3.69it/s]

train:  35%|███▍      | 1227/3515 [06:15<10:19,  3.70it/s]
overall:  31%|███       | 1228/4000 [06:18<12:26,  3.71it/s]

train:  35%|███▍      | 1228/3515 [06:15<10:15,  3.71it/s]
overall:  31%|███       | 1229/4000 [06:18<10:13,  4.52it/s]

train:  35%|███▍      | 1229/3515 [06:15<08:25,  4.52it/s]
overall:  31%|███       | 1231/4000 [06:18<09:28,  4.87it/s]

train:  35%|███▌      | 1231/3515 [06:15<07:48,  4.87it/s]
overall:  31%|███       | 1232/4000 [06:18<10:18,  4.47it/s]

train:  35%|███▌      | 1232/3515 [06:16<08:30,  4.47it/s]
overall:  31%|███       | 1233/4000 [06:19<10:56,  4.21it/s]

train:  35%|███▌      | 1233/3515 [06:16<09:01,  4.21it/s]
overall:  31%|███       | 1234/4000 [06:19<13:06,  3.52it/s]

train:  35%|███▌      | 1234/3515 [06:16<10:48,  3.52it/s]
overall:  31%|███       | 1235/4000 [06:19<12:35,  3.66it/s]

train:  35%|███▌      | 1235/3515 [06:17<10:23,  3.66it/s]
overall:  31%|███       | 1236/4000 [06:20<11:52,  3.88it/s]

train:  35%|███▌      | 1236/3515 [06:17<09:47,  3.88it/s]
overall:  31%|███       | 1237/4000 [06:20<12:06,  3.80it/s]

train:  35%|███▌      | 1237/3515 [06:17<09:58,  3.80it/s]
overall:  31%|███       | 1238/4000 [06:20<12:13,  3.76it/s]

train:  35%|███▌      | 1238/3515 [06:17<10:04,  3.76it/s]
overall:  31%|███       | 1239/4000 [06:20<13:25,  3.43it/s]

train:  35%|███▌      | 1239/3515 [06:18<11:04,  3.43it/s]train loss: 0.5896121263504028
train per: 0.1738947072467843

overall:  31%|███       | 1240/4000 [06:21<13:12,  3.48it/s]

train:  35%|███▌      | 1240/3515 [06:18<10:53,  3.48it/s]
overall:  31%|███       | 1241/4000 [06:21<12:56,  3.55it/s]

train:  35%|███▌      | 1241/3515 [06:18<10:39,  3.55it/s]
overall:  31%|███       | 1242/4000 [06:21<11:58,  3.84it/s]

train:  35%|███▌      | 1242/3515 [06:19<09:52,  3.84it/s]
overall:  31%|███       | 1243/4000 [06:21<11:46,  3.90it/s]

train:  35%|███▌      | 1243/3515 [06:19<09:42,  3.90it/s]
overall:  31%|███       | 1244/4000 [06:22<12:27,  3.69it/s]

train:  35%|███▌      | 1244/3515 [06:19<10:15,  3.69it/s]
overall:  31%|███       | 1245/4000 [06:22<11:33,  3.98it/s]

train:  35%|███▌      | 1245/3515 [06:19<09:31,  3.98it/s]
overall:  31%|███       | 1246/4000 [06:22<11:52,  3.86it/s]

train:  35%|███▌      | 1246/3515 [06:20<09:47,  3.86it/s]
overall:  31%|███       | 1247/4000 [06:23<12:17,  3.73it/s]

train:  35%|███▌      | 1247/3515 [06:20<10:07,  3.73it/s]
overall:  31%|███       | 1248/4000 [06:23<12:09,  3.77it/s]

train:  36%|███▌      | 1248/3515 [06:20<10:00,  3.77it/s]
overall:  31%|███▏      | 1250/4000 [06:23<09:56,  4.61it/s]

train:  36%|███▌      | 1250/3515 [06:20<08:11,  4.61it/s]
overall:  31%|███▏      | 1251/4000 [06:23<10:37,  4.31it/s]

train:  36%|███▌      | 1251/3515 [06:21<08:44,  4.31it/s]
overall:  31%|███▏      | 1252/4000 [06:24<11:00,  4.16it/s]

train:  36%|███▌      | 1252/3515 [06:21<09:03,  4.16it/s]
overall:  31%|███▏      | 1253/4000 [06:24<11:25,  4.01it/s]

train:  36%|███▌      | 1253/3515 [06:21<09:24,  4.01it/s]
overall:  31%|███▏      | 1254/4000 [06:24<11:36,  3.94it/s]

train:  36%|███▌      | 1254/3515 [06:22<09:33,  3.94it/s]
overall:  31%|███▏      | 1255/4000 [06:24<11:52,  3.86it/s]

train:  36%|███▌      | 1255/3515 [06:22<09:46,  3.86it/s]
overall:  31%|███▏      | 1256/4000 [06:25<11:42,  3.91it/s]

train:  36%|███▌      | 1256/3515 [06:22<09:38,  3.91it/s]
overall:  31%|███▏      | 1257/4000 [06:25<11:51,  3.86it/s]

train:  36%|███▌      | 1257/3515 [06:22<09:45,  3.86it/s]
overall:  31%|███▏      | 1258/4000 [06:25<14:37,  3.12it/s]

train:  36%|███▌      | 1258/3515 [06:23<12:02,  3.12it/s]
overall:  31%|███▏      | 1259/4000 [06:26<13:59,  3.26it/s]

train:  36%|███▌      | 1259/3515 [06:23<11:31,  3.26it/s]train loss: 0.5757993459701538
train per: 0.16558057617310354

overall:  32%|███▏      | 1260/4000 [06:26<13:31,  3.38it/s]

train:  36%|███▌      | 1260/3515 [06:23<11:10,  3.36it/s]
overall:  32%|███▏      | 1261/4000 [06:26<13:12,  3.45it/s]

train:  36%|███▌      | 1261/3515 [06:24<10:51,  3.46it/s]
overall:  32%|███▏      | 1263/4000 [06:27<11:22,  4.01it/s]

train:  36%|███▌      | 1263/3515 [06:24<09:20,  4.01it/s]
overall:  32%|███▏      | 1264/4000 [06:27<10:55,  4.17it/s]

train:  36%|███▌      | 1264/3515 [06:24<08:59,  4.17it/s]
overall:  32%|███▏      | 1265/4000 [06:27<11:15,  4.05it/s]

train:  36%|███▌      | 1265/3515 [06:25<09:15,  4.05it/s]
overall:  32%|███▏      | 1266/4000 [06:27<11:34,  3.94it/s]

train:  36%|███▌      | 1266/3515 [06:25<09:31,  3.94it/s]
overall:  32%|███▏      | 1267/4000 [06:28<11:47,  3.87it/s]

train:  36%|███▌      | 1267/3515 [06:25<09:41,  3.87it/s]
overall:  32%|███▏      | 1268/4000 [06:28<10:04,  4.52it/s]

train:  36%|███▌      | 1268/3515 [06:25<08:17,  4.52it/s]
overall:  32%|███▏      | 1269/4000 [06:28<10:09,  4.48it/s]

train:  36%|███▌      | 1269/3515 [06:25<08:21,  4.48it/s]
overall:  32%|███▏      | 1270/4000 [06:28<10:48,  4.21it/s]

train:  36%|███▌      | 1270/3515 [06:26<08:53,  4.21it/s]
overall:  32%|███▏      | 1271/4000 [06:29<11:08,  4.08it/s]

train:  36%|███▌      | 1271/3515 [06:26<09:09,  4.08it/s]
overall:  32%|███▏      | 1272/4000 [06:29<11:24,  3.98it/s]

train:  36%|███▌      | 1272/3515 [06:26<09:23,  3.98it/s]
overall:  32%|███▏      | 1273/4000 [06:29<11:40,  3.89it/s]

train:  36%|███▌      | 1273/3515 [06:26<09:36,  3.89it/s]
overall:  32%|███▏      | 1274/4000 [06:29<11:50,  3.84it/s]

train:  36%|███▌      | 1274/3515 [06:27<09:43,  3.84it/s]
overall:  32%|███▏      | 1275/4000 [06:30<11:49,  3.84it/s]

train:  36%|███▋      | 1275/3515 [06:27<09:43,  3.84it/s]
overall:  32%|███▏      | 1276/4000 [06:30<11:05,  4.09it/s]

train:  36%|███▋      | 1276/3515 [06:27<09:07,  4.09it/s]
overall:  32%|███▏      | 1277/4000 [06:30<11:19,  4.01it/s]

train:  36%|███▋      | 1277/3515 [06:27<09:18,  4.01it/s]
overall:  32%|███▏      | 1278/4000 [06:30<11:38,  3.90it/s]

train:  36%|███▋      | 1278/3515 [06:28<09:34,  3.90it/s]
overall:  32%|███▏      | 1279/4000 [06:31<11:48,  3.84it/s]

train:  36%|███▋      | 1279/3515 [06:28<09:42,  3.84it/s]train loss: 0.6027844548225403
train per: 0.17495154916815583

overall:  32%|███▏      | 1280/4000 [06:31<12:01,  3.77it/s]

train:  36%|███▋      | 1280/3515 [06:28<09:52,  3.77it/s]
overall:  32%|███▏      | 1281/4000 [06:31<12:08,  3.73it/s]

train:  36%|███▋      | 1281/3515 [06:29<09:58,  3.73it/s]
overall:  32%|███▏      | 1282/4000 [06:32<14:25,  3.14it/s]

train:  36%|███▋      | 1282/3515 [06:29<11:50,  3.14it/s]
overall:  32%|███▏      | 1283/4000 [06:32<13:38,  3.32it/s]

train:  37%|███▋      | 1283/3515 [06:29<11:12,  3.32it/s]
overall:  32%|███▏      | 1284/4000 [06:32<13:01,  3.47it/s]

train:  37%|███▋      | 1284/3515 [06:30<10:42,  3.47it/s]
overall:  32%|███▏      | 1285/4000 [06:32<12:36,  3.59it/s]

train:  37%|███▋      | 1285/3515 [06:30<10:21,  3.59it/s]
overall:  32%|███▏      | 1286/4000 [06:33<12:25,  3.64it/s]

train:  37%|███▋      | 1286/3515 [06:30<10:12,  3.64it/s]
overall:  32%|███▏      | 1287/4000 [06:33<13:25,  3.37it/s]

train:  37%|███▋      | 1287/3515 [06:30<11:01,  3.37it/s]
overall:  32%|███▏      | 1288/4000 [06:33<12:41,  3.56it/s]

train:  37%|███▋      | 1288/3515 [06:31<10:25,  3.56it/s]
overall:  32%|███▏      | 1289/4000 [06:34<12:14,  3.69it/s]

train:  37%|███▋      | 1289/3515 [06:31<10:03,  3.69it/s]
overall:  32%|███▏      | 1290/4000 [06:34<12:11,  3.70it/s]

train:  37%|███▋      | 1290/3515 [06:31<10:00,  3.70it/s]
overall:  32%|███▏      | 1291/4000 [06:34<12:11,  3.71it/s]

train:  37%|███▋      | 1291/3515 [06:31<10:00,  3.71it/s]
overall:  32%|███▏      | 1292/4000 [06:34<13:17,  3.39it/s]

train:  37%|███▋      | 1292/3515 [06:32<10:54,  3.39it/s]
overall:  32%|███▏      | 1293/4000 [06:35<12:52,  3.50it/s]

train:  37%|███▋      | 1293/3515 [06:32<10:34,  3.50it/s]
overall:  32%|███▏      | 1294/4000 [06:35<12:33,  3.59it/s]

train:  37%|███▋      | 1294/3515 [06:32<10:18,  3.59it/s]
overall:  32%|███▏      | 1295/4000 [06:35<12:19,  3.66it/s]

train:  37%|███▋      | 1295/3515 [06:33<10:07,  3.66it/s]
overall:  32%|███▏      | 1297/4000 [06:36<10:23,  4.34it/s]

train:  37%|███▋      | 1297/3515 [06:33<08:31,  4.34it/s]
overall:  32%|███▏      | 1298/4000 [06:36<10:45,  4.19it/s]

train:  37%|███▋      | 1298/3515 [06:33<08:49,  4.19it/s]
overall:  32%|███▏      | 1299/4000 [06:36<11:02,  4.07it/s]

train:  37%|███▋      | 1299/3515 [06:33<09:03,  4.07it/s]train loss: 0.6126611232757568
train per: 0.16930640317737092

overall:  32%|███▎      | 1300/4000 [06:36<11:38,  3.87it/s]

train:  37%|███▋      | 1300/3515 [06:34<09:33,  3.86it/s]
overall:  33%|███▎      | 1301/4000 [06:37<11:46,  3.82it/s]

train:  37%|███▋      | 1301/3515 [06:34<09:39,  3.82it/s]
overall:  33%|███▎      | 1302/4000 [06:37<11:50,  3.80it/s]

train:  37%|███▋      | 1302/3515 [06:34<09:42,  3.80it/s]
overall:  33%|███▎      | 1303/4000 [06:37<10:22,  4.33it/s]

train:  37%|███▋      | 1303/3515 [06:34<08:30,  4.34it/s]
overall:  33%|███▎      | 1304/4000 [06:37<10:41,  4.20it/s]

train:  37%|███▋      | 1304/3515 [06:35<08:45,  4.20it/s]
overall:  33%|███▎      | 1305/4000 [06:38<10:51,  4.14it/s]

train:  37%|███▋      | 1305/3515 [06:35<08:54,  4.14it/s]
overall:  33%|███▎      | 1306/4000 [06:38<13:31,  3.32it/s]

train:  37%|███▋      | 1306/3515 [06:35<11:05,  3.32it/s]
overall:  33%|███▎      | 1307/4000 [06:38<12:45,  3.52it/s]

train:  37%|███▋      | 1307/3515 [06:36<10:27,  3.52it/s]
overall:  33%|███▎      | 1308/4000 [06:39<12:32,  3.58it/s]

train:  37%|███▋      | 1308/3515 [06:36<10:17,  3.58it/s]
overall:  33%|███▎      | 1309/4000 [06:39<12:21,  3.63it/s]

train:  37%|███▋      | 1309/3515 [06:36<10:07,  3.63it/s]
overall:  33%|███▎      | 1310/4000 [06:39<12:25,  3.61it/s]

train:  37%|███▋      | 1310/3515 [06:36<10:11,  3.61it/s]
overall:  33%|███▎      | 1311/4000 [06:39<13:21,  3.36it/s]

train:  37%|███▋      | 1311/3515 [06:37<10:56,  3.36it/s]
overall:  33%|███▎      | 1312/4000 [06:40<11:18,  3.96it/s]

train:  37%|███▋      | 1312/3515 [06:37<09:16,  3.96it/s]
overall:  33%|███▎      | 1313/4000 [06:40<11:31,  3.89it/s]

train:  37%|███▋      | 1313/3515 [06:37<09:26,  3.89it/s]
overall:  33%|███▎      | 1314/4000 [06:40<11:46,  3.80it/s]

train:  37%|███▋      | 1314/3515 [06:37<09:39,  3.80it/s]
overall:  33%|███▎      | 1315/4000 [06:40<11:51,  3.77it/s]

train:  37%|███▋      | 1315/3515 [06:38<09:42,  3.77it/s]
overall:  33%|███▎      | 1316/4000 [06:41<12:44,  3.51it/s]

train:  37%|███▋      | 1316/3515 [06:38<10:26,  3.51it/s]
overall:  33%|███▎      | 1317/4000 [06:41<12:11,  3.67it/s]

train:  37%|███▋      | 1317/3515 [06:38<09:59,  3.67it/s]
overall:  33%|███▎      | 1318/4000 [06:41<12:06,  3.69it/s]

train:  37%|███▋      | 1318/3515 [06:39<09:54,  3.69it/s]
overall:  33%|███▎      | 1319/4000 [06:42<12:28,  3.58it/s]

train:  38%|███▊      | 1319/3515 [06:39<10:12,  3.58it/s]train loss: 0.6086606383323669
train per: 0.1739781013229461

overall:  33%|███▎      | 1320/4000 [06:42<10:31,  4.24it/s]

train:  38%|███▊      | 1320/3515 [06:39<08:37,  4.24it/s]
overall:  33%|███▎      | 1321/4000 [06:42<10:46,  4.15it/s]

train:  38%|███▊      | 1321/3515 [06:39<08:48,  4.15it/s]
overall:  33%|███▎      | 1322/4000 [06:42<11:01,  4.05it/s]

train:  38%|███▊      | 1322/3515 [06:40<09:01,  4.05it/s]
overall:  33%|███▎      | 1324/4000 [06:43<09:25,  4.74it/s]

train:  38%|███▊      | 1324/3515 [06:40<07:42,  4.74it/s]
overall:  33%|███▎      | 1325/4000 [06:43<09:41,  4.60it/s]

train:  38%|███▊      | 1325/3515 [06:40<07:55,  4.60it/s]
overall:  33%|███▎      | 1326/4000 [06:43<10:17,  4.33it/s]

train:  38%|███▊      | 1326/3515 [06:40<08:25,  4.33it/s]
overall:  33%|███▎      | 1327/4000 [06:43<10:40,  4.17it/s]

train:  38%|███▊      | 1327/3515 [06:41<08:44,  4.17it/s]
overall:  33%|███▎      | 1328/4000 [06:44<11:06,  4.01it/s]

train:  38%|███▊      | 1328/3515 [06:41<09:05,  4.01it/s]
overall:  33%|███▎      | 1329/4000 [06:44<11:00,  4.05it/s]

train:  38%|███▊      | 1329/3515 [06:41<09:00,  4.05it/s]
overall:  33%|███▎      | 1330/4000 [06:44<13:29,  3.30it/s]

train:  38%|███▊      | 1330/3515 [06:42<11:02,  3.30it/s]
overall:  33%|███▎      | 1331/4000 [06:44<12:53,  3.45it/s]

train:  38%|███▊      | 1331/3515 [06:42<10:32,  3.45it/s]
overall:  33%|███▎      | 1332/4000 [06:45<12:29,  3.56it/s]

train:  38%|███▊      | 1332/3515 [06:42<10:13,  3.56it/s]
overall:  33%|███▎      | 1333/4000 [06:45<10:37,  4.18it/s]

train:  38%|███▊      | 1333/3515 [06:42<08:41,  4.18it/s]
overall:  33%|███▎      | 1334/4000 [06:45<11:08,  3.99it/s]

train:  38%|███▊      | 1334/3515 [06:43<09:06,  3.99it/s]
overall:  33%|███▎      | 1335/4000 [06:45<11:39,  3.81it/s]

train:  38%|███▊      | 1335/3515 [06:43<09:31,  3.81it/s]
overall:  33%|███▎      | 1336/4000 [06:46<11:22,  3.90it/s]

train:  38%|███▊      | 1336/3515 [06:43<09:18,  3.90it/s]
overall:  33%|███▎      | 1337/4000 [06:46<11:08,  3.98it/s]

train:  38%|███▊      | 1337/3515 [06:43<09:06,  3.98it/s]
overall:  33%|███▎      | 1338/4000 [06:46<11:33,  3.84it/s]

train:  38%|███▊      | 1338/3515 [06:44<09:27,  3.84it/s]
overall:  33%|███▎      | 1339/4000 [06:46<11:44,  3.78it/s]

train:  38%|███▊      | 1339/3515 [06:44<09:36,  3.78it/s]train loss: 0.5771068334579468
train per: 0.1696265601105027

overall:  34%|███▎      | 1340/4000 [06:47<12:40,  3.50it/s]

train:  38%|███▊      | 1340/3515 [06:44<10:22,  3.50it/s]
overall:  34%|███▎      | 1341/4000 [06:47<12:35,  3.52it/s]

train:  38%|███▊      | 1341/3515 [06:44<10:17,  3.52it/s]
overall:  34%|███▎      | 1342/4000 [06:47<12:22,  3.58it/s]

train:  38%|███▊      | 1342/3515 [06:45<10:06,  3.58it/s]
overall:  34%|███▎      | 1343/4000 [06:48<11:26,  3.87it/s]

train:  38%|███▊      | 1343/3515 [06:45<09:21,  3.87it/s]
overall:  34%|███▎      | 1344/4000 [06:48<11:27,  3.86it/s]

train:  38%|███▊      | 1344/3515 [06:45<09:21,  3.86it/s]
overall:  34%|███▎      | 1345/4000 [06:48<11:36,  3.81it/s]

train:  38%|███▊      | 1345/3515 [06:45<09:29,  3.81it/s]
overall:  34%|███▎      | 1346/4000 [06:48<11:29,  3.85it/s]

train:  38%|███▊      | 1346/3515 [06:46<09:23,  3.85it/s]
overall:  34%|███▎      | 1347/4000 [06:49<11:27,  3.86it/s]

train:  38%|███▊      | 1347/3515 [06:46<09:21,  3.86it/s]
overall:  34%|███▎      | 1348/4000 [06:49<11:34,  3.82it/s]

train:  38%|███▊      | 1348/3515 [06:46<09:27,  3.82it/s]
overall:  34%|███▎      | 1349/4000 [06:49<11:42,  3.77it/s]

train:  38%|███▊      | 1349/3515 [06:47<09:33,  3.77it/s]
overall:  34%|███▍      | 1350/4000 [06:49<11:42,  3.77it/s]

train:  38%|███▊      | 1350/3515 [06:47<09:34,  3.77it/s]
overall:  34%|███▍      | 1351/4000 [06:50<11:25,  3.86it/s]

train:  38%|███▊      | 1351/3515 [06:47<09:20,  3.86it/s]
overall:  34%|███▍      | 1352/4000 [06:50<11:37,  3.79it/s]

train:  38%|███▊      | 1352/3515 [06:47<09:30,  3.79it/s]
overall:  34%|███▍      | 1353/4000 [06:50<11:32,  3.82it/s]

train:  38%|███▊      | 1353/3515 [06:48<09:26,  3.82it/s]
overall:  34%|███▍      | 1354/4000 [06:51<13:54,  3.17it/s]

train:  39%|███▊      | 1354/3515 [06:48<11:21,  3.17it/s]
overall:  34%|███▍      | 1355/4000 [06:51<13:13,  3.33it/s]

train:  39%|███▊      | 1355/3515 [06:48<10:48,  3.33it/s]
overall:  34%|███▍      | 1356/4000 [06:51<12:51,  3.43it/s]

train:  39%|███▊      | 1356/3515 [06:49<10:29,  3.43it/s]
overall:  34%|███▍      | 1357/4000 [06:51<12:41,  3.47it/s]

train:  39%|███▊      | 1357/3515 [06:49<10:22,  3.47it/s]
overall:  34%|███▍      | 1358/4000 [06:52<12:28,  3.53it/s]

train:  39%|███▊      | 1358/3515 [06:49<10:11,  3.53it/s]
overall:  34%|███▍      | 1359/4000 [06:52<13:11,  3.34it/s]

train:  39%|███▊      | 1359/3515 [06:49<10:46,  3.34it/s]train loss: 0.5829333066940308
train per: 0.17247112549602076

overall:  34%|███▍      | 1360/4000 [06:52<12:54,  3.41it/s]

train:  39%|███▊      | 1360/3515 [06:50<10:32,  3.41it/s]
overall:  34%|███▍      | 1361/4000 [06:53<12:24,  3.55it/s]

train:  39%|███▊      | 1361/3515 [06:50<10:07,  3.55it/s]
overall:  34%|███▍      | 1362/4000 [06:53<12:32,  3.51it/s]

train:  39%|███▊      | 1362/3515 [06:50<10:13,  3.51it/s]
overall:  34%|███▍      | 1363/4000 [06:53<12:15,  3.59it/s]

train:  39%|███▉      | 1363/3515 [06:51<09:59,  3.59it/s]
overall:  34%|███▍      | 1364/4000 [06:54<12:59,  3.38it/s]

train:  39%|███▉      | 1364/3515 [06:51<10:35,  3.38it/s]
overall:  34%|███▍      | 1365/4000 [06:54<12:40,  3.47it/s]

train:  39%|███▉      | 1365/3515 [06:51<10:20,  3.47it/s]
overall:  34%|███▍      | 1367/4000 [06:54<10:21,  4.23it/s]

train:  39%|███▉      | 1367/3515 [06:51<08:27,  4.23it/s]
overall:  34%|███▍      | 1368/4000 [06:54<10:26,  4.20it/s]

train:  39%|███▉      | 1368/3515 [06:52<08:31,  4.20it/s]
overall:  34%|███▍      | 1369/4000 [06:55<10:01,  4.37it/s]

train:  39%|███▉      | 1369/3515 [06:52<08:10,  4.37it/s]
overall:  34%|███▍      | 1370/4000 [06:55<10:26,  4.20it/s]

train:  39%|███▉      | 1370/3515 [06:52<08:30,  4.20it/s]
overall:  34%|███▍      | 1371/4000 [06:55<10:54,  4.02it/s]

train:  39%|███▉      | 1371/3515 [06:52<08:53,  4.02it/s]
overall:  34%|███▍      | 1372/4000 [06:55<11:12,  3.91it/s]

train:  39%|███▉      | 1372/3515 [06:53<09:08,  3.91it/s]
overall:  34%|███▍      | 1373/4000 [06:56<11:23,  3.84it/s]

train:  39%|███▉      | 1373/3515 [06:53<09:17,  3.84it/s]
overall:  34%|███▍      | 1374/4000 [06:56<11:16,  3.88it/s]

train:  39%|███▉      | 1374/3515 [06:53<09:11,  3.88it/s]
overall:  34%|███▍      | 1375/4000 [06:56<11:26,  3.82it/s]

train:  39%|███▉      | 1375/3515 [06:54<09:19,  3.82it/s]
overall:  34%|███▍      | 1376/4000 [06:56<11:35,  3.77it/s]

train:  39%|███▉      | 1376/3515 [06:54<09:26,  3.77it/s]
overall:  34%|███▍      | 1377/4000 [06:57<11:33,  3.78it/s]

train:  39%|███▉      | 1377/3515 [06:54<09:24,  3.78it/s]
overall:  34%|███▍      | 1378/4000 [06:57<13:51,  3.15it/s]

train:  39%|███▉      | 1378/3515 [06:55<11:17,  3.15it/s]
overall:  34%|███▍      | 1379/4000 [06:57<13:04,  3.34it/s]

train:  39%|███▉      | 1379/3515 [06:55<10:39,  3.34it/s]train loss: 0.5711101293563843
train per: 0.16722334641733916

overall:  34%|███▍      | 1380/4000 [06:58<11:52,  3.68it/s]

train:  39%|███▉      | 1380/3515 [06:55<09:41,  3.67it/s]
overall:  35%|███▍      | 1381/4000 [06:58<12:04,  3.62it/s]

train:  39%|███▉      | 1381/3515 [06:55<09:50,  3.62it/s]
overall:  35%|███▍      | 1382/4000 [06:58<11:49,  3.69it/s]

train:  39%|███▉      | 1382/3515 [06:56<09:37,  3.69it/s]
overall:  35%|███▍      | 1383/4000 [06:58<12:35,  3.47it/s]

train:  39%|███▉      | 1383/3515 [06:56<10:15,  3.47it/s]
overall:  35%|███▍      | 1384/4000 [06:59<11:26,  3.81it/s]

train:  39%|███▉      | 1384/3515 [06:56<09:18,  3.81it/s]
overall:  35%|███▍      | 1385/4000 [06:59<10:02,  4.34it/s]

train:  39%|███▉      | 1385/3515 [06:56<08:10,  4.34it/s]
overall:  35%|███▍      | 1386/4000 [06:59<10:26,  4.17it/s]

train:  39%|███▉      | 1386/3515 [06:56<08:30,  4.17it/s]
overall:  35%|███▍      | 1387/4000 [06:59<10:05,  4.32it/s]

train:  39%|███▉      | 1387/3515 [06:57<08:13,  4.32it/s]
overall:  35%|███▍      | 1388/4000 [06:59<09:10,  4.74it/s]

train:  39%|███▉      | 1388/3515 [06:57<07:28,  4.74it/s]
overall:  35%|███▍      | 1389/4000 [07:00<09:57,  4.37it/s]

train:  40%|███▉      | 1389/3515 [06:57<08:06,  4.37it/s]
overall:  35%|███▍      | 1390/4000 [07:00<10:30,  4.14it/s]

train:  40%|███▉      | 1390/3515 [06:57<08:33,  4.14it/s]
overall:  35%|███▍      | 1391/4000 [07:00<10:54,  3.99it/s]

train:  40%|███▉      | 1391/3515 [06:58<08:52,  3.99it/s]
overall:  35%|███▍      | 1392/4000 [07:01<11:04,  3.93it/s]

train:  40%|███▉      | 1392/3515 [06:58<09:00,  3.93it/s]
overall:  35%|███▍      | 1393/4000 [07:01<11:11,  3.88it/s]

train:  40%|███▉      | 1393/3515 [06:58<09:06,  3.88it/s]
overall:  35%|███▍      | 1394/4000 [07:01<11:08,  3.90it/s]

train:  40%|███▉      | 1394/3515 [06:58<09:04,  3.90it/s]
overall:  35%|███▍      | 1395/4000 [07:01<11:20,  3.83it/s]

train:  40%|███▉      | 1395/3515 [06:59<09:13,  3.83it/s]
overall:  35%|███▍      | 1396/4000 [07:02<11:17,  3.85it/s]

train:  40%|███▉      | 1396/3515 [06:59<09:11,  3.85it/s]
overall:  35%|███▍      | 1397/4000 [07:02<11:15,  3.85it/s]

train:  40%|███▉      | 1397/3515 [06:59<09:09,  3.85it/s]
overall:  35%|███▍      | 1398/4000 [07:02<11:22,  3.81it/s]

train:  40%|███▉      | 1398/3515 [07:00<09:15,  3.81it/s]
overall:  35%|███▍      | 1399/4000 [07:02<11:20,  3.82it/s]

train:  40%|███▉      | 1399/3515 [07:00<09:13,  3.82it/s]train loss: 0.5645450353622437
train per: 0.16304261038511697

overall:  35%|███▌      | 1400/4000 [07:03<09:44,  4.45it/s]

train:  40%|███▉      | 1400/3515 [07:00<07:56,  4.44it/s]
overall:  35%|███▌      | 1401/4000 [07:03<10:06,  4.29it/s]

train:  40%|███▉      | 1401/3515 [07:00<08:13,  4.29it/s]
overall:  35%|███▌      | 1402/4000 [07:03<10:56,  3.96it/s]

train:  40%|███▉      | 1402/3515 [07:00<08:54,  3.96it/s]
overall:  35%|███▌      | 1403/4000 [07:03<10:58,  3.94it/s]

train:  40%|███▉      | 1403/3515 [07:01<08:55,  3.94it/s]
overall:  35%|███▌      | 1404/4000 [07:04<09:40,  4.47it/s]

train:  40%|███▉      | 1404/3515 [07:01<07:51,  4.47it/s]
overall:  35%|███▌      | 1405/4000 [07:04<08:07,  5.32it/s]

train:  40%|███▉      | 1405/3515 [07:01<06:36,  5.32it/s]
overall:  35%|███▌      | 1406/4000 [07:04<09:22,  4.61it/s]

train:  40%|████      | 1406/3515 [07:01<07:37,  4.61it/s]
overall:  35%|███▌      | 1407/4000 [07:04<11:09,  3.87it/s]

train:  40%|████      | 1407/3515 [07:02<09:04,  3.87it/s]
overall:  35%|███▌      | 1408/4000 [07:05<11:06,  3.89it/s]

train:  40%|████      | 1408/3515 [07:02<09:02,  3.89it/s]
overall:  35%|███▌      | 1409/4000 [07:05<11:14,  3.84it/s]

train:  40%|████      | 1409/3515 [07:02<09:08,  3.84it/s]
overall:  35%|███▌      | 1411/4000 [07:05<09:29,  4.55it/s]

train:  40%|████      | 1411/3515 [07:02<07:42,  4.55it/s]
overall:  35%|███▌      | 1412/4000 [07:05<10:40,  4.04it/s]

train:  40%|████      | 1412/3515 [07:03<08:40,  4.04it/s]
overall:  35%|███▌      | 1413/4000 [07:06<09:51,  4.37it/s]

train:  40%|████      | 1413/3515 [07:03<08:00,  4.37it/s]
overall:  35%|███▌      | 1414/4000 [07:06<10:21,  4.16it/s]

train:  40%|████      | 1414/3515 [07:03<08:25,  4.16it/s]
overall:  35%|███▌      | 1415/4000 [07:06<09:09,  4.70it/s]

train:  40%|████      | 1415/3515 [07:03<07:26,  4.70it/s]
overall:  35%|███▌      | 1417/4000 [07:06<08:17,  5.19it/s]

train:  40%|████      | 1417/3515 [07:04<06:43,  5.19it/s]
overall:  35%|███▌      | 1418/4000 [07:07<08:47,  4.89it/s]

train:  40%|████      | 1418/3515 [07:04<07:08,  4.89it/s]
overall:  35%|███▌      | 1419/4000 [07:07<09:01,  4.77it/s]

train:  40%|████      | 1419/3515 [07:04<07:19,  4.77it/s]train loss: 0.5863474607467651
train per: 0.1724807939497205

overall:  36%|███▌      | 1420/4000 [07:07<09:52,  4.36it/s]

train:  40%|████      | 1420/3515 [07:04<08:01,  4.35it/s]
overall:  36%|███▌      | 1421/4000 [07:07<10:23,  4.14it/s]

train:  40%|████      | 1421/3515 [07:05<08:25,  4.14it/s]
overall:  36%|███▌      | 1422/4000 [07:08<09:09,  4.69it/s]

train:  40%|████      | 1422/3515 [07:05<07:25,  4.70it/s]
overall:  36%|███▌      | 1423/4000 [07:08<09:08,  4.70it/s]

train:  40%|████      | 1423/3515 [07:05<07:25,  4.70it/s]
overall:  36%|███▌      | 1424/4000 [07:08<09:30,  4.52it/s]

train:  41%|████      | 1424/3515 [07:05<07:42,  4.52it/s]
overall:  36%|███▌      | 1425/4000 [07:08<09:26,  4.55it/s]

train:  41%|████      | 1425/3515 [07:06<07:39,  4.55it/s]
overall:  36%|███▌      | 1426/4000 [07:09<12:08,  3.54it/s]

train:  41%|████      | 1426/3515 [07:06<09:50,  3.54it/s]
overall:  36%|███▌      | 1427/4000 [07:09<11:45,  3.65it/s]

train:  41%|████      | 1427/3515 [07:06<09:32,  3.65it/s]
overall:  36%|███▌      | 1428/4000 [07:09<11:42,  3.66it/s]

train:  41%|████      | 1428/3515 [07:07<09:29,  3.66it/s]
overall:  36%|███▌      | 1429/4000 [07:09<11:31,  3.72it/s]

train:  41%|████      | 1429/3515 [07:07<09:20,  3.72it/s]
overall:  36%|███▌      | 1430/4000 [07:10<11:26,  3.74it/s]

train:  41%|████      | 1430/3515 [07:07<09:17,  3.74it/s]
overall:  36%|███▌      | 1431/4000 [07:10<12:23,  3.46it/s]

train:  41%|████      | 1431/3515 [07:07<10:02,  3.46it/s]
overall:  36%|███▌      | 1432/4000 [07:10<12:07,  3.53it/s]

train:  41%|████      | 1432/3515 [07:08<09:50,  3.53it/s]
overall:  36%|███▌      | 1433/4000 [07:11<11:43,  3.65it/s]

train:  41%|████      | 1433/3515 [07:08<09:30,  3.65it/s]
overall:  36%|███▌      | 1434/4000 [07:11<11:48,  3.62it/s]

train:  41%|████      | 1434/3515 [07:08<09:34,  3.62it/s]
overall:  36%|███▌      | 1435/4000 [07:11<11:38,  3.67it/s]

train:  41%|████      | 1435/3515 [07:08<09:26,  3.67it/s]
overall:  36%|███▌      | 1436/4000 [07:11<12:06,  3.53it/s]

train:  41%|████      | 1436/3515 [07:09<09:48,  3.53it/s]
overall:  36%|███▌      | 1437/4000 [07:12<11:55,  3.58it/s]

train:  41%|████      | 1437/3515 [07:09<09:39,  3.58it/s]
overall:  36%|███▌      | 1438/4000 [07:12<11:45,  3.63it/s]

train:  41%|████      | 1438/3515 [07:09<09:32,  3.63it/s]
overall:  36%|███▌      | 1439/4000 [07:12<11:36,  3.68it/s]

train:  41%|████      | 1439/3515 [07:10<09:24,  3.68it/s]train loss: 0.5984925031661987
train per: 0.17289095975087218

overall:  36%|███▌      | 1440/4000 [07:12<11:29,  3.71it/s]

train:  41%|████      | 1440/3515 [07:10<09:19,  3.71it/s]
overall:  36%|███▌      | 1441/4000 [07:13<11:29,  3.71it/s]

train:  41%|████      | 1441/3515 [07:10<09:18,  3.71it/s]
overall:  36%|███▌      | 1442/4000 [07:13<11:31,  3.70it/s]

train:  41%|████      | 1442/3515 [07:10<09:20,  3.70it/s]
overall:  36%|███▌      | 1443/4000 [07:13<11:09,  3.82it/s]

train:  41%|████      | 1443/3515 [07:11<09:02,  3.82it/s]
overall:  36%|███▌      | 1444/4000 [07:14<11:14,  3.79it/s]

train:  41%|████      | 1444/3515 [07:11<09:06,  3.79it/s]
overall:  36%|███▌      | 1445/4000 [07:14<11:16,  3.78it/s]

train:  41%|████      | 1445/3515 [07:11<09:07,  3.78it/s]
overall:  36%|███▌      | 1446/4000 [07:14<11:18,  3.76it/s]

train:  41%|████      | 1446/3515 [07:11<09:09,  3.76it/s]
overall:  36%|███▌      | 1447/4000 [07:14<10:13,  4.16it/s]

train:  41%|████      | 1447/3515 [07:12<08:16,  4.16it/s]
overall:  36%|███▌      | 1448/4000 [07:15<10:39,  3.99it/s]

train:  41%|████      | 1448/3515 [07:12<08:38,  3.99it/s]
overall:  36%|███▌      | 1449/4000 [07:15<10:58,  3.87it/s]

train:  41%|████      | 1449/3515 [07:12<08:53,  3.87it/s]
overall:  36%|███▋      | 1450/4000 [07:15<13:12,  3.22it/s]

train:  41%|████▏     | 1450/3515 [07:13<10:41,  3.22it/s]
overall:  36%|███▋      | 1451/4000 [07:15<12:10,  3.49it/s]

train:  41%|████▏     | 1451/3515 [07:13<09:51,  3.49it/s]
overall:  36%|███▋      | 1452/4000 [07:16<11:24,  3.72it/s]

train:  41%|████▏     | 1452/3515 [07:13<09:14,  3.72it/s]
overall:  36%|███▋      | 1453/4000 [07:16<09:37,  4.41it/s]

train:  41%|████▏     | 1453/3515 [07:13<07:47,  4.41it/s]
overall:  36%|███▋      | 1454/4000 [07:16<10:11,  4.16it/s]

train:  41%|████▏     | 1454/3515 [07:13<08:14,  4.16it/s]
overall:  36%|███▋      | 1455/4000 [07:16<11:13,  3.78it/s]

train:  41%|████▏     | 1455/3515 [07:14<09:05,  3.78it/s]
overall:  36%|███▋      | 1456/4000 [07:17<10:43,  3.96it/s]

train:  41%|████▏     | 1456/3515 [07:14<08:40,  3.96it/s]
overall:  36%|███▋      | 1457/4000 [07:17<10:07,  4.19it/s]

train:  41%|████▏     | 1457/3515 [07:14<08:11,  4.19it/s]
overall:  36%|███▋      | 1459/4000 [07:17<06:59,  6.06it/s]

train:  42%|████▏     | 1459/3515 [07:14<05:39,  6.06it/s]train loss: 0.5734618902206421
train per: 0.1676918735891648

overall:  36%|███▋      | 1460/4000 [07:17<08:44,  4.85it/s]

train:  42%|████▏     | 1460/3515 [07:15<07:04,  4.84it/s]
overall:  37%|███▋      | 1461/4000 [07:18<09:23,  4.51it/s]

train:  42%|████▏     | 1461/3515 [07:15<07:35,  4.51it/s]
overall:  37%|███▋      | 1462/4000 [07:18<09:59,  4.24it/s]

train:  42%|████▏     | 1462/3515 [07:15<08:04,  4.24it/s]
overall:  37%|███▋      | 1463/4000 [07:18<10:15,  4.12it/s]

train:  42%|████▏     | 1463/3515 [07:15<08:18,  4.12it/s]
overall:  37%|███▋      | 1464/4000 [07:18<10:32,  4.01it/s]

train:  42%|████▏     | 1464/3515 [07:16<08:31,  4.01it/s]
overall:  37%|███▋      | 1465/4000 [07:18<08:44,  4.83it/s]

train:  42%|████▏     | 1465/3515 [07:16<07:04,  4.83it/s]
overall:  37%|███▋      | 1466/4000 [07:19<09:33,  4.42it/s]

train:  42%|████▏     | 1466/3515 [07:16<07:44,  4.42it/s]
overall:  37%|███▋      | 1467/4000 [07:19<10:02,  4.21it/s]

train:  42%|████▏     | 1467/3515 [07:16<08:06,  4.21it/s]
overall:  37%|███▋      | 1468/4000 [07:19<10:31,  4.01it/s]

train:  42%|████▏     | 1468/3515 [07:17<08:30,  4.01it/s]
overall:  37%|███▋      | 1469/4000 [07:20<10:50,  3.89it/s]

train:  42%|████▏     | 1469/3515 [07:17<08:45,  3.89it/s]
overall:  37%|███▋      | 1470/4000 [07:20<11:03,  3.81it/s]

train:  42%|████▏     | 1470/3515 [07:17<08:56,  3.81it/s]
overall:  37%|███▋      | 1471/4000 [07:20<10:45,  3.92it/s]

train:  42%|████▏     | 1471/3515 [07:17<08:41,  3.92it/s]
overall:  37%|███▋      | 1472/4000 [07:20<11:01,  3.82it/s]

train:  42%|████▏     | 1472/3515 [07:18<08:54,  3.82it/s]
overall:  37%|███▋      | 1473/4000 [07:21<10:57,  3.85it/s]

train:  42%|████▏     | 1473/3515 [07:18<08:50,  3.85it/s]
overall:  37%|███▋      | 1474/4000 [07:21<13:03,  3.22it/s]

train:  42%|████▏     | 1474/3515 [07:18<10:33,  3.22it/s]
overall:  37%|███▋      | 1475/4000 [07:21<10:31,  4.00it/s]

train:  42%|████▏     | 1475/3515 [07:19<08:30,  4.00it/s]
overall:  37%|███▋      | 1476/4000 [07:21<10:52,  3.87it/s]

train:  42%|████▏     | 1476/3515 [07:19<08:46,  3.87it/s]
overall:  37%|███▋      | 1477/4000 [07:22<10:55,  3.85it/s]

train:  42%|████▏     | 1477/3515 [07:19<08:49,  3.85it/s]
overall:  37%|███▋      | 1478/4000 [07:22<10:59,  3.82it/s]

train:  42%|████▏     | 1478/3515 [07:19<08:52,  3.82it/s]
overall:  37%|███▋      | 1479/4000 [07:22<10:24,  4.03it/s]

train:  42%|████▏     | 1479/3515 [07:20<08:24,  4.03it/s]train loss: 0.6090326905250549
train per: 0.1751987130961393

overall:  37%|███▋      | 1480/4000 [07:22<10:24,  4.03it/s]

train:  42%|████▏     | 1480/3515 [07:20<08:24,  4.03it/s]
overall:  37%|███▋      | 1481/4000 [07:23<10:30,  3.99it/s]

train:  42%|████▏     | 1481/3515 [07:20<08:29,  4.00it/s]
overall:  37%|███▋      | 1482/4000 [07:23<09:53,  4.25it/s]

train:  42%|████▏     | 1482/3515 [07:20<07:58,  4.25it/s]
overall:  37%|███▋      | 1483/4000 [07:23<10:11,  4.12it/s]

train:  42%|████▏     | 1483/3515 [07:21<08:13,  4.12it/s]
overall:  37%|███▋      | 1484/4000 [07:23<11:13,  3.74it/s]

train:  42%|████▏     | 1484/3515 [07:21<09:03,  3.74it/s]
overall:  37%|███▋      | 1485/4000 [07:24<10:00,  4.19it/s]

train:  42%|████▏     | 1485/3515 [07:21<08:04,  4.19it/s]
overall:  37%|███▋      | 1486/4000 [07:24<09:43,  4.31it/s]

train:  42%|████▏     | 1486/3515 [07:21<07:51,  4.31it/s]
overall:  37%|███▋      | 1487/4000 [07:24<10:25,  4.02it/s]

train:  42%|████▏     | 1487/3515 [07:22<08:24,  4.02it/s]
overall:  37%|███▋      | 1488/4000 [07:24<10:41,  3.91it/s]

train:  42%|████▏     | 1488/3515 [07:22<08:37,  3.92it/s]
overall:  37%|███▋      | 1489/4000 [07:25<10:39,  3.93it/s]

train:  42%|████▏     | 1489/3515 [07:22<08:36,  3.93it/s]
overall:  37%|███▋      | 1490/4000 [07:25<08:52,  4.71it/s]

train:  42%|████▏     | 1490/3515 [07:22<07:09,  4.71it/s]
overall:  37%|███▋      | 1491/4000 [07:25<09:15,  4.52it/s]

train:  42%|████▏     | 1491/3515 [07:22<07:28,  4.52it/s]
overall:  37%|███▋      | 1492/4000 [07:25<09:39,  4.33it/s]

train:  42%|████▏     | 1492/3515 [07:23<07:47,  4.33it/s]
overall:  37%|███▋      | 1493/4000 [07:26<10:11,  4.10it/s]

train:  42%|████▏     | 1493/3515 [07:23<08:13,  4.10it/s]
overall:  37%|███▋      | 1494/4000 [07:26<10:37,  3.93it/s]

train:  43%|████▎     | 1494/3515 [07:23<08:34,  3.93it/s]
overall:  37%|███▋      | 1495/4000 [07:26<10:28,  3.98it/s]

train:  43%|████▎     | 1495/3515 [07:23<08:27,  3.98it/s]
overall:  37%|███▋      | 1496/4000 [07:26<10:34,  3.95it/s]

train:  43%|████▎     | 1496/3515 [07:24<08:31,  3.95it/s]
overall:  37%|███▋      | 1497/4000 [07:27<10:45,  3.88it/s]

train:  43%|████▎     | 1497/3515 [07:24<08:40,  3.88it/s]
overall:  37%|███▋      | 1498/4000 [07:27<13:03,  3.19it/s]

train:  43%|████▎     | 1498/3515 [07:24<10:31,  3.19it/s]
overall:  37%|███▋      | 1499/4000 [07:27<12:09,  3.43it/s]

train:  43%|████▎     | 1499/3515 [07:25<09:47,  3.43it/s]train loss: 0.6514393091201782
train per: 0.1739201127310173

overall:  38%|███▊      | 1500/4000 [07:27<10:17,  4.05it/s]

train:  43%|████▎     | 1500/3515 [07:25<08:18,  4.04it/s]
overall:  38%|███▊      | 1501/4000 [07:28<09:42,  4.29it/s]

train:  43%|████▎     | 1501/3515 [07:25<07:49,  4.29it/s]
overall:  38%|███▊      | 1502/4000 [07:28<09:25,  4.41it/s]

train:  43%|████▎     | 1502/3515 [07:25<07:35,  4.41it/s]
overall:  38%|███▊      | 1503/4000 [07:28<10:54,  3.81it/s]

train:  43%|████▎     | 1503/3515 [07:26<08:47,  3.81it/s]
overall:  38%|███▊      | 1504/4000 [07:28<10:13,  4.07it/s]

train:  43%|████▎     | 1504/3515 [07:26<08:13,  4.07it/s]
overall:  38%|███▊      | 1505/4000 [07:29<10:32,  3.94it/s]

train:  43%|████▎     | 1505/3515 [07:26<08:29,  3.94it/s]
overall:  38%|███▊      | 1506/4000 [07:29<10:56,  3.80it/s]

train:  43%|████▎     | 1506/3515 [07:26<08:48,  3.80it/s]
overall:  38%|███▊      | 1507/4000 [07:29<10:56,  3.80it/s]

train:  43%|████▎     | 1507/3515 [07:27<08:48,  3.80it/s]
overall:  38%|███▊      | 1508/4000 [07:30<11:36,  3.58it/s]

train:  43%|████▎     | 1508/3515 [07:27<09:21,  3.58it/s]
overall:  38%|███▊      | 1509/4000 [07:30<11:33,  3.59it/s]

train:  43%|████▎     | 1509/3515 [07:27<09:18,  3.59it/s]
overall:  38%|███▊      | 1510/4000 [07:30<11:27,  3.62it/s]

train:  43%|████▎     | 1510/3515 [07:27<09:13,  3.62it/s]
overall:  38%|███▊      | 1512/4000 [07:30<09:25,  4.40it/s]

train:  43%|████▎     | 1512/3515 [07:28<07:35,  4.40it/s]
overall:  38%|███▊      | 1513/4000 [07:31<09:46,  4.24it/s]

train:  43%|████▎     | 1513/3515 [07:28<07:51,  4.24it/s]
overall:  38%|███▊      | 1514/4000 [07:31<10:14,  4.05it/s]

train:  43%|████▎     | 1514/3515 [07:28<08:14,  4.05it/s]
overall:  38%|███▊      | 1515/4000 [07:31<10:13,  4.05it/s]

train:  43%|████▎     | 1515/3515 [07:29<08:13,  4.05it/s]
overall:  38%|███▊      | 1516/4000 [07:31<10:25,  3.97it/s]

train:  43%|████▎     | 1516/3515 [07:29<08:23,  3.97it/s]
overall:  38%|███▊      | 1517/4000 [07:32<10:39,  3.88it/s]

train:  43%|████▎     | 1517/3515 [07:29<08:34,  3.88it/s]
overall:  38%|███▊      | 1518/4000 [07:32<10:50,  3.82it/s]

train:  43%|████▎     | 1518/3515 [07:29<08:43,  3.82it/s]
overall:  38%|███▊      | 1519/4000 [07:32<10:14,  4.04it/s]

train:  43%|████▎     | 1519/3515 [07:30<08:13,  4.04it/s]train loss: 0.5651698112487793
train per: 0.1670048309178744

overall:  38%|███▊      | 1520/4000 [07:32<10:26,  3.96it/s]

train:  43%|████▎     | 1520/3515 [07:30<08:24,  3.96it/s]
overall:  38%|███▊      | 1521/4000 [07:33<10:39,  3.88it/s]

train:  43%|████▎     | 1521/3515 [07:30<08:34,  3.88it/s]
overall:  38%|███▊      | 1522/4000 [07:33<12:37,  3.27it/s]

train:  43%|████▎     | 1522/3515 [07:31<10:09,  3.27it/s]
overall:  38%|███▊      | 1523/4000 [07:33<11:49,  3.49it/s]

train:  43%|████▎     | 1523/3515 [07:31<09:30,  3.49it/s]
overall:  38%|███▊      | 1524/4000 [07:34<11:34,  3.56it/s]

train:  43%|████▎     | 1524/3515 [07:31<09:18,  3.56it/s]
overall:  38%|███▊      | 1525/4000 [07:34<10:46,  3.83it/s]

train:  43%|████▎     | 1525/3515 [07:31<08:39,  3.83it/s]
overall:  38%|███▊      | 1526/4000 [07:34<10:21,  3.98it/s]

train:  43%|████▎     | 1526/3515 [07:32<08:19,  3.98it/s]
overall:  38%|███▊      | 1527/4000 [07:35<11:48,  3.49it/s]

train:  43%|████▎     | 1527/3515 [07:32<09:29,  3.49it/s]
overall:  38%|███▊      | 1529/4000 [07:35<09:27,  4.35it/s]

train:  43%|████▎     | 1529/3515 [07:32<07:36,  4.35it/s]
overall:  38%|███▊      | 1530/4000 [07:35<09:46,  4.21it/s]

train:  44%|████▎     | 1530/3515 [07:32<07:51,  4.21it/s]
overall:  38%|███▊      | 1531/4000 [07:35<09:59,  4.12it/s]

train:  44%|████▎     | 1531/3515 [07:33<08:01,  4.12it/s]
overall:  38%|███▊      | 1532/4000 [07:36<10:54,  3.77it/s]

train:  44%|████▎     | 1532/3515 [07:33<08:46,  3.77it/s]
overall:  38%|███▊      | 1533/4000 [07:36<10:55,  3.76it/s]

train:  44%|████▎     | 1533/3515 [07:33<08:46,  3.76it/s]
overall:  38%|███▊      | 1534/4000 [07:36<10:57,  3.75it/s]

train:  44%|████▎     | 1534/3515 [07:34<08:47,  3.75it/s]
overall:  38%|███▊      | 1535/4000 [07:36<11:08,  3.69it/s]

train:  44%|████▎     | 1535/3515 [07:34<08:56,  3.69it/s]
overall:  38%|███▊      | 1536/4000 [07:37<11:09,  3.68it/s]

train:  44%|████▎     | 1536/3515 [07:34<08:57,  3.68it/s]
overall:  38%|███▊      | 1537/4000 [07:37<11:09,  3.68it/s]

train:  44%|████▎     | 1537/3515 [07:34<08:57,  3.68it/s]
overall:  38%|███▊      | 1538/4000 [07:37<11:14,  3.65it/s]

train:  44%|████▍     | 1538/3515 [07:35<09:01,  3.65it/s]
overall:  38%|███▊      | 1539/4000 [07:38<11:05,  3.70it/s]

train:  44%|████▍     | 1539/3515 [07:35<08:54,  3.70it/s]train loss: 0.5694109797477722
train per: 0.1652615466848077

overall:  38%|███▊      | 1540/4000 [07:38<11:08,  3.68it/s]

train:  44%|████▍     | 1540/3515 [07:35<08:56,  3.68it/s]
overall:  39%|███▊      | 1541/4000 [07:38<11:02,  3.71it/s]

train:  44%|████▍     | 1541/3515 [07:35<08:51,  3.71it/s]
overall:  39%|███▊      | 1542/4000 [07:38<11:03,  3.71it/s]

train:  44%|████▍     | 1542/3515 [07:36<08:52,  3.71it/s]
overall:  39%|███▊      | 1543/4000 [07:39<10:57,  3.74it/s]

train:  44%|████▍     | 1543/3515 [07:36<08:47,  3.74it/s]
overall:  39%|███▊      | 1544/4000 [07:39<09:04,  4.51it/s]

train:  44%|████▍     | 1544/3515 [07:36<07:17,  4.51it/s]
overall:  39%|███▊      | 1545/4000 [07:39<09:31,  4.29it/s]

train:  44%|████▍     | 1545/3515 [07:36<07:38,  4.29it/s]
overall:  39%|███▊      | 1546/4000 [07:39<11:50,  3.46it/s]

train:  44%|████▍     | 1546/3515 [07:37<09:29,  3.46it/s]
overall:  39%|███▊      | 1547/4000 [07:40<11:23,  3.59it/s]

train:  44%|████▍     | 1547/3515 [07:37<09:08,  3.59it/s]
overall:  39%|███▊      | 1548/4000 [07:40<10:31,  3.88it/s]

train:  44%|████▍     | 1548/3515 [07:37<08:26,  3.88it/s]
overall:  39%|███▊      | 1549/4000 [07:40<10:47,  3.79it/s]

train:  44%|████▍     | 1549/3515 [07:38<08:39,  3.79it/s]
overall:  39%|███▉      | 1550/4000 [07:40<10:14,  3.99it/s]

train:  44%|████▍     | 1550/3515 [07:38<08:12,  3.99it/s]
overall:  39%|███▉      | 1551/4000 [07:41<11:45,  3.47it/s]

train:  44%|████▍     | 1551/3515 [07:38<09:26,  3.47it/s]
overall:  39%|███▉      | 1552/4000 [07:41<11:35,  3.52it/s]

train:  44%|████▍     | 1552/3515 [07:38<09:17,  3.52it/s]
overall:  39%|███▉      | 1553/4000 [07:41<11:25,  3.57it/s]

train:  44%|████▍     | 1553/3515 [07:39<09:09,  3.57it/s]
overall:  39%|███▉      | 1554/4000 [07:42<11:16,  3.62it/s]

train:  44%|████▍     | 1554/3515 [07:39<09:02,  3.62it/s]
overall:  39%|███▉      | 1555/4000 [07:42<09:19,  4.37it/s]

train:  44%|████▍     | 1555/3515 [07:39<07:28,  4.37it/s]
overall:  39%|███▉      | 1556/4000 [07:42<10:30,  3.88it/s]

train:  44%|████▍     | 1556/3515 [07:39<08:25,  3.88it/s]
overall:  39%|███▉      | 1557/4000 [07:42<08:42,  4.68it/s]

train:  44%|████▍     | 1557/3515 [07:40<06:58,  4.67it/s]
overall:  39%|███▉      | 1558/4000 [07:42<09:20,  4.36it/s]

train:  44%|████▍     | 1558/3515 [07:40<07:28,  4.36it/s]
overall:  39%|███▉      | 1559/4000 [07:43<10:00,  4.07it/s]

train:  44%|████▍     | 1559/3515 [07:40<08:00,  4.07it/s]train loss: 0.5873633623123169
train per: 0.17213625506283103

overall:  39%|███▉      | 1560/4000 [07:43<09:19,  4.36it/s]

train:  44%|████▍     | 1560/3515 [07:40<07:28,  4.36it/s]
overall:  39%|███▉      | 1561/4000 [07:43<09:50,  4.13it/s]

train:  44%|████▍     | 1561/3515 [07:41<07:53,  4.13it/s]
overall:  39%|███▉      | 1562/4000 [07:43<10:16,  3.96it/s]

train:  44%|████▍     | 1562/3515 [07:41<08:13,  3.96it/s]
overall:  39%|███▉      | 1563/4000 [07:44<10:19,  3.93it/s]

train:  44%|████▍     | 1563/3515 [07:41<08:16,  3.93it/s]
overall:  39%|███▉      | 1564/4000 [07:44<10:34,  3.84it/s]

train:  44%|████▍     | 1564/3515 [07:41<08:27,  3.84it/s]
overall:  39%|███▉      | 1565/4000 [07:44<10:32,  3.85it/s]

train:  45%|████▍     | 1565/3515 [07:42<08:26,  3.85it/s]
overall:  39%|███▉      | 1566/4000 [07:45<10:42,  3.79it/s]

train:  45%|████▍     | 1566/3515 [07:42<08:34,  3.79it/s]
overall:  39%|███▉      | 1567/4000 [07:45<10:06,  4.01it/s]

train:  45%|████▍     | 1567/3515 [07:42<08:05,  4.01it/s]
overall:  39%|███▉      | 1568/4000 [07:45<10:12,  3.97it/s]

train:  45%|████▍     | 1568/3515 [07:42<08:10,  3.97it/s]
overall:  39%|███▉      | 1569/4000 [07:45<10:20,  3.92it/s]

train:  45%|████▍     | 1569/3515 [07:43<08:16,  3.92it/s]
overall:  39%|███▉      | 1570/4000 [07:46<12:06,  3.34it/s]

train:  45%|████▍     | 1570/3515 [07:43<09:41,  3.34it/s]
overall:  39%|███▉      | 1571/4000 [07:46<11:26,  3.54it/s]

train:  45%|████▍     | 1571/3515 [07:43<09:09,  3.54it/s]
overall:  39%|███▉      | 1572/4000 [07:46<11:20,  3.57it/s]

train:  45%|████▍     | 1572/3515 [07:44<09:04,  3.57it/s]
overall:  39%|███▉      | 1573/4000 [07:46<11:24,  3.54it/s]

train:  45%|████▍     | 1573/3515 [07:44<09:07,  3.54it/s]
overall:  39%|███▉      | 1574/4000 [07:47<11:02,  3.66it/s]

train:  45%|████▍     | 1574/3515 [07:44<08:50,  3.66it/s]
overall:  39%|███▉      | 1575/4000 [07:47<12:06,  3.34it/s]

train:  45%|████▍     | 1575/3515 [07:44<09:41,  3.34it/s]
overall:  39%|███▉      | 1576/4000 [07:47<11:00,  3.67it/s]

train:  45%|████▍     | 1576/3515 [07:45<08:48,  3.67it/s]
overall:  39%|███▉      | 1577/4000 [07:48<10:46,  3.75it/s]

train:  45%|████▍     | 1577/3515 [07:45<08:37,  3.75it/s]
overall:  39%|███▉      | 1578/4000 [07:48<10:58,  3.68it/s]

train:  45%|████▍     | 1578/3515 [07:45<08:46,  3.68it/s]
overall:  39%|███▉      | 1579/4000 [07:48<10:58,  3.68it/s]

train:  45%|████▍     | 1579/3515 [07:45<08:46,  3.68it/s]train loss: 0.5840966105461121
train per: 0.16790608484168187

overall:  40%|███▉      | 1580/4000 [07:48<11:35,  3.48it/s]

train:  45%|████▍     | 1580/3515 [07:46<09:16,  3.48it/s]
overall:  40%|███▉      | 1581/4000 [07:49<10:52,  3.71it/s]

train:  45%|████▍     | 1581/3515 [07:46<08:41,  3.71it/s]
overall:  40%|███▉      | 1583/4000 [07:49<08:56,  4.51it/s]

train:  45%|████▌     | 1583/3515 [07:46<07:08,  4.51it/s]
overall:  40%|███▉      | 1584/4000 [07:49<08:18,  4.85it/s]

train:  45%|████▌     | 1584/3515 [07:46<06:38,  4.85it/s]
overall:  40%|███▉      | 1585/4000 [07:49<08:40,  4.64it/s]

train:  45%|████▌     | 1585/3515 [07:47<06:55,  4.64it/s]
overall:  40%|███▉      | 1586/4000 [07:50<09:11,  4.37it/s]

train:  45%|████▌     | 1586/3515 [07:47<07:20,  4.38it/s]
overall:  40%|███▉      | 1587/4000 [07:50<08:16,  4.86it/s]

train:  45%|████▌     | 1587/3515 [07:47<06:37,  4.86it/s]
overall:  40%|███▉      | 1588/4000 [07:50<08:46,  4.58it/s]

train:  45%|████▌     | 1588/3515 [07:47<07:00,  4.58it/s]
overall:  40%|███▉      | 1589/4000 [07:50<09:29,  4.24it/s]

train:  45%|████▌     | 1589/3515 [07:48<07:34,  4.24it/s]
overall:  40%|███▉      | 1590/4000 [07:51<09:44,  4.12it/s]

train:  45%|████▌     | 1590/3515 [07:48<07:47,  4.12it/s]
overall:  40%|███▉      | 1591/4000 [07:51<10:09,  3.95it/s]

train:  45%|████▌     | 1591/3515 [07:48<08:07,  3.95it/s]
overall:  40%|███▉      | 1592/4000 [07:51<10:17,  3.90it/s]

train:  45%|████▌     | 1592/3515 [07:48<08:13,  3.90it/s]
overall:  40%|███▉      | 1593/4000 [07:51<10:22,  3.87it/s]

train:  45%|████▌     | 1593/3515 [07:49<08:16,  3.87it/s]
overall:  40%|███▉      | 1594/4000 [07:52<12:06,  3.31it/s]

train:  45%|████▌     | 1594/3515 [07:49<09:39,  3.31it/s]
overall:  40%|███▉      | 1595/4000 [07:52<11:37,  3.45it/s]

train:  45%|████▌     | 1595/3515 [07:49<09:16,  3.45it/s]
overall:  40%|███▉      | 1596/4000 [07:52<11:21,  3.53it/s]

train:  45%|████▌     | 1596/3515 [07:50<09:03,  3.53it/s]
overall:  40%|███▉      | 1597/4000 [07:53<11:17,  3.54it/s]

train:  45%|████▌     | 1597/3515 [07:50<09:01,  3.54it/s]
overall:  40%|███▉      | 1598/4000 [07:53<09:27,  4.23it/s]

train:  45%|████▌     | 1598/3515 [07:50<07:32,  4.23it/s]
overall:  40%|███▉      | 1599/4000 [07:53<10:44,  3.72it/s]

train:  45%|████▌     | 1599/3515 [07:50<08:34,  3.72it/s]train loss: 0.60442054271698
train per: 0.17405459554219885

overall:  40%|████      | 1600/4000 [07:53<10:44,  3.72it/s]

train:  46%|████▌     | 1600/3515 [07:51<08:34,  3.72it/s]
overall:  40%|████      | 1601/4000 [07:54<10:43,  3.73it/s]

train:  46%|████▌     | 1601/3515 [07:51<08:33,  3.73it/s]
overall:  40%|████      | 1602/4000 [07:54<10:29,  3.81it/s]

train:  46%|████▌     | 1602/3515 [07:51<08:22,  3.81it/s]
overall:  40%|████      | 1603/4000 [07:54<10:25,  3.83it/s]

train:  46%|████▌     | 1603/3515 [07:51<08:19,  3.83it/s]
overall:  40%|████      | 1604/4000 [07:54<11:15,  3.55it/s]

train:  46%|████▌     | 1604/3515 [07:52<08:58,  3.55it/s]
overall:  40%|████      | 1605/4000 [07:55<10:55,  3.66it/s]

train:  46%|████▌     | 1605/3515 [07:52<08:42,  3.66it/s]
overall:  40%|████      | 1606/4000 [07:55<10:03,  3.96it/s]

train:  46%|████▌     | 1606/3515 [07:52<08:01,  3.96it/s]
overall:  40%|████      | 1607/4000 [07:55<09:18,  4.29it/s]

train:  46%|████▌     | 1607/3515 [07:52<07:25,  4.29it/s]
overall:  40%|████      | 1608/4000 [07:55<09:31,  4.18it/s]

train:  46%|████▌     | 1608/3515 [07:53<07:35,  4.18it/s]
overall:  40%|████      | 1609/4000 [07:56<09:34,  4.16it/s]

train:  46%|████▌     | 1609/3515 [07:53<07:37,  4.16it/s]
overall:  40%|████      | 1610/4000 [07:56<09:47,  4.07it/s]

train:  46%|████▌     | 1610/3515 [07:53<07:48,  4.07it/s]
overall:  40%|████      | 1611/4000 [07:56<10:10,  3.91it/s]

train:  46%|████▌     | 1611/3515 [07:53<08:06,  3.91it/s]
overall:  40%|████      | 1612/4000 [07:56<10:02,  3.96it/s]

train:  46%|████▌     | 1612/3515 [07:54<08:00,  3.96it/s]
overall:  40%|████      | 1613/4000 [07:57<09:08,  4.35it/s]

train:  46%|████▌     | 1613/3515 [07:54<07:16,  4.35it/s]
overall:  40%|████      | 1614/4000 [07:57<09:39,  4.12it/s]

train:  46%|████▌     | 1614/3515 [07:54<07:41,  4.12it/s]
overall:  40%|████      | 1615/4000 [07:57<09:55,  4.01it/s]

train:  46%|████▌     | 1615/3515 [07:54<07:54,  4.01it/s]
overall:  40%|████      | 1616/4000 [07:57<10:17,  3.86it/s]

train:  46%|████▌     | 1616/3515 [07:55<08:11,  3.86it/s]
overall:  40%|████      | 1617/4000 [07:58<10:14,  3.88it/s]

train:  46%|████▌     | 1617/3515 [07:55<08:09,  3.88it/s]
overall:  40%|████      | 1618/4000 [07:58<12:01,  3.30it/s]

train:  46%|████▌     | 1618/3515 [07:55<09:34,  3.30it/s]
overall:  40%|████      | 1619/4000 [07:58<11:34,  3.43it/s]

train:  46%|████▌     | 1619/3515 [07:56<09:13,  3.43it/s]train loss: 0.5687097311019897
train per: 0.16519065190651908

overall:  40%|████      | 1620/4000 [07:59<11:12,  3.54it/s]

train:  46%|████▌     | 1620/3515 [07:56<08:56,  3.53it/s]
overall:  41%|████      | 1621/4000 [07:59<11:05,  3.57it/s]

train:  46%|████▌     | 1621/3515 [07:56<08:50,  3.57it/s]
overall:  41%|████      | 1622/4000 [07:59<10:55,  3.63it/s]

train:  46%|████▌     | 1622/3515 [07:56<08:41,  3.63it/s]
overall:  41%|████      | 1623/4000 [07:59<11:17,  3.51it/s]

train:  46%|████▌     | 1623/3515 [07:57<08:58,  3.51it/s]
overall:  41%|████      | 1624/4000 [08:00<10:58,  3.61it/s]

train:  46%|████▌     | 1624/3515 [07:57<08:43,  3.61it/s]
overall:  41%|████      | 1625/4000 [08:00<10:47,  3.67it/s]

train:  46%|████▌     | 1625/3515 [07:57<08:34,  3.67it/s]
overall:  41%|████      | 1626/4000 [08:00<10:47,  3.67it/s]

train:  46%|████▋     | 1626/3515 [07:58<08:35,  3.67it/s]
overall:  41%|████      | 1627/4000 [08:00<10:46,  3.67it/s]

train:  46%|████▋     | 1627/3515 [07:58<08:34,  3.67it/s]
overall:  41%|████      | 1628/4000 [08:01<11:14,  3.52it/s]

train:  46%|████▋     | 1628/3515 [07:58<08:56,  3.52it/s]
overall:  41%|████      | 1629/4000 [08:01<11:01,  3.58it/s]

train:  46%|████▋     | 1629/3515 [07:58<08:46,  3.58it/s]
overall:  41%|████      | 1630/4000 [08:01<10:48,  3.65it/s]

train:  46%|████▋     | 1630/3515 [07:59<08:36,  3.65it/s]
overall:  41%|████      | 1631/4000 [08:02<10:56,  3.61it/s]

train:  46%|████▋     | 1631/3515 [07:59<08:42,  3.61it/s]
overall:  41%|████      | 1632/4000 [08:02<10:28,  3.77it/s]

train:  46%|████▋     | 1632/3515 [07:59<08:20,  3.77it/s]
overall:  41%|████      | 1633/4000 [08:02<10:31,  3.75it/s]

train:  46%|████▋     | 1633/3515 [07:59<08:21,  3.75it/s]
overall:  41%|████      | 1634/4000 [08:02<10:24,  3.79it/s]

train:  46%|████▋     | 1634/3515 [08:00<08:16,  3.79it/s]
overall:  41%|████      | 1635/4000 [08:03<10:31,  3.74it/s]

train:  47%|████▋     | 1635/3515 [08:00<08:22,  3.74it/s]
overall:  41%|████      | 1637/4000 [08:03<08:59,  4.38it/s]

train:  47%|████▋     | 1637/3515 [08:00<07:08,  4.38it/s]
overall:  41%|████      | 1638/4000 [08:03<09:24,  4.18it/s]

train:  47%|████▋     | 1638/3515 [08:01<07:28,  4.18it/s]
overall:  41%|████      | 1639/4000 [08:03<09:02,  4.35it/s]

train:  47%|████▋     | 1639/3515 [08:01<07:10,  4.35it/s]train loss: 0.5537768602371216
train per: 0.15909836449693052

overall:  41%|████      | 1640/4000 [08:04<09:27,  4.16it/s]

train:  47%|████▋     | 1640/3515 [08:01<07:31,  4.15it/s]
overall:  41%|████      | 1641/4000 [08:04<09:19,  4.22it/s]

train:  47%|████▋     | 1641/3515 [08:01<07:24,  4.22it/s]
overall:  41%|████      | 1642/4000 [08:04<11:21,  3.46it/s]

train:  47%|████▋     | 1642/3515 [08:02<09:01,  3.46it/s]
overall:  41%|████      | 1643/4000 [08:05<11:03,  3.55it/s]

train:  47%|████▋     | 1643/3515 [08:02<08:46,  3.55it/s]
overall:  41%|████      | 1644/4000 [08:05<10:55,  3.59it/s]

train:  47%|████▋     | 1644/3515 [08:02<08:40,  3.60it/s]
overall:  41%|████      | 1645/4000 [08:05<10:57,  3.58it/s]

train:  47%|████▋     | 1645/3515 [08:03<08:42,  3.58it/s]
overall:  41%|████      | 1646/4000 [08:05<10:38,  3.69it/s]

train:  47%|████▋     | 1646/3515 [08:03<08:26,  3.69it/s]
overall:  41%|████      | 1647/4000 [08:06<11:26,  3.43it/s]

train:  47%|████▋     | 1647/3515 [08:03<09:04,  3.43it/s]
overall:  41%|████      | 1648/4000 [08:06<11:04,  3.54it/s]

train:  47%|████▋     | 1648/3515 [08:03<08:47,  3.54it/s]
overall:  41%|████      | 1649/4000 [08:06<10:53,  3.60it/s]

train:  47%|████▋     | 1649/3515 [08:04<08:38,  3.60it/s]
overall:  41%|████▏     | 1650/4000 [08:07<11:02,  3.55it/s]

train:  47%|████▋     | 1650/3515 [08:04<08:45,  3.55it/s]
overall:  41%|████▏     | 1651/4000 [08:07<10:56,  3.58it/s]

train:  47%|████▋     | 1651/3515 [08:04<08:41,  3.58it/s]
overall:  41%|████▏     | 1652/4000 [08:07<11:08,  3.51it/s]

train:  47%|████▋     | 1652/3515 [08:05<08:50,  3.51it/s]
overall:  41%|████▏     | 1653/4000 [08:07<10:49,  3.62it/s]

train:  47%|████▋     | 1653/3515 [08:05<08:34,  3.62it/s]
overall:  41%|████▏     | 1654/4000 [08:08<10:24,  3.75it/s]

train:  47%|████▋     | 1654/3515 [08:05<08:15,  3.75it/s]
overall:  41%|████▏     | 1655/4000 [08:08<10:42,  3.65it/s]

train:  47%|████▋     | 1655/3515 [08:05<08:29,  3.65it/s]
overall:  41%|████▏     | 1656/4000 [08:08<10:39,  3.67it/s]

train:  47%|████▋     | 1656/3515 [08:06<08:27,  3.67it/s]
overall:  41%|████▏     | 1657/4000 [08:08<10:17,  3.79it/s]

train:  47%|████▋     | 1657/3515 [08:06<08:09,  3.79it/s]
overall:  41%|████▏     | 1658/4000 [08:09<10:18,  3.79it/s]

train:  47%|████▋     | 1658/3515 [08:06<08:10,  3.79it/s]
overall:  41%|████▏     | 1659/4000 [08:09<09:45,  4.00it/s]

train:  47%|████▋     | 1659/3515 [08:06<07:43,  4.00it/s]train loss: 0.5589413642883301
train per: 0.16123938762900752

overall:  42%|████▏     | 1660/4000 [08:09<10:08,  3.85it/s]

train:  47%|████▋     | 1660/3515 [08:07<08:02,  3.84it/s]
overall:  42%|████▏     | 1661/4000 [08:10<10:15,  3.80it/s]

train:  47%|████▋     | 1661/3515 [08:07<08:08,  3.80it/s]
overall:  42%|████▏     | 1662/4000 [08:10<10:16,  3.79it/s]

train:  47%|████▋     | 1662/3515 [08:07<08:08,  3.79it/s]
overall:  42%|████▏     | 1663/4000 [08:10<10:17,  3.78it/s]

train:  47%|████▋     | 1663/3515 [08:07<08:09,  3.78it/s]
overall:  42%|████▏     | 1664/4000 [08:10<10:26,  3.73it/s]

train:  47%|████▋     | 1664/3515 [08:08<08:16,  3.73it/s]
overall:  42%|████▏     | 1665/4000 [08:11<10:23,  3.75it/s]

train:  47%|████▋     | 1665/3515 [08:08<08:13,  3.75it/s]
overall:  42%|████▏     | 1666/4000 [08:11<12:00,  3.24it/s]

train:  47%|████▋     | 1666/3515 [08:08<09:31,  3.24it/s]
overall:  42%|████▏     | 1667/4000 [08:11<11:10,  3.48it/s]

train:  47%|████▋     | 1667/3515 [08:09<08:50,  3.48it/s]
overall:  42%|████▏     | 1668/4000 [08:12<11:01,  3.53it/s]

train:  47%|████▋     | 1668/3515 [08:09<08:43,  3.53it/s]
overall:  42%|████▏     | 1669/4000 [08:12<10:51,  3.58it/s]

train:  47%|████▋     | 1669/3515 [08:09<08:35,  3.58it/s]
overall:  42%|████▏     | 1670/4000 [08:12<10:23,  3.74it/s]

train:  48%|████▊     | 1670/3515 [08:09<08:13,  3.74it/s]
overall:  42%|████▏     | 1671/4000 [08:12<11:31,  3.37it/s]

train:  48%|████▊     | 1671/3515 [08:10<09:07,  3.37it/s]
overall:  42%|████▏     | 1672/4000 [08:12<09:17,  4.17it/s]

train:  48%|████▊     | 1672/3515 [08:10<07:21,  4.17it/s]
overall:  42%|████▏     | 1673/4000 [08:13<09:31,  4.07it/s]

train:  48%|████▊     | 1673/3515 [08:10<07:32,  4.07it/s]
overall:  42%|████▏     | 1674/4000 [08:13<09:40,  4.01it/s]

train:  48%|████▊     | 1674/3515 [08:10<07:39,  4.01it/s]
overall:  42%|████▏     | 1675/4000 [08:13<09:36,  4.03it/s]

train:  48%|████▊     | 1675/3515 [08:11<07:36,  4.03it/s]
overall:  42%|████▏     | 1676/4000 [08:14<10:12,  3.80it/s]

train:  48%|████▊     | 1676/3515 [08:11<08:04,  3.79it/s]
overall:  42%|████▏     | 1677/4000 [08:14<10:09,  3.81it/s]

train:  48%|████▊     | 1677/3515 [08:11<08:01,  3.81it/s]
overall:  42%|████▏     | 1678/4000 [08:14<10:08,  3.82it/s]

train:  48%|████▊     | 1678/3515 [08:11<08:01,  3.82it/s]
overall:  42%|████▏     | 1679/4000 [08:14<10:15,  3.77it/s]

train:  48%|████▊     | 1679/3515 [08:12<08:07,  3.77it/s]train loss: 0.562867283821106
train per: 0.1627646356615587

overall:  42%|████▏     | 1681/4000 [08:14<06:39,  5.81it/s]

train:  48%|████▊     | 1681/3515 [08:12<05:15,  5.81it/s]
overall:  42%|████▏     | 1682/4000 [08:15<06:21,  6.08it/s]

train:  48%|████▊     | 1682/3515 [08:12<05:01,  6.08it/s]
overall:  42%|████▏     | 1683/4000 [08:15<07:18,  5.28it/s]

train:  48%|████▊     | 1683/3515 [08:12<05:46,  5.28it/s]
overall:  42%|████▏     | 1684/4000 [08:15<08:01,  4.81it/s]

train:  48%|████▊     | 1684/3515 [08:12<06:20,  4.81it/s]
overall:  42%|████▏     | 1685/4000 [08:15<08:49,  4.37it/s]

train:  48%|████▊     | 1685/3515 [08:13<06:58,  4.37it/s]
overall:  42%|████▏     | 1686/4000 [08:16<08:58,  4.30it/s]

train:  48%|████▊     | 1686/3515 [08:13<07:05,  4.30it/s]
overall:  42%|████▏     | 1687/4000 [08:16<09:17,  4.15it/s]

train:  48%|████▊     | 1687/3515 [08:13<07:20,  4.15it/s]
overall:  42%|████▏     | 1688/4000 [08:16<08:09,  4.72it/s]

train:  48%|████▊     | 1688/3515 [08:13<06:27,  4.72it/s]
overall:  42%|████▏     | 1689/4000 [08:16<08:48,  4.37it/s]

train:  48%|████▊     | 1689/3515 [08:14<06:57,  4.37it/s]
overall:  42%|████▏     | 1690/4000 [08:17<10:57,  3.51it/s]

train:  48%|████▊     | 1690/3515 [08:14<08:39,  3.51it/s]
overall:  42%|████▏     | 1691/4000 [08:17<10:41,  3.60it/s]

train:  48%|████▊     | 1691/3515 [08:14<08:27,  3.60it/s]
overall:  42%|████▏     | 1692/4000 [08:17<10:34,  3.64it/s]

train:  48%|████▊     | 1692/3515 [08:15<08:21,  3.64it/s]
overall:  42%|████▏     | 1693/4000 [08:18<10:38,  3.61it/s]

train:  48%|████▊     | 1693/3515 [08:15<08:24,  3.61it/s]
overall:  42%|████▏     | 1694/4000 [08:18<10:31,  3.65it/s]

train:  48%|████▊     | 1694/3515 [08:15<08:18,  3.65it/s]
overall:  42%|████▏     | 1695/4000 [08:18<11:28,  3.35it/s]

train:  48%|████▊     | 1695/3515 [08:16<09:03,  3.35it/s]
overall:  42%|████▏     | 1696/4000 [08:18<10:37,  3.61it/s]

train:  48%|████▊     | 1696/3515 [08:16<08:23,  3.61it/s]
overall:  42%|████▏     | 1697/4000 [08:19<09:43,  3.95it/s]

train:  48%|████▊     | 1697/3515 [08:16<07:40,  3.95it/s]
overall:  42%|████▏     | 1698/4000 [08:19<09:58,  3.85it/s]

train:  48%|████▊     | 1698/3515 [08:16<07:52,  3.85it/s]
overall:  42%|████▏     | 1699/4000 [08:19<10:06,  3.79it/s]

train:  48%|████▊     | 1699/3515 [08:17<07:58,  3.79it/s]train loss: 0.5780056118965149
train per: 0.17117433262313572

overall:  42%|████▎     | 1700/4000 [08:19<10:42,  3.58it/s]

train:  48%|████▊     | 1700/3515 [08:17<08:27,  3.58it/s]
overall:  43%|████▎     | 1701/4000 [08:20<10:34,  3.62it/s]

train:  48%|████▊     | 1701/3515 [08:17<08:20,  3.62it/s]
overall:  43%|████▎     | 1702/4000 [08:20<09:03,  4.23it/s]

train:  48%|████▊     | 1702/3515 [08:17<07:08,  4.23it/s]
overall:  43%|████▎     | 1703/4000 [08:20<09:38,  3.97it/s]

train:  48%|████▊     | 1703/3515 [08:18<07:36,  3.97it/s]
overall:  43%|████▎     | 1705/4000 [08:20<08:07,  4.71it/s]

train:  49%|████▊     | 1705/3515 [08:18<06:24,  4.71it/s]
overall:  43%|████▎     | 1706/4000 [08:21<08:32,  4.47it/s]

train:  49%|████▊     | 1706/3515 [08:18<06:44,  4.47it/s]
overall:  43%|████▎     | 1707/4000 [08:21<07:49,  4.88it/s]

train:  49%|████▊     | 1707/3515 [08:18<06:10,  4.88it/s]
overall:  43%|████▎     | 1708/4000 [08:21<08:34,  4.45it/s]

train:  49%|████▊     | 1708/3515 [08:19<06:45,  4.45it/s]
overall:  43%|████▎     | 1709/4000 [08:21<08:15,  4.62it/s]

train:  49%|████▊     | 1709/3515 [08:19<06:30,  4.62it/s]
overall:  43%|████▎     | 1710/4000 [08:22<08:43,  4.38it/s]

train:  49%|████▊     | 1710/3515 [08:19<06:52,  4.38it/s]
overall:  43%|████▎     | 1711/4000 [08:22<08:42,  4.38it/s]

train:  49%|████▊     | 1711/3515 [08:19<06:51,  4.38it/s]
overall:  43%|████▎     | 1712/4000 [08:22<08:54,  4.28it/s]

train:  49%|████▊     | 1712/3515 [08:19<07:01,  4.28it/s]
overall:  43%|████▎     | 1713/4000 [08:22<09:17,  4.10it/s]

train:  49%|████▊     | 1713/3515 [08:20<07:19,  4.10it/s]
overall:  43%|████▎     | 1714/4000 [08:23<11:06,  3.43it/s]

train:  49%|████▉     | 1714/3515 [08:20<08:44,  3.43it/s]
overall:  43%|████▎     | 1715/4000 [08:23<10:56,  3.48it/s]

train:  49%|████▉     | 1715/3515 [08:20<08:37,  3.48it/s]
overall:  43%|████▎     | 1716/4000 [08:23<10:34,  3.60it/s]

train:  49%|████▉     | 1716/3515 [08:21<08:19,  3.60it/s]
overall:  43%|████▎     | 1717/4000 [08:24<10:32,  3.61it/s]

train:  49%|████▉     | 1717/3515 [08:21<08:18,  3.61it/s]
overall:  43%|████▎     | 1718/4000 [08:24<09:58,  3.81it/s]

train:  49%|████▉     | 1718/3515 [08:21<07:51,  3.81it/s]
overall:  43%|████▎     | 1719/4000 [08:24<10:44,  3.54it/s]

train:  49%|████▉     | 1719/3515 [08:22<08:27,  3.54it/s]train loss: 0.5828923583030701
train per: 0.1716964778189268

overall:  43%|████▎     | 1720/4000 [08:24<10:35,  3.59it/s]

train:  49%|████▉     | 1720/3515 [08:22<08:21,  3.58it/s]
overall:  43%|████▎     | 1721/4000 [08:25<10:22,  3.66it/s]

train:  49%|████▉     | 1721/3515 [08:22<08:09,  3.66it/s]
overall:  43%|████▎     | 1723/4000 [08:25<08:18,  4.57it/s]

train:  49%|████▉     | 1723/3515 [08:22<06:32,  4.57it/s]
overall:  43%|████▎     | 1724/4000 [08:25<09:18,  4.07it/s]

train:  49%|████▉     | 1724/3515 [08:23<07:19,  4.07it/s]
overall:  43%|████▎     | 1725/4000 [08:26<09:35,  3.95it/s]

train:  49%|████▉     | 1725/3515 [08:23<07:33,  3.95it/s]
overall:  43%|████▎     | 1726/4000 [08:26<09:39,  3.92it/s]

train:  49%|████▉     | 1726/3515 [08:23<07:35,  3.92it/s]
overall:  43%|████▎     | 1727/4000 [08:26<09:55,  3.81it/s]

train:  49%|████▉     | 1727/3515 [08:24<07:48,  3.81it/s]
overall:  43%|████▎     | 1728/4000 [08:26<09:25,  4.02it/s]

train:  49%|████▉     | 1728/3515 [08:24<07:24,  4.02it/s]
overall:  43%|████▎     | 1729/4000 [08:27<09:19,  4.06it/s]

train:  49%|████▉     | 1729/3515 [08:24<07:19,  4.06it/s]
overall:  43%|████▎     | 1730/4000 [08:27<09:32,  3.96it/s]

train:  49%|████▉     | 1730/3515 [08:24<07:30,  3.96it/s]
overall:  43%|████▎     | 1731/4000 [08:27<09:30,  3.98it/s]

train:  49%|████▉     | 1731/3515 [08:24<07:28,  3.98it/s]
overall:  43%|████▎     | 1732/4000 [08:27<08:00,  4.72it/s]

train:  49%|████▉     | 1732/3515 [08:25<06:17,  4.72it/s]
overall:  43%|████▎     | 1733/4000 [08:27<08:26,  4.48it/s]

train:  49%|████▉     | 1733/3515 [08:25<06:38,  4.48it/s]
overall:  43%|████▎     | 1734/4000 [08:28<08:56,  4.22it/s]

train:  49%|████▉     | 1734/3515 [08:25<07:02,  4.22it/s]
overall:  43%|████▎     | 1735/4000 [08:28<08:59,  4.20it/s]

train:  49%|████▉     | 1735/3515 [08:25<07:03,  4.20it/s]
overall:  43%|████▎     | 1736/4000 [08:28<09:19,  4.05it/s]

train:  49%|████▉     | 1736/3515 [08:26<07:19,  4.05it/s]
overall:  43%|████▎     | 1737/4000 [08:29<09:25,  4.00it/s]

train:  49%|████▉     | 1737/3515 [08:26<07:24,  4.00it/s]
overall:  43%|████▎     | 1738/4000 [08:29<11:00,  3.42it/s]

train:  49%|████▉     | 1738/3515 [08:26<08:39,  3.42it/s]
overall:  43%|████▎     | 1739/4000 [08:29<10:37,  3.55it/s]

train:  49%|████▉     | 1739/3515 [08:27<08:20,  3.55it/s]train loss: 0.5897344350814819
train per: 0.16193805381676288

overall:  44%|████▎     | 1740/4000 [08:29<10:35,  3.56it/s]

train:  50%|████▉     | 1740/3515 [08:27<08:19,  3.55it/s]
overall:  44%|████▎     | 1741/4000 [08:30<10:02,  3.75it/s]

train:  50%|████▉     | 1741/3515 [08:27<07:52,  3.75it/s]
overall:  44%|████▎     | 1742/4000 [08:30<09:26,  3.98it/s]

train:  50%|████▉     | 1742/3515 [08:27<07:25,  3.98it/s]
overall:  44%|████▎     | 1743/4000 [08:30<10:17,  3.66it/s]

train:  50%|████▉     | 1743/3515 [08:28<08:04,  3.66it/s]
overall:  44%|████▎     | 1744/4000 [08:30<10:05,  3.73it/s]

train:  50%|████▉     | 1744/3515 [08:28<07:55,  3.73it/s]
overall:  44%|████▎     | 1745/4000 [08:31<09:58,  3.77it/s]

train:  50%|████▉     | 1745/3515 [08:28<07:50,  3.77it/s]
overall:  44%|████▎     | 1746/4000 [08:31<10:10,  3.70it/s]

train:  50%|████▉     | 1746/3515 [08:28<07:58,  3.70it/s]
overall:  44%|████▎     | 1747/4000 [08:31<09:27,  3.97it/s]

train:  50%|████▉     | 1747/3515 [08:29<07:25,  3.97it/s]
overall:  44%|████▎     | 1748/4000 [08:32<10:25,  3.60it/s]

train:  50%|████▉     | 1748/3515 [08:29<08:10,  3.60it/s]
overall:  44%|████▎     | 1749/4000 [08:32<10:11,  3.68it/s]

train:  50%|████▉     | 1749/3515 [08:29<08:00,  3.68it/s]
overall:  44%|████▍     | 1751/4000 [08:32<08:32,  4.39it/s]

train:  50%|████▉     | 1751/3515 [08:30<06:42,  4.39it/s]
overall:  44%|████▍     | 1752/4000 [08:32<08:47,  4.26it/s]

train:  50%|████▉     | 1752/3515 [08:30<06:53,  4.26it/s]
overall:  44%|████▍     | 1753/4000 [08:33<08:41,  4.31it/s]

train:  50%|████▉     | 1753/3515 [08:30<06:49,  4.31it/s]
overall:  44%|████▍     | 1754/4000 [08:33<08:50,  4.23it/s]

train:  50%|████▉     | 1754/3515 [08:30<06:55,  4.23it/s]
overall:  44%|████▍     | 1755/4000 [08:33<09:03,  4.13it/s]

train:  50%|████▉     | 1755/3515 [08:31<07:05,  4.13it/s]
overall:  44%|████▍     | 1756/4000 [08:33<07:58,  4.69it/s]

train:  50%|████▉     | 1756/3515 [08:31<06:14,  4.69it/s]
overall:  44%|████▍     | 1757/4000 [08:33<07:53,  4.74it/s]

train:  50%|████▉     | 1757/3515 [08:31<06:10,  4.74it/s]
overall:  44%|████▍     | 1758/4000 [08:34<08:32,  4.37it/s]

train:  50%|█████     | 1758/3515 [08:31<06:41,  4.37it/s]
overall:  44%|████▍     | 1759/4000 [08:34<08:50,  4.23it/s]

train:  50%|█████     | 1759/3515 [08:31<06:55,  4.23it/s]train loss: 0.5739034414291382
train per: 0.16257910318042315

overall:  44%|████▍     | 1760/4000 [08:34<09:20,  4.00it/s]

train:  50%|█████     | 1760/3515 [08:32<07:19,  4.00it/s]
overall:  44%|████▍     | 1761/4000 [08:35<09:23,  3.97it/s]

train:  50%|█████     | 1761/3515 [08:32<07:21,  3.97it/s]
overall:  44%|████▍     | 1762/4000 [08:35<10:32,  3.54it/s]

train:  50%|█████     | 1762/3515 [08:32<08:15,  3.54it/s]
overall:  44%|████▍     | 1763/4000 [08:35<10:16,  3.63it/s]

train:  50%|█████     | 1763/3515 [08:33<08:02,  3.63it/s]
overall:  44%|████▍     | 1764/4000 [08:35<10:04,  3.70it/s]

train:  50%|█████     | 1764/3515 [08:33<07:53,  3.70it/s]
overall:  44%|████▍     | 1765/4000 [08:36<10:16,  3.63it/s]

train:  50%|█████     | 1765/3515 [08:33<08:02,  3.63it/s]
overall:  44%|████▍     | 1766/4000 [08:36<10:05,  3.69it/s]

train:  50%|█████     | 1766/3515 [08:33<07:53,  3.69it/s]
overall:  44%|████▍     | 1767/4000 [08:36<10:35,  3.51it/s]

train:  50%|█████     | 1767/3515 [08:34<08:17,  3.51it/s]
overall:  44%|████▍     | 1768/4000 [08:37<10:26,  3.56it/s]

train:  50%|█████     | 1768/3515 [08:34<08:10,  3.56it/s]
overall:  44%|████▍     | 1769/4000 [08:37<08:43,  4.27it/s]

train:  50%|█████     | 1769/3515 [08:34<06:49,  4.27it/s]
overall:  44%|████▍     | 1770/4000 [08:37<09:19,  3.98it/s]

train:  50%|█████     | 1770/3515 [08:34<07:17,  3.98it/s]
overall:  44%|████▍     | 1771/4000 [08:37<08:32,  4.35it/s]

train:  50%|█████     | 1771/3515 [08:35<06:40,  4.35it/s]
overall:  44%|████▍     | 1772/4000 [08:38<09:49,  3.78it/s]

train:  50%|█████     | 1772/3515 [08:35<07:41,  3.78it/s]
overall:  44%|████▍     | 1773/4000 [08:38<09:47,  3.79it/s]

train:  50%|█████     | 1773/3515 [08:35<07:39,  3.79it/s]
overall:  44%|████▍     | 1774/4000 [08:38<09:42,  3.82it/s]

train:  50%|█████     | 1774/3515 [08:35<07:35,  3.82it/s]
overall:  44%|████▍     | 1775/4000 [08:38<08:34,  4.32it/s]

train:  50%|█████     | 1775/3515 [08:36<06:42,  4.32it/s]
overall:  44%|████▍     | 1776/4000 [08:38<08:53,  4.17it/s]

train:  51%|█████     | 1776/3515 [08:36<06:56,  4.17it/s]
overall:  44%|████▍     | 1777/4000 [08:39<09:08,  4.05it/s]

train:  51%|█████     | 1777/3515 [08:36<07:09,  4.05it/s]
overall:  44%|████▍     | 1778/4000 [08:39<09:21,  3.96it/s]

train:  51%|█████     | 1778/3515 [08:36<07:18,  3.96it/s]
overall:  44%|████▍     | 1779/4000 [08:39<09:38,  3.84it/s]

train:  51%|█████     | 1779/3515 [08:37<07:32,  3.84it/s]train loss: 0.5866485238075256
train per: 0.16887368678231127

overall:  44%|████▍     | 1780/4000 [08:40<09:38,  3.84it/s]

train:  51%|█████     | 1780/3515 [08:37<07:32,  3.84it/s]
overall:  45%|████▍     | 1781/4000 [08:40<09:19,  3.97it/s]

train:  51%|█████     | 1781/3515 [08:37<07:16,  3.97it/s]
overall:  45%|████▍     | 1782/4000 [08:40<09:30,  3.89it/s]

train:  51%|█████     | 1782/3515 [08:37<07:25,  3.89it/s]
overall:  45%|████▍     | 1783/4000 [08:40<09:34,  3.86it/s]

train:  51%|█████     | 1783/3515 [08:38<07:28,  3.86it/s]
overall:  45%|████▍     | 1784/4000 [08:41<09:45,  3.79it/s]

train:  51%|█████     | 1784/3515 [08:38<07:37,  3.79it/s]
overall:  45%|████▍     | 1785/4000 [08:41<09:46,  3.78it/s]

train:  51%|█████     | 1785/3515 [08:38<07:37,  3.78it/s]
overall:  45%|████▍     | 1786/4000 [08:41<11:39,  3.17it/s]

train:  51%|█████     | 1786/3515 [08:39<09:05,  3.17it/s]
overall:  45%|████▍     | 1787/4000 [08:42<11:08,  3.31it/s]

train:  51%|█████     | 1787/3515 [08:39<08:42,  3.31it/s]
overall:  45%|████▍     | 1788/4000 [08:42<10:37,  3.47it/s]

train:  51%|█████     | 1788/3515 [08:39<08:17,  3.47it/s]
overall:  45%|████▍     | 1789/4000 [08:42<10:34,  3.48it/s]

train:  51%|█████     | 1789/3515 [08:39<08:15,  3.48it/s]
overall:  45%|████▍     | 1790/4000 [08:42<10:11,  3.61it/s]

train:  51%|█████     | 1790/3515 [08:40<07:57,  3.61it/s]
overall:  45%|████▍     | 1791/4000 [08:43<11:07,  3.31it/s]

train:  51%|█████     | 1791/3515 [08:40<08:41,  3.31it/s]
overall:  45%|████▍     | 1792/4000 [08:43<10:52,  3.38it/s]

train:  51%|█████     | 1792/3515 [08:40<08:29,  3.38it/s]
overall:  45%|████▍     | 1793/4000 [08:43<10:28,  3.51it/s]

train:  51%|█████     | 1793/3515 [08:41<08:10,  3.51it/s]
overall:  45%|████▍     | 1794/4000 [08:44<10:25,  3.53it/s]

train:  51%|█████     | 1794/3515 [08:41<08:08,  3.53it/s]
overall:  45%|████▍     | 1795/4000 [08:44<10:08,  3.62it/s]

train:  51%|█████     | 1795/3515 [08:41<07:54,  3.62it/s]
overall:  45%|████▍     | 1796/4000 [08:44<10:18,  3.56it/s]

train:  51%|█████     | 1796/3515 [08:41<08:02,  3.56it/s]
overall:  45%|████▍     | 1797/4000 [08:44<09:51,  3.72it/s]

train:  51%|█████     | 1797/3515 [08:42<07:41,  3.72it/s]
overall:  45%|████▍     | 1798/4000 [08:45<09:20,  3.93it/s]

train:  51%|█████     | 1798/3515 [08:42<07:17,  3.93it/s]
overall:  45%|████▍     | 1799/4000 [08:45<09:34,  3.83it/s]

train:  51%|█████     | 1799/3515 [08:42<07:27,  3.83it/s]train loss: 0.5640689134597778
train per: 0.16380443051105323

overall:  45%|████▌     | 1800/4000 [08:45<09:35,  3.82it/s]

train:  51%|█████     | 1800/3515 [08:42<07:29,  3.82it/s]
overall:  45%|████▌     | 1801/4000 [08:45<09:46,  3.75it/s]

train:  51%|█████     | 1801/3515 [08:43<07:36,  3.75it/s]
overall:  45%|████▌     | 1802/4000 [08:46<09:53,  3.70it/s]

train:  51%|█████▏    | 1802/3515 [08:43<07:42,  3.71it/s]
overall:  45%|████▌     | 1803/4000 [08:46<09:52,  3.71it/s]

train:  51%|█████▏    | 1803/3515 [08:43<07:41,  3.71it/s]
overall:  45%|████▌     | 1804/4000 [08:46<09:48,  3.73it/s]

train:  51%|█████▏    | 1804/3515 [08:44<07:38,  3.73it/s]
overall:  45%|████▌     | 1805/4000 [08:46<09:48,  3.73it/s]

train:  51%|█████▏    | 1805/3515 [08:44<07:38,  3.73it/s]
overall:  45%|████▌     | 1806/4000 [08:47<09:53,  3.70it/s]

train:  51%|█████▏    | 1806/3515 [08:44<07:42,  3.70it/s]
overall:  45%|████▌     | 1807/4000 [08:47<09:55,  3.68it/s]

train:  51%|█████▏    | 1807/3515 [08:44<07:43,  3.68it/s]
overall:  45%|████▌     | 1808/4000 [08:47<10:00,  3.65it/s]

train:  51%|█████▏    | 1808/3515 [08:45<07:47,  3.65it/s]
overall:  45%|████▌     | 1809/4000 [08:48<09:52,  3.70it/s]

train:  51%|█████▏    | 1809/3515 [08:45<07:41,  3.70it/s]
overall:  45%|████▌     | 1810/4000 [08:48<11:29,  3.18it/s]

train:  51%|█████▏    | 1810/3515 [08:45<08:56,  3.18it/s]
overall:  45%|████▌     | 1811/4000 [08:48<10:54,  3.34it/s]

train:  52%|█████▏    | 1811/3515 [08:46<08:29,  3.34it/s]
overall:  45%|████▌     | 1812/4000 [08:48<10:26,  3.49it/s]

train:  52%|█████▏    | 1812/3515 [08:46<08:07,  3.49it/s]
overall:  45%|████▌     | 1813/4000 [08:49<10:18,  3.54it/s]

train:  52%|█████▏    | 1813/3515 [08:46<08:01,  3.54it/s]
overall:  45%|████▌     | 1814/4000 [08:49<10:09,  3.59it/s]

train:  52%|█████▏    | 1814/3515 [08:46<07:54,  3.59it/s]
overall:  45%|████▌     | 1815/4000 [08:49<10:59,  3.32it/s]

train:  52%|█████▏    | 1815/3515 [08:47<08:32,  3.32it/s]
overall:  45%|████▌     | 1816/4000 [08:50<10:25,  3.49it/s]

train:  52%|█████▏    | 1816/3515 [08:47<08:06,  3.49it/s]
overall:  45%|████▌     | 1817/4000 [08:50<09:55,  3.67it/s]

train:  52%|█████▏    | 1817/3515 [08:47<07:42,  3.67it/s]
overall:  45%|████▌     | 1818/4000 [08:50<10:06,  3.60it/s]

train:  52%|█████▏    | 1818/3515 [08:47<07:51,  3.60it/s]
overall:  45%|████▌     | 1819/4000 [08:50<09:55,  3.66it/s]

train:  52%|█████▏    | 1819/3515 [08:48<07:43,  3.66it/s]train loss: 0.5611256957054138
train per: 0.16337843341327662

overall:  46%|████▌     | 1820/4000 [08:51<10:07,  3.59it/s]

train:  52%|█████▏    | 1820/3515 [08:48<07:52,  3.59it/s]
overall:  46%|████▌     | 1821/4000 [08:51<10:14,  3.54it/s]

train:  52%|█████▏    | 1821/3515 [08:48<07:57,  3.54it/s]
overall:  46%|████▌     | 1822/4000 [08:51<10:05,  3.60it/s]

train:  52%|█████▏    | 1822/3515 [08:49<07:50,  3.60it/s]
overall:  46%|████▌     | 1823/4000 [08:52<10:03,  3.61it/s]

train:  52%|█████▏    | 1823/3515 [08:49<07:49,  3.61it/s]
overall:  46%|████▌     | 1824/4000 [08:52<09:58,  3.64it/s]

train:  52%|█████▏    | 1824/3515 [08:49<07:44,  3.64it/s]
overall:  46%|████▌     | 1825/4000 [08:52<09:21,  3.87it/s]

train:  52%|█████▏    | 1825/3515 [08:49<07:16,  3.87it/s]
overall:  46%|████▌     | 1826/4000 [08:52<09:32,  3.80it/s]

train:  52%|█████▏    | 1826/3515 [08:50<07:24,  3.80it/s]
overall:  46%|████▌     | 1827/4000 [08:53<09:31,  3.80it/s]

train:  52%|█████▏    | 1827/3515 [08:50<07:24,  3.80it/s]
overall:  46%|████▌     | 1828/4000 [08:53<09:19,  3.88it/s]

train:  52%|█████▏    | 1828/3515 [08:50<07:14,  3.88it/s]
overall:  46%|████▌     | 1829/4000 [08:53<09:30,  3.80it/s]

train:  52%|█████▏    | 1829/3515 [08:50<07:23,  3.80it/s]
overall:  46%|████▌     | 1830/4000 [08:53<09:26,  3.83it/s]

train:  52%|█████▏    | 1830/3515 [08:51<07:19,  3.83it/s]
overall:  46%|████▌     | 1831/4000 [08:53<08:07,  4.45it/s]

train:  52%|█████▏    | 1831/3515 [08:51<06:18,  4.45it/s]
overall:  46%|████▌     | 1832/4000 [08:54<08:39,  4.17it/s]

train:  52%|█████▏    | 1832/3515 [08:51<06:43,  4.17it/s]
overall:  46%|████▌     | 1833/4000 [08:54<09:00,  4.01it/s]

train:  52%|█████▏    | 1833/3515 [08:51<06:59,  4.01it/s]
overall:  46%|████▌     | 1834/4000 [08:54<09:16,  3.89it/s]

train:  52%|█████▏    | 1834/3515 [08:52<07:12,  3.89it/s]
overall:  46%|████▌     | 1835/4000 [08:55<09:25,  3.83it/s]

train:  52%|█████▏    | 1835/3515 [08:52<07:18,  3.83it/s]
overall:  46%|████▌     | 1836/4000 [08:55<08:26,  4.28it/s]

train:  52%|█████▏    | 1836/3515 [08:52<06:32,  4.28it/s]
overall:  46%|████▌     | 1837/4000 [08:55<08:53,  4.05it/s]

train:  52%|█████▏    | 1837/3515 [08:52<06:54,  4.05it/s]
overall:  46%|████▌     | 1838/4000 [08:55<09:18,  3.87it/s]

train:  52%|█████▏    | 1838/3515 [08:53<07:13,  3.87it/s]
overall:  46%|████▌     | 1839/4000 [08:56<10:20,  3.48it/s]

train:  52%|█████▏    | 1839/3515 [08:53<08:01,  3.48it/s]train loss: 0.5726312398910522
train per: 0.16714821977021954

overall:  46%|████▌     | 1840/4000 [08:56<10:10,  3.54it/s]

train:  52%|█████▏    | 1840/3515 [08:53<07:53,  3.54it/s]
overall:  46%|████▌     | 1841/4000 [08:56<10:07,  3.55it/s]

train:  52%|█████▏    | 1841/3515 [08:54<07:51,  3.55it/s]
overall:  46%|████▌     | 1842/4000 [08:56<10:00,  3.59it/s]

train:  52%|█████▏    | 1842/3515 [08:54<07:45,  3.59it/s]
overall:  46%|████▌     | 1843/4000 [08:57<09:42,  3.71it/s]

train:  52%|█████▏    | 1843/3515 [08:54<07:31,  3.71it/s]
overall:  46%|████▌     | 1844/4000 [08:57<10:22,  3.46it/s]

train:  52%|█████▏    | 1844/3515 [08:54<08:02,  3.46it/s]
overall:  46%|████▌     | 1845/4000 [08:57<10:13,  3.51it/s]

train:  52%|█████▏    | 1845/3515 [08:55<07:55,  3.51it/s]
overall:  46%|████▌     | 1846/4000 [08:58<09:42,  3.70it/s]

train:  53%|█████▎    | 1846/3515 [08:55<07:31,  3.70it/s]
overall:  46%|████▌     | 1847/4000 [08:58<09:49,  3.65it/s]

train:  53%|█████▎    | 1847/3515 [08:55<07:36,  3.65it/s]
overall:  46%|████▌     | 1848/4000 [08:58<08:28,  4.24it/s]

train:  53%|█████▎    | 1848/3515 [08:55<06:33,  4.24it/s]
overall:  46%|████▌     | 1849/4000 [08:58<08:45,  4.09it/s]

train:  53%|█████▎    | 1849/3515 [08:56<06:46,  4.09it/s]
overall:  46%|████▋     | 1850/4000 [08:59<09:04,  3.95it/s]

train:  53%|█████▎    | 1850/3515 [08:56<07:01,  3.95it/s]
overall:  46%|████▋     | 1851/4000 [08:59<08:56,  4.01it/s]

train:  53%|█████▎    | 1851/3515 [08:56<06:55,  4.01it/s]
overall:  46%|████▋     | 1852/4000 [08:59<09:10,  3.90it/s]

train:  53%|█████▎    | 1852/3515 [08:56<07:06,  3.90it/s]
overall:  46%|████▋     | 1853/4000 [08:59<09:15,  3.86it/s]

train:  53%|█████▎    | 1853/3515 [08:57<07:10,  3.86it/s]
overall:  46%|████▋     | 1854/4000 [09:00<09:20,  3.83it/s]

train:  53%|█████▎    | 1854/3515 [08:57<07:13,  3.83it/s]
overall:  46%|████▋     | 1855/4000 [09:00<09:23,  3.80it/s]

train:  53%|█████▎    | 1855/3515 [08:57<07:16,  3.80it/s]
overall:  46%|████▋     | 1856/4000 [09:00<09:29,  3.76it/s]

train:  53%|█████▎    | 1856/3515 [08:57<07:20,  3.76it/s]
overall:  46%|████▋     | 1857/4000 [09:01<12:01,  2.97it/s]

train:  53%|█████▎    | 1857/3515 [08:58<09:17,  2.97it/s]
overall:  46%|████▋     | 1858/4000 [09:01<11:19,  3.15it/s]

train:  53%|█████▎    | 1858/3515 [08:58<08:45,  3.15it/s]
overall:  46%|████▋     | 1859/4000 [09:01<10:28,  3.41it/s]

train:  53%|█████▎    | 1859/3515 [08:58<08:05,  3.41it/s]train loss: 0.5919003486633301
train per: 0.16456810181424317

overall:  46%|████▋     | 1860/4000 [09:01<10:26,  3.41it/s]

train:  53%|█████▎    | 1860/3515 [08:59<08:05,  3.41it/s]
overall:  47%|████▋     | 1862/4000 [09:02<08:24,  4.24it/s]

train:  53%|█████▎    | 1862/3515 [08:59<06:30,  4.24it/s]
overall:  47%|████▋     | 1863/4000 [09:02<08:42,  4.09it/s]

train:  53%|█████▎    | 1863/3515 [08:59<06:43,  4.09it/s]
overall:  47%|████▋     | 1864/4000 [09:02<08:48,  4.04it/s]

train:  53%|█████▎    | 1864/3515 [09:00<06:48,  4.04it/s]
overall:  47%|████▋     | 1865/4000 [09:03<08:54,  4.00it/s]

train:  53%|█████▎    | 1865/3515 [09:00<06:52,  4.00it/s]
overall:  47%|████▋     | 1866/4000 [09:03<09:07,  3.90it/s]

train:  53%|█████▎    | 1866/3515 [09:00<07:03,  3.90it/s]
overall:  47%|████▋     | 1867/4000 [09:03<09:11,  3.87it/s]

train:  53%|█████▎    | 1867/3515 [09:00<07:06,  3.87it/s]
overall:  47%|████▋     | 1868/4000 [09:03<09:47,  3.63it/s]

train:  53%|█████▎    | 1868/3515 [09:01<07:33,  3.63it/s]
overall:  47%|████▋     | 1869/4000 [09:04<08:47,  4.04it/s]

train:  53%|█████▎    | 1869/3515 [09:01<06:47,  4.04it/s]
overall:  47%|████▋     | 1870/4000 [09:04<08:55,  3.98it/s]

train:  53%|█████▎    | 1870/3515 [09:01<06:53,  3.98it/s]
overall:  47%|████▋     | 1872/4000 [09:04<07:24,  4.79it/s]

train:  53%|█████▎    | 1872/3515 [09:02<05:43,  4.79it/s]
overall:  47%|████▋     | 1873/4000 [09:04<07:54,  4.48it/s]

train:  53%|█████▎    | 1873/3515 [09:02<06:06,  4.48it/s]
overall:  47%|████▋     | 1874/4000 [09:05<08:13,  4.31it/s]

train:  53%|█████▎    | 1874/3515 [09:02<06:20,  4.31it/s]
overall:  47%|████▋     | 1875/4000 [09:05<08:16,  4.28it/s]

train:  53%|█████▎    | 1875/3515 [09:02<06:23,  4.28it/s]
overall:  47%|████▋     | 1876/4000 [09:05<08:26,  4.20it/s]

train:  53%|█████▎    | 1876/3515 [09:03<06:30,  4.20it/s]
overall:  47%|████▋     | 1877/4000 [09:05<08:50,  4.01it/s]

train:  53%|█████▎    | 1877/3515 [09:03<06:48,  4.01it/s]
overall:  47%|████▋     | 1878/4000 [09:06<08:54,  3.97it/s]

train:  53%|█████▎    | 1878/3515 [09:03<06:51,  3.97it/s]
overall:  47%|████▋     | 1879/4000 [09:06<09:09,  3.86it/s]

train:  53%|█████▎    | 1879/3515 [09:03<07:03,  3.86it/s]train loss: 0.5606261491775513
train per: 0.164014298954058

overall:  47%|████▋     | 1880/4000 [09:06<08:44,  4.04it/s]

train:  53%|█████▎    | 1880/3515 [09:04<06:44,  4.04it/s]
overall:  47%|████▋     | 1881/4000 [09:07<10:58,  3.22it/s]

train:  54%|█████▎    | 1881/3515 [09:04<08:27,  3.22it/s]
overall:  47%|████▋     | 1882/4000 [09:07<10:34,  3.34it/s]

train:  54%|█████▎    | 1882/3515 [09:04<08:09,  3.34it/s]
overall:  47%|████▋     | 1883/4000 [09:07<10:16,  3.44it/s]

train:  54%|█████▎    | 1883/3515 [09:05<07:54,  3.44it/s]
overall:  47%|████▋     | 1884/4000 [09:07<09:46,  3.61it/s]

train:  54%|█████▎    | 1884/3515 [09:05<07:32,  3.61it/s]
overall:  47%|████▋     | 1885/4000 [09:08<09:24,  3.75it/s]

train:  54%|█████▎    | 1885/3515 [09:05<07:14,  3.75it/s]
overall:  47%|████▋     | 1886/4000 [09:08<09:24,  3.75it/s]

train:  54%|█████▎    | 1886/3515 [09:05<07:14,  3.75it/s]
overall:  47%|████▋     | 1888/4000 [09:08<07:59,  4.40it/s]

train:  54%|█████▎    | 1888/3515 [09:06<06:09,  4.40it/s]
overall:  47%|████▋     | 1889/4000 [09:09<08:16,  4.25it/s]

train:  54%|█████▎    | 1889/3515 [09:06<06:22,  4.25it/s]
overall:  47%|████▋     | 1890/4000 [09:09<08:32,  4.12it/s]

train:  54%|█████▍    | 1890/3515 [09:06<06:34,  4.12it/s]
overall:  47%|████▋     | 1891/4000 [09:09<08:44,  4.02it/s]

train:  54%|█████▍    | 1891/3515 [09:06<06:43,  4.02it/s]
overall:  47%|████▋     | 1892/4000 [09:09<10:13,  3.44it/s]

train:  54%|█████▍    | 1892/3515 [09:07<07:52,  3.44it/s]
overall:  47%|████▋     | 1893/4000 [09:10<10:02,  3.49it/s]

train:  54%|█████▍    | 1893/3515 [09:07<07:44,  3.49it/s]
overall:  47%|████▋     | 1894/4000 [09:10<09:47,  3.58it/s]

train:  54%|█████▍    | 1894/3515 [09:07<07:32,  3.58it/s]
overall:  47%|████▋     | 1895/4000 [09:10<09:39,  3.63it/s]

train:  54%|█████▍    | 1895/3515 [09:08<07:26,  3.63it/s]
overall:  47%|████▋     | 1896/4000 [09:11<09:06,  3.85it/s]

train:  54%|█████▍    | 1896/3515 [09:08<07:00,  3.85it/s]
overall:  47%|████▋     | 1897/4000 [09:11<09:06,  3.85it/s]

train:  54%|█████▍    | 1897/3515 [09:08<07:00,  3.85it/s]
overall:  47%|████▋     | 1898/4000 [09:11<09:08,  3.83it/s]

train:  54%|█████▍    | 1898/3515 [09:08<07:02,  3.83it/s]
overall:  47%|████▋     | 1899/4000 [09:11<09:12,  3.80it/s]

train:  54%|█████▍    | 1899/3515 [09:09<07:04,  3.80it/s]train loss: 0.5911617875099182
train per: 0.16811949069539667

overall:  48%|████▊     | 1901/4000 [09:12<07:43,  4.53it/s]

train:  54%|█████▍    | 1901/3515 [09:09<05:56,  4.53it/s]
overall:  48%|████▊     | 1902/4000 [09:12<08:11,  4.27it/s]

train:  54%|█████▍    | 1902/3515 [09:09<06:17,  4.27it/s]
overall:  48%|████▊     | 1903/4000 [09:12<07:08,  4.90it/s]

train:  54%|█████▍    | 1903/3515 [09:09<05:29,  4.90it/s]
overall:  48%|████▊     | 1904/4000 [09:12<07:51,  4.44it/s]

train:  54%|█████▍    | 1904/3515 [09:10<06:02,  4.44it/s]
overall:  48%|████▊     | 1905/4000 [09:13<09:59,  3.49it/s]

train:  54%|█████▍    | 1905/3515 [09:10<07:40,  3.49it/s]
overall:  48%|████▊     | 1906/4000 [09:13<09:33,  3.65it/s]

train:  54%|█████▍    | 1906/3515 [09:10<07:20,  3.65it/s]
overall:  48%|████▊     | 1907/4000 [09:13<09:31,  3.66it/s]

train:  54%|█████▍    | 1907/3515 [09:11<07:18,  3.66it/s]
overall:  48%|████▊     | 1908/4000 [09:14<09:27,  3.69it/s]

train:  54%|█████▍    | 1908/3515 [09:11<07:15,  3.69it/s]
overall:  48%|████▊     | 1909/4000 [09:14<09:29,  3.67it/s]

train:  54%|█████▍    | 1909/3515 [09:11<07:17,  3.67it/s]
overall:  48%|████▊     | 1910/4000 [09:14<09:21,  3.72it/s]

train:  54%|█████▍    | 1910/3515 [09:11<07:11,  3.72it/s]
overall:  48%|████▊     | 1911/4000 [09:14<09:16,  3.75it/s]

train:  54%|█████▍    | 1911/3515 [09:12<07:07,  3.75it/s]
overall:  48%|████▊     | 1912/4000 [09:15<09:12,  3.78it/s]

train:  54%|█████▍    | 1912/3515 [09:12<07:04,  3.78it/s]
overall:  48%|████▊     | 1913/4000 [09:15<09:10,  3.79it/s]

train:  54%|█████▍    | 1913/3515 [09:12<07:02,  3.79it/s]
overall:  48%|████▊     | 1914/4000 [09:15<08:54,  3.90it/s]

train:  54%|█████▍    | 1914/3515 [09:12<06:50,  3.90it/s]
overall:  48%|████▊     | 1915/4000 [09:15<09:03,  3.83it/s]

train:  54%|█████▍    | 1915/3515 [09:13<06:57,  3.83it/s]
overall:  48%|████▊     | 1916/4000 [09:16<10:17,  3.38it/s]

train:  55%|█████▍    | 1916/3515 [09:13<07:53,  3.38it/s]
overall:  48%|████▊     | 1917/4000 [09:16<09:23,  3.70it/s]

train:  55%|█████▍    | 1917/3515 [09:13<07:12,  3.70it/s]
overall:  48%|████▊     | 1918/4000 [09:16<09:16,  3.74it/s]

train:  55%|█████▍    | 1918/3515 [09:14<07:06,  3.74it/s]
overall:  48%|████▊     | 1919/4000 [09:17<09:28,  3.66it/s]

train:  55%|█████▍    | 1919/3515 [09:14<07:16,  3.66it/s]train loss: 0.5845912098884583
train per: 0.1679559551188236

overall:  48%|████▊     | 1920/4000 [09:17<09:24,  3.68it/s]

train:  55%|█████▍    | 1920/3515 [09:14<07:13,  3.68it/s]
overall:  48%|████▊     | 1921/4000 [09:17<09:18,  3.72it/s]

train:  55%|█████▍    | 1921/3515 [09:14<07:07,  3.72it/s]
overall:  48%|████▊     | 1922/4000 [09:17<09:09,  3.78it/s]

train:  55%|█████▍    | 1922/3515 [09:15<07:00,  3.79it/s]
overall:  48%|████▊     | 1923/4000 [09:18<09:13,  3.76it/s]

train:  55%|█████▍    | 1923/3515 [09:15<07:03,  3.76it/s]
overall:  48%|████▊     | 1924/4000 [09:18<07:59,  4.33it/s]

train:  55%|█████▍    | 1924/3515 [09:15<06:07,  4.33it/s]
overall:  48%|████▊     | 1925/4000 [09:18<08:23,  4.12it/s]

train:  55%|█████▍    | 1925/3515 [09:15<06:25,  4.12it/s]
overall:  48%|████▊     | 1926/4000 [09:18<08:34,  4.03it/s]

train:  55%|█████▍    | 1926/3515 [09:16<06:34,  4.03it/s]
overall:  48%|████▊     | 1927/4000 [09:18<07:51,  4.40it/s]

train:  55%|█████▍    | 1927/3515 [09:16<06:00,  4.40it/s]
overall:  48%|████▊     | 1928/4000 [09:19<08:24,  4.11it/s]

train:  55%|█████▍    | 1928/3515 [09:16<06:26,  4.11it/s]
overall:  48%|████▊     | 1929/4000 [09:19<10:46,  3.20it/s]

train:  55%|█████▍    | 1929/3515 [09:17<08:15,  3.20it/s]
overall:  48%|████▊     | 1930/4000 [09:19<10:19,  3.34it/s]

train:  55%|█████▍    | 1930/3515 [09:17<07:54,  3.34it/s]
overall:  48%|████▊     | 1931/4000 [09:20<08:36,  4.00it/s]

train:  55%|█████▍    | 1931/3515 [09:17<06:35,  4.00it/s]
overall:  48%|████▊     | 1932/4000 [09:20<09:01,  3.82it/s]

train:  55%|█████▍    | 1932/3515 [09:17<06:54,  3.82it/s]
overall:  48%|████▊     | 1933/4000 [09:20<08:44,  3.94it/s]

train:  55%|█████▍    | 1933/3515 [09:17<06:41,  3.94it/s]
overall:  48%|████▊     | 1935/4000 [09:20<07:31,  4.58it/s]

train:  55%|█████▌    | 1935/3515 [09:18<05:45,  4.58it/s]
overall:  48%|████▊     | 1937/4000 [09:21<06:53,  4.99it/s]

train:  55%|█████▌    | 1937/3515 [09:18<05:16,  4.99it/s]
overall:  48%|████▊     | 1938/4000 [09:21<07:22,  4.66it/s]

train:  55%|█████▌    | 1938/3515 [09:18<05:38,  4.66it/s]
overall:  48%|████▊     | 1939/4000 [09:21<07:20,  4.68it/s]

train:  55%|█████▌    | 1939/3515 [09:19<05:36,  4.68it/s]train loss: 0.5768903493881226
train per: 0.16591820368885324

overall:  48%|████▊     | 1940/4000 [09:22<08:50,  3.88it/s]

train:  55%|█████▌    | 1940/3515 [09:19<06:46,  3.88it/s]
overall:  49%|████▊     | 1941/4000 [09:22<09:12,  3.73it/s]

train:  55%|█████▌    | 1941/3515 [09:19<07:02,  3.73it/s]
overall:  49%|████▊     | 1942/4000 [09:22<09:16,  3.70it/s]

train:  55%|█████▌    | 1942/3515 [09:20<07:05,  3.70it/s]
overall:  49%|████▊     | 1943/4000 [09:23<09:17,  3.69it/s]

train:  55%|█████▌    | 1943/3515 [09:20<07:05,  3.69it/s]
overall:  49%|████▊     | 1944/4000 [09:23<08:45,  3.91it/s]

train:  55%|█████▌    | 1944/3515 [09:20<06:41,  3.91it/s]
overall:  49%|████▊     | 1945/4000 [09:23<08:49,  3.88it/s]

train:  55%|█████▌    | 1945/3515 [09:20<06:44,  3.88it/s]
overall:  49%|████▊     | 1946/4000 [09:23<07:45,  4.41it/s]

train:  55%|█████▌    | 1946/3515 [09:21<05:55,  4.41it/s]
overall:  49%|████▊     | 1947/4000 [09:23<07:39,  4.47it/s]

train:  55%|█████▌    | 1947/3515 [09:21<05:50,  4.47it/s]
overall:  49%|████▊     | 1948/4000 [09:24<07:46,  4.40it/s]

train:  55%|█████▌    | 1948/3515 [09:21<05:56,  4.40it/s]
overall:  49%|████▊     | 1949/4000 [09:24<08:13,  4.15it/s]

train:  55%|█████▌    | 1949/3515 [09:21<06:16,  4.15it/s]
overall:  49%|████▉     | 1950/4000 [09:24<08:28,  4.03it/s]

train:  55%|█████▌    | 1950/3515 [09:22<06:28,  4.03it/s]
overall:  49%|████▉     | 1951/4000 [09:24<08:36,  3.97it/s]

train:  56%|█████▌    | 1951/3515 [09:22<06:34,  3.97it/s]
overall:  49%|████▉     | 1952/4000 [09:25<08:41,  3.92it/s]

train:  56%|█████▌    | 1952/3515 [09:22<06:38,  3.92it/s]
overall:  49%|████▉     | 1953/4000 [09:25<10:37,  3.21it/s]

train:  56%|█████▌    | 1953/3515 [09:22<08:06,  3.21it/s]
overall:  49%|████▉     | 1954/4000 [09:25<10:13,  3.34it/s]

train:  56%|█████▌    | 1954/3515 [09:23<07:47,  3.34it/s]
overall:  49%|████▉     | 1955/4000 [09:26<09:55,  3.43it/s]

train:  56%|█████▌    | 1955/3515 [09:23<07:34,  3.43it/s]
overall:  49%|████▉     | 1956/4000 [09:26<09:44,  3.50it/s]

train:  56%|█████▌    | 1956/3515 [09:23<07:25,  3.50it/s]
overall:  49%|████▉     | 1958/4000 [09:26<07:54,  4.31it/s]

train:  56%|█████▌    | 1958/3515 [09:24<06:01,  4.31it/s]
overall:  49%|████▉     | 1959/4000 [09:27<08:07,  4.19it/s]

train:  56%|█████▌    | 1959/3515 [09:24<06:11,  4.19it/s]train loss: 0.5925744771957397
train per: 0.1725608715786395

overall:  49%|████▉     | 1960/4000 [09:27<08:18,  4.10it/s]

train:  56%|█████▌    | 1960/3515 [09:24<06:20,  4.09it/s]
overall:  49%|████▉     | 1961/4000 [09:27<08:32,  3.98it/s]

train:  56%|█████▌    | 1961/3515 [09:24<06:30,  3.98it/s]
overall:  49%|████▉     | 1962/4000 [09:27<08:46,  3.87it/s]

train:  56%|█████▌    | 1962/3515 [09:25<06:41,  3.87it/s]
overall:  49%|████▉     | 1963/4000 [09:28<08:54,  3.81it/s]

train:  56%|█████▌    | 1963/3515 [09:25<06:47,  3.81it/s]
overall:  49%|████▉     | 1964/4000 [09:28<09:54,  3.42it/s]

train:  56%|█████▌    | 1964/3515 [09:25<07:33,  3.42it/s]
overall:  49%|████▉     | 1965/4000 [09:28<09:38,  3.52it/s]

train:  56%|█████▌    | 1965/3515 [09:26<07:20,  3.52it/s]
overall:  49%|████▉     | 1966/4000 [09:28<09:23,  3.61it/s]

train:  56%|█████▌    | 1966/3515 [09:26<07:09,  3.61it/s]
overall:  49%|████▉     | 1967/4000 [09:29<09:26,  3.59it/s]

train:  56%|█████▌    | 1967/3515 [09:26<07:11,  3.59it/s]
overall:  49%|████▉     | 1968/4000 [09:29<09:23,  3.61it/s]

train:  56%|█████▌    | 1968/3515 [09:26<07:08,  3.61it/s]
overall:  49%|████▉     | 1970/4000 [09:29<06:49,  4.95it/s]

train:  56%|█████▌    | 1970/3515 [09:27<05:11,  4.96it/s]
overall:  49%|████▉     | 1971/4000 [09:30<07:11,  4.71it/s]

train:  56%|█████▌    | 1971/3515 [09:27<05:28,  4.71it/s]
overall:  49%|████▉     | 1972/4000 [09:30<07:39,  4.41it/s]

train:  56%|█████▌    | 1972/3515 [09:27<05:49,  4.41it/s]
overall:  49%|████▉     | 1973/4000 [09:30<08:08,  4.15it/s]

train:  56%|█████▌    | 1973/3515 [09:27<06:11,  4.15it/s]
overall:  49%|████▉     | 1974/4000 [09:30<08:24,  4.01it/s]

train:  56%|█████▌    | 1974/3515 [09:28<06:23,  4.01it/s]
overall:  49%|████▉     | 1975/4000 [09:30<06:59,  4.83it/s]

train:  56%|█████▌    | 1975/3515 [09:28<05:18,  4.83it/s]
overall:  49%|████▉     | 1976/4000 [09:31<06:59,  4.82it/s]

train:  56%|█████▌    | 1976/3515 [09:28<05:19,  4.82it/s]
overall:  49%|████▉     | 1977/4000 [09:31<09:33,  3.53it/s]

train:  56%|█████▌    | 1977/3515 [09:28<07:16,  3.53it/s]
overall:  49%|████▉     | 1978/4000 [09:31<09:21,  3.60it/s]

train:  56%|█████▋    | 1978/3515 [09:29<07:06,  3.60it/s]
overall:  49%|████▉     | 1979/4000 [09:32<09:10,  3.67it/s]

train:  56%|█████▋    | 1979/3515 [09:29<06:58,  3.67it/s]train loss: 0.5987033247947693
train per: 0.17096867749419953

overall:  50%|████▉     | 1980/4000 [09:32<09:14,  3.65it/s]

train:  56%|█████▋    | 1980/3515 [09:29<07:01,  3.64it/s]
overall:  50%|████▉     | 1981/4000 [09:32<09:06,  3.70it/s]

train:  56%|█████▋    | 1981/3515 [09:30<06:55,  3.70it/s]
overall:  50%|████▉     | 1982/4000 [09:32<07:23,  4.55it/s]

train:  56%|█████▋    | 1982/3515 [09:30<05:36,  4.55it/s]
overall:  50%|████▉     | 1983/4000 [09:33<07:33,  4.44it/s]

train:  56%|█████▋    | 1983/3515 [09:30<05:44,  4.44it/s]
overall:  50%|████▉     | 1984/4000 [09:33<08:01,  4.18it/s]

train:  56%|█████▋    | 1984/3515 [09:30<06:05,  4.18it/s]
overall:  50%|████▉     | 1985/4000 [09:33<08:12,  4.09it/s]

train:  56%|█████▋    | 1985/3515 [09:30<06:14,  4.09it/s]
overall:  50%|████▉     | 1986/4000 [09:33<08:04,  4.16it/s]

train:  57%|█████▋    | 1986/3515 [09:31<06:07,  4.16it/s]
overall:  50%|████▉     | 1987/4000 [09:34<08:25,  3.98it/s]

train:  57%|█████▋    | 1987/3515 [09:31<06:23,  3.99it/s]
overall:  50%|████▉     | 1988/4000 [09:34<09:33,  3.51it/s]

train:  57%|█████▋    | 1988/3515 [09:31<07:15,  3.51it/s]
overall:  50%|████▉     | 1989/4000 [09:34<09:34,  3.50it/s]

train:  57%|█████▋    | 1989/3515 [09:32<07:15,  3.50it/s]
overall:  50%|████▉     | 1990/4000 [09:34<09:14,  3.62it/s]

train:  57%|█████▋    | 1990/3515 [09:32<07:01,  3.62it/s]
overall:  50%|████▉     | 1991/4000 [09:35<09:13,  3.63it/s]

train:  57%|█████▋    | 1991/3515 [09:32<07:00,  3.63it/s]
overall:  50%|████▉     | 1992/4000 [09:35<09:11,  3.64it/s]

train:  57%|█████▋    | 1992/3515 [09:32<06:58,  3.64it/s]
overall:  50%|████▉     | 1993/4000 [09:35<09:02,  3.70it/s]

train:  57%|█████▋    | 1993/3515 [09:33<06:51,  3.70it/s]
overall:  50%|████▉     | 1994/4000 [09:36<08:56,  3.74it/s]

train:  57%|█████▋    | 1994/3515 [09:33<06:46,  3.74it/s]
overall:  50%|████▉     | 1995/4000 [09:36<09:00,  3.71it/s]

train:  57%|█████▋    | 1995/3515 [09:33<06:50,  3.71it/s]
overall:  50%|████▉     | 1997/4000 [09:36<07:42,  4.33it/s]

train:  57%|█████▋    | 1997/3515 [09:34<05:50,  4.33it/s]
overall:  50%|████▉     | 1998/4000 [09:36<07:57,  4.19it/s]

train:  57%|█████▋    | 1998/3515 [09:34<06:01,  4.19it/s]
overall:  50%|████▉     | 1999/4000 [09:37<08:06,  4.11it/s]

train:  57%|█████▋    | 1999/3515 [09:34<06:08,  4.11it/s]train loss: 0.5877385139465332
train per: 0.1723030821917808
[LibriPhone] - Taking the first phoneme sequences for a deterministic behavior.



word -> phonemes:   0%|          | 0/2703 [00:00<?, ?it/s]


word -> phonemes:  12%|█▏        | 311/2703 [00:00<00:00, 3102.41it/s]


word -> phonemes:  23%|██▎       | 622/2703 [00:00<00:00, 3070.26it/s]


word -> phonemes:  35%|███▌      | 955/2703 [00:00<00:00, 3187.45it/s]


word -> phonemes:  47%|████▋     | 1274/2703 [00:00<00:00, 3084.10it/s]


word -> phonemes:  60%|█████▉    | 1611/2703 [00:00<00:00, 3184.29it/s]


word -> phonemes:  72%|███████▏  | 1933/2703 [00:00<00:00, 3194.30it/s]


word -> phonemes:  83%|████████▎ | 2253/2703 [00:00<00:00, 3177.78it/s]


word -> phonemes:  95%|█████████▌| 2577/2703 [00:00<00:00, 3195.43it/s]
word -> phonemes: 100%|██████████| 2703/2703 [00:00<00:00, 3164.36it/s]



dev:   0%|          | 0/2703 [00:00<?, ?it/s]


dev:   0%|          | 1/2703 [00:00<43:06,  1.04it/s]


dev:   0%|          | 4/2703 [00:01<09:53,  4.55it/s]


dev:   0%|          | 6/2703 [00:01<06:33,  6.85it/s]


dev:   0%|          | 9/2703 [00:01<04:17, 10.44it/s]


dev:   0%|          | 12/2703 [00:01<03:16, 13.72it/s]


dev:   1%|          | 15/2703 [00:01<02:41, 16.67it/s]


dev:   1%|          | 18/2703 [00:01<02:20, 19.15it/s]


dev:   1%|          | 21/2703 [00:01<02:05, 21.45it/s]


dev:   1%|          | 24/2703 [00:01<01:56, 23.03it/s]


dev:   1%|          | 27/2703 [00:02<01:52, 23.76it/s]


dev:   1%|          | 30/2703 [00:02<01:48, 24.61it/s]


dev:   1%|▏         | 34/2703 [00:02<01:38, 26.97it/s]


dev:   1%|▏         | 37/2703 [00:02<01:36, 27.65it/s]


dev:   2%|▏         | 41/2703 [00:02<01:33, 28.59it/s]


dev:   2%|▏         | 45/2703 [00:02<01:26, 30.79it/s]


dev:   2%|▏         | 49/2703 [00:02<01:23, 31.68it/s]


dev:   2%|▏         | 53/2703 [00:02<01:24, 31.36it/s]


dev:   2%|▏         | 57/2703 [00:02<01:22, 32.22it/s]


dev:   2%|▏         | 61/2703 [00:03<01:20, 32.90it/s]


dev:   2%|▏         | 65/2703 [00:03<01:17, 33.86it/s]


dev:   3%|▎         | 69/2703 [00:03<01:16, 34.41it/s]


dev:   3%|▎         | 73/2703 [00:03<01:17, 34.01it/s]


dev:   3%|▎         | 77/2703 [00:03<01:16, 34.32it/s]


dev:   3%|▎         | 82/2703 [00:03<01:12, 36.26it/s]


dev:   3%|▎         | 86/2703 [00:03<01:14, 35.14it/s]


dev:   3%|▎         | 90/2703 [00:03<01:14, 35.09it/s]


dev:   3%|▎         | 94/2703 [00:03<01:14, 34.88it/s]


dev:   4%|▎         | 98/2703 [00:04<01:12, 35.99it/s]


dev:   4%|▍         | 102/2703 [00:04<01:12, 35.82it/s]


dev:   4%|▍         | 106/2703 [00:04<01:11, 36.44it/s]


dev:   4%|▍         | 110/2703 [00:04<01:11, 36.49it/s]


dev:   4%|▍         | 114/2703 [00:04<01:10, 36.97it/s]


dev:   4%|▍         | 118/2703 [00:04<01:10, 36.92it/s]


dev:   5%|▍         | 122/2703 [00:04<01:09, 37.25it/s]


dev:   5%|▍         | 126/2703 [00:04<01:09, 37.15it/s]


dev:   5%|▍         | 131/2703 [00:04<01:06, 38.67it/s]


dev:   5%|▍         | 135/2703 [00:05<01:07, 37.96it/s]


dev:   5%|▌         | 139/2703 [00:05<01:06, 38.30it/s]


dev:   5%|▌         | 143/2703 [00:05<01:06, 38.39it/s]


dev:   5%|▌         | 147/2703 [00:05<01:07, 37.83it/s]


dev:   6%|▌         | 151/2703 [00:05<01:09, 36.62it/s]


dev:   6%|▌         | 156/2703 [00:05<01:06, 38.24it/s]


dev:   6%|▌         | 160/2703 [00:05<01:07, 37.67it/s]


dev:   6%|▌         | 164/2703 [00:05<01:06, 38.23it/s]


dev:   6%|▋         | 169/2703 [00:05<01:05, 38.56it/s]


dev:   6%|▋         | 174/2703 [00:06<01:04, 39.42it/s]


dev:   7%|▋         | 179/2703 [00:06<01:01, 41.15it/s]


dev:   7%|▋         | 184/2703 [00:06<01:02, 40.34it/s]


dev:   7%|▋         | 189/2703 [00:06<01:01, 40.68it/s]


dev:   7%|▋         | 194/2703 [00:06<01:00, 41.24it/s]


dev:   7%|▋         | 199/2703 [00:06<00:59, 41.83it/s]


dev:   8%|▊         | 204/2703 [00:06<00:59, 42.14it/s]


dev:   8%|▊         | 209/2703 [00:06<01:00, 41.08it/s]


dev:   8%|▊         | 214/2703 [00:07<00:58, 42.22it/s]


dev:   8%|▊         | 219/2703 [00:07<00:59, 41.96it/s]


dev:   8%|▊         | 224/2703 [00:07<00:58, 42.20it/s]


dev:   8%|▊         | 229/2703 [00:07<00:59, 41.80it/s]


dev:   9%|▊         | 234/2703 [00:07<00:57, 42.90it/s]


dev:   9%|▉         | 239/2703 [00:07<00:55, 44.38it/s]


dev:   9%|▉         | 244/2703 [00:07<00:56, 43.56it/s]


dev:   9%|▉         | 249/2703 [00:07<00:55, 44.11it/s]


dev:   9%|▉         | 254/2703 [00:07<00:56, 43.65it/s]


dev:  10%|▉         | 259/2703 [00:08<00:54, 44.86it/s]


dev:  10%|▉         | 264/2703 [00:08<00:55, 44.08it/s]


dev:  10%|▉         | 269/2703 [00:08<00:54, 44.76it/s]


dev:  10%|█         | 274/2703 [00:08<00:53, 45.60it/s]


dev:  10%|█         | 279/2703 [00:08<00:54, 44.80it/s]


dev:  11%|█         | 284/2703 [00:08<00:52, 45.91it/s]


dev:  11%|█         | 289/2703 [00:08<00:52, 46.15it/s]


dev:  11%|█         | 294/2703 [00:08<00:51, 46.87it/s]


dev:  11%|█         | 299/2703 [00:08<00:50, 47.30it/s]


dev:  11%|█         | 304/2703 [00:09<00:52, 45.89it/s]


dev:  11%|█▏        | 309/2703 [00:09<00:51, 46.87it/s]


dev:  12%|█▏        | 314/2703 [00:09<00:52, 45.59it/s]


dev:  12%|█▏        | 319/2703 [00:09<00:51, 46.37it/s]


dev:  12%|█▏        | 324/2703 [00:09<00:50, 46.75it/s]


dev:  12%|█▏        | 329/2703 [00:09<00:49, 47.64it/s]


dev:  12%|█▏        | 335/2703 [00:09<00:48, 49.24it/s]


dev:  13%|█▎        | 340/2703 [00:09<00:48, 48.40it/s]


dev:  13%|█▎        | 346/2703 [00:09<00:47, 49.29it/s]


dev:  13%|█▎        | 351/2703 [00:10<00:48, 48.48it/s]


dev:  13%|█▎        | 357/2703 [00:10<00:46, 49.94it/s]


dev:  13%|█▎        | 362/2703 [00:10<00:47, 49.36it/s]


dev:  14%|█▎        | 367/2703 [00:10<00:47, 49.12it/s]


dev:  14%|█▍        | 373/2703 [00:10<00:46, 49.71it/s]


dev:  14%|█▍        | 379/2703 [00:10<00:45, 50.86it/s]


dev:  14%|█▍        | 385/2703 [00:10<00:46, 50.21it/s]


dev:  14%|█▍        | 391/2703 [00:10<00:47, 48.33it/s]


dev:  15%|█▍        | 397/2703 [00:10<00:47, 48.72it/s]


dev:  15%|█▍        | 403/2703 [00:11<00:46, 49.67it/s]


dev:  15%|█▌        | 409/2703 [00:11<00:46, 49.79it/s]


dev:  15%|█▌        | 414/2703 [00:11<00:46, 49.72it/s]


dev:  16%|█▌        | 419/2703 [00:11<00:46, 49.29it/s]


dev:  16%|█▌        | 425/2703 [00:11<00:45, 50.24it/s]


dev:  16%|█▌        | 431/2703 [00:11<00:45, 49.94it/s]


dev:  16%|█▌        | 437/2703 [00:11<00:45, 50.18it/s]


dev:  16%|█▋        | 443/2703 [00:11<00:43, 51.38it/s]


dev:  17%|█▋        | 449/2703 [00:11<00:44, 50.10it/s]


dev:  17%|█▋        | 455/2703 [00:12<00:43, 51.24it/s]


dev:  17%|█▋        | 461/2703 [00:12<00:44, 50.78it/s]


dev:  17%|█▋        | 467/2703 [00:12<00:43, 51.27it/s]


dev:  17%|█▋        | 473/2703 [00:12<00:43, 51.13it/s]


dev:  18%|█▊        | 479/2703 [00:12<00:42, 51.84it/s]


dev:  18%|█▊        | 485/2703 [00:12<00:42, 51.67it/s]


dev:  18%|█▊        | 491/2703 [00:12<00:42, 52.10it/s]


dev:  18%|█▊        | 497/2703 [00:12<00:41, 53.71it/s]


dev:  19%|█▊        | 503/2703 [00:13<00:40, 54.30it/s]


dev:  19%|█▉        | 509/2703 [00:13<00:41, 53.47it/s]


dev:  19%|█▉        | 515/2703 [00:13<00:40, 53.95it/s]


dev:  19%|█▉        | 521/2703 [00:13<00:40, 53.42it/s]


dev:  19%|█▉        | 527/2703 [00:13<00:40, 54.35it/s]


dev:  20%|█▉        | 533/2703 [00:13<00:40, 53.03it/s]


dev:  20%|█▉        | 539/2703 [00:13<00:39, 54.44it/s]


dev:  20%|██        | 545/2703 [00:13<00:39, 54.77it/s]


dev:  20%|██        | 551/2703 [00:13<00:40, 53.64it/s]


dev:  21%|██        | 557/2703 [00:13<00:38, 55.15it/s]


dev:  21%|██        | 563/2703 [00:14<00:38, 54.97it/s]


dev:  21%|██        | 569/2703 [00:14<00:38, 55.26it/s]


dev:  21%|██▏       | 575/2703 [00:14<00:39, 54.53it/s]


dev:  21%|██▏       | 581/2703 [00:14<00:39, 54.25it/s]


dev:  22%|██▏       | 587/2703 [00:14<00:39, 53.60it/s]


dev:  22%|██▏       | 593/2703 [00:14<00:39, 53.97it/s]


dev:  22%|██▏       | 599/2703 [00:14<00:38, 54.01it/s]


dev:  22%|██▏       | 605/2703 [00:14<00:37, 55.54it/s]


dev:  23%|██▎       | 611/2703 [00:14<00:38, 54.10it/s]


dev:  23%|██▎       | 617/2703 [00:15<00:37, 55.21it/s]


dev:  23%|██▎       | 623/2703 [00:15<00:37, 55.16it/s]


dev:  23%|██▎       | 630/2703 [00:15<00:37, 55.96it/s]


dev:  24%|██▎       | 636/2703 [00:15<00:37, 55.30it/s]


dev:  24%|██▍       | 642/2703 [00:15<00:36, 56.32it/s]


dev:  24%|██▍       | 649/2703 [00:15<00:35, 57.68it/s]


dev:  24%|██▍       | 655/2703 [00:15<00:35, 56.91it/s]


dev:  24%|██▍       | 661/2703 [00:15<00:36, 56.23it/s]


dev:  25%|██▍       | 667/2703 [00:15<00:36, 55.89it/s]


dev:  25%|██▍       | 673/2703 [00:16<00:35, 56.85it/s]


dev:  25%|██▌       | 679/2703 [00:16<00:37, 54.68it/s]


dev:  25%|██▌       | 686/2703 [00:16<00:35, 56.80it/s]


dev:  26%|██▌       | 692/2703 [00:16<00:35, 56.47it/s]


dev:  26%|██▌       | 698/2703 [00:16<00:35, 57.14it/s]


dev:  26%|██▌       | 704/2703 [00:16<00:35, 56.69it/s]


dev:  26%|██▋       | 710/2703 [00:16<00:34, 57.30it/s]


dev:  26%|██▋       | 716/2703 [00:16<00:34, 57.71it/s]


dev:  27%|██▋       | 722/2703 [00:16<00:34, 57.91it/s]


dev:  27%|██▋       | 728/2703 [00:17<00:34, 57.44it/s]


dev:  27%|██▋       | 734/2703 [00:17<00:34, 57.48it/s]


dev:  27%|██▋       | 740/2703 [00:17<00:34, 57.26it/s]


dev:  28%|██▊       | 746/2703 [00:17<00:34, 57.46it/s]


dev:  28%|██▊       | 752/2703 [00:17<00:33, 58.19it/s]


dev:  28%|██▊       | 759/2703 [00:17<00:32, 59.57it/s]


dev:  28%|██▊       | 765/2703 [00:17<00:33, 57.68it/s]


dev:  29%|██▊       | 772/2703 [00:17<00:32, 59.26it/s]


dev:  29%|██▉       | 778/2703 [00:17<00:32, 58.93it/s]


dev:  29%|██▉       | 785/2703 [00:18<00:32, 59.54it/s]


dev:  29%|██▉       | 791/2703 [00:18<00:32, 59.35it/s]


dev:  30%|██▉       | 798/2703 [00:18<00:31, 60.52it/s]


dev:  30%|██▉       | 805/2703 [00:18<00:31, 60.24it/s]


dev:  30%|███       | 812/2703 [00:18<00:31, 60.30it/s]


dev:  30%|███       | 819/2703 [00:18<00:31, 60.47it/s]


dev:  31%|███       | 826/2703 [00:18<00:30, 61.06it/s]


dev:  31%|███       | 833/2703 [00:18<00:30, 60.35it/s]


dev:  31%|███       | 840/2703 [00:18<00:30, 60.77it/s]


dev:  31%|███▏      | 847/2703 [00:19<00:30, 60.67it/s]


dev:  32%|███▏      | 854/2703 [00:19<00:29, 62.00it/s]


dev:  32%|███▏      | 861/2703 [00:19<00:30, 60.70it/s]


dev:  32%|███▏      | 868/2703 [00:19<00:29, 61.98it/s]


dev:  32%|███▏      | 875/2703 [00:19<00:29, 62.54it/s]


dev:  33%|███▎      | 882/2703 [00:19<00:29, 61.61it/s]


dev:  33%|███▎      | 889/2703 [00:19<00:28, 62.58it/s]


dev:  33%|███▎      | 896/2703 [00:19<00:29, 62.16it/s]


dev:  33%|███▎      | 903/2703 [00:19<00:29, 61.81it/s]


dev:  34%|███▎      | 910/2703 [00:20<00:28, 62.34it/s]


dev:  34%|███▍      | 917/2703 [00:20<00:29, 61.41it/s]


dev:  34%|███▍      | 924/2703 [00:20<00:30, 59.27it/s]


dev:  34%|███▍      | 931/2703 [00:20<00:28, 61.12it/s]


dev:  35%|███▍      | 938/2703 [00:20<00:28, 60.91it/s]


dev:  35%|███▍      | 945/2703 [00:20<00:28, 61.67it/s]


dev:  35%|███▌      | 952/2703 [00:20<00:29, 60.06it/s]


dev:  35%|███▌      | 959/2703 [00:20<00:28, 61.62it/s]


dev:  36%|███▌      | 966/2703 [00:20<00:28, 61.84it/s]


dev:  36%|███▌      | 973/2703 [00:21<00:27, 63.56it/s]


dev:  36%|███▋      | 980/2703 [00:21<00:26, 64.23it/s]


dev:  37%|███▋      | 987/2703 [00:21<00:26, 64.48it/s]


dev:  37%|███▋      | 994/2703 [00:21<00:26, 63.92it/s]


dev:  37%|███▋      | 1001/2703 [00:21<00:26, 64.30it/s]


dev:  37%|███▋      | 1008/2703 [00:21<00:26, 62.82it/s]


dev:  38%|███▊      | 1015/2703 [00:21<00:26, 63.62it/s]


dev:  38%|███▊      | 1022/2703 [00:21<00:26, 63.24it/s]


dev:  38%|███▊      | 1029/2703 [00:21<00:26, 63.58it/s]


dev:  38%|███▊      | 1036/2703 [00:22<00:26, 63.63it/s]


dev:  39%|███▊      | 1043/2703 [00:22<00:26, 62.95it/s]


dev:  39%|███▉      | 1050/2703 [00:22<00:26, 63.51it/s]


dev:  39%|███▉      | 1057/2703 [00:22<00:25, 64.48it/s]


dev:  39%|███▉      | 1064/2703 [00:22<00:25, 65.15it/s]


dev:  40%|███▉      | 1071/2703 [00:22<00:25, 64.60it/s]


dev:  40%|███▉      | 1078/2703 [00:22<00:25, 63.95it/s]


dev:  40%|████      | 1085/2703 [00:22<00:24, 65.43it/s]


dev:  40%|████      | 1092/2703 [00:22<00:25, 62.79it/s]


dev:  41%|████      | 1099/2703 [00:23<00:24, 64.62it/s]


dev:  41%|████      | 1106/2703 [00:23<00:24, 64.55it/s]


dev:  41%|████      | 1113/2703 [00:23<00:24, 64.03it/s]


dev:  41%|████▏     | 1120/2703 [00:23<00:24, 64.88it/s]


dev:  42%|████▏     | 1127/2703 [00:23<00:24, 65.62it/s]


dev:  42%|████▏     | 1134/2703 [00:23<00:24, 65.01it/s]


dev:  42%|████▏     | 1142/2703 [00:23<00:23, 66.89it/s]


dev:  43%|████▎     | 1149/2703 [00:23<00:24, 64.05it/s]


dev:  43%|████▎     | 1156/2703 [00:23<00:23, 65.11it/s]


dev:  43%|████▎     | 1163/2703 [00:24<00:23, 65.21it/s]


dev:  43%|████▎     | 1170/2703 [00:24<00:24, 63.84it/s]


dev:  44%|████▎     | 1177/2703 [00:24<00:23, 64.24it/s]


dev:  44%|████▍     | 1184/2703 [00:24<00:23, 65.35it/s]


dev:  44%|████▍     | 1191/2703 [00:24<00:22, 65.80it/s]


dev:  44%|████▍     | 1198/2703 [00:24<00:23, 64.72it/s]


dev:  45%|████▍     | 1206/2703 [00:24<00:22, 66.81it/s]


dev:  45%|████▍     | 1213/2703 [00:24<00:22, 65.94it/s]


dev:  45%|████▌     | 1220/2703 [00:24<00:22, 64.69it/s]


dev:  45%|████▌     | 1228/2703 [00:25<00:22, 66.43it/s]


dev:  46%|████▌     | 1236/2703 [00:25<00:21, 67.84it/s]


dev:  46%|████▌     | 1243/2703 [00:25<00:21, 68.05it/s]


dev:  46%|████▌     | 1250/2703 [00:25<00:21, 66.20it/s]


dev:  47%|████▋     | 1257/2703 [00:25<00:21, 67.08it/s]


dev:  47%|████▋     | 1265/2703 [00:25<00:21, 68.29it/s]


dev:  47%|████▋     | 1272/2703 [00:25<00:21, 67.75it/s]


dev:  47%|████▋     | 1280/2703 [00:25<00:20, 69.06it/s]


dev:  48%|████▊     | 1287/2703 [00:25<00:20, 69.19it/s]


dev:  48%|████▊     | 1294/2703 [00:25<00:20, 69.19it/s]


dev:  48%|████▊     | 1301/2703 [00:26<00:20, 69.28it/s]


dev:  48%|████▊     | 1309/2703 [00:26<00:19, 70.04it/s]


dev:  49%|████▊     | 1317/2703 [00:26<00:19, 70.56it/s]


dev:  49%|████▉     | 1325/2703 [00:26<00:19, 69.36it/s]


dev:  49%|████▉     | 1332/2703 [00:26<00:19, 69.08it/s]


dev:  50%|████▉     | 1340/2703 [00:26<00:19, 69.61it/s]


dev:  50%|████▉     | 1347/2703 [00:26<00:19, 68.63it/s]


dev:  50%|█████     | 1354/2703 [00:26<00:19, 68.74it/s]


dev:  50%|█████     | 1362/2703 [00:26<00:19, 70.35it/s]


dev:  51%|█████     | 1370/2703 [00:27<00:19, 69.18it/s]


dev:  51%|█████     | 1377/2703 [00:27<00:19, 68.03it/s]


dev:  51%|█████     | 1384/2703 [00:27<00:19, 67.67it/s]


dev:  51%|█████▏    | 1392/2703 [00:27<00:18, 69.06it/s]


dev:  52%|█████▏    | 1400/2703 [00:27<00:18, 69.93it/s]


dev:  52%|█████▏    | 1407/2703 [00:27<00:19, 67.75it/s]


dev:  52%|█████▏    | 1414/2703 [00:27<00:19, 67.14it/s]


dev:  53%|█████▎    | 1422/2703 [00:27<00:18, 68.99it/s]


dev:  53%|█████▎    | 1430/2703 [00:27<00:18, 67.92it/s]


dev:  53%|█████▎    | 1438/2703 [00:28<00:18, 69.23it/s]


dev:  53%|█████▎    | 1446/2703 [00:28<00:17, 70.15it/s]


dev:  54%|█████▍    | 1454/2703 [00:28<00:17, 70.02it/s]


dev:  54%|█████▍    | 1462/2703 [00:28<00:17, 69.81it/s]


dev:  54%|█████▍    | 1470/2703 [00:28<00:17, 70.17it/s]


dev:  55%|█████▍    | 1478/2703 [00:28<00:17, 70.38it/s]


dev:  55%|█████▍    | 1486/2703 [00:28<00:17, 69.52it/s]


dev:  55%|█████▌    | 1494/2703 [00:28<00:17, 70.09it/s]


dev:  56%|█████▌    | 1502/2703 [00:28<00:17, 69.97it/s]


dev:  56%|█████▌    | 1509/2703 [00:29<00:17, 68.33it/s]


dev:  56%|█████▌    | 1517/2703 [00:29<00:16, 70.02it/s]


dev:  56%|█████▋    | 1525/2703 [00:29<00:16, 69.57it/s]


dev:  57%|█████▋    | 1533/2703 [00:29<00:16, 70.60it/s]


dev:  57%|█████▋    | 1541/2703 [00:29<00:16, 69.51it/s]


dev:  57%|█████▋    | 1549/2703 [00:29<00:16, 70.67it/s]


dev:  58%|█████▊    | 1557/2703 [00:29<00:16, 71.24it/s]


dev:  58%|█████▊    | 1565/2703 [00:29<00:16, 69.94it/s]


dev:  58%|█████▊    | 1573/2703 [00:29<00:15, 70.70it/s]


dev:  58%|█████▊    | 1581/2703 [00:30<00:16, 69.86it/s]


dev:  59%|█████▉    | 1589/2703 [00:30<00:15, 70.32it/s]


dev:  59%|█████▉    | 1597/2703 [00:30<00:15, 70.63it/s]


dev:  59%|█████▉    | 1605/2703 [00:30<00:15, 71.56it/s]


dev:  60%|█████▉    | 1613/2703 [00:30<00:15, 71.23it/s]


dev:  60%|█████▉    | 1621/2703 [00:30<00:15, 70.08it/s]


dev:  60%|██████    | 1629/2703 [00:30<00:15, 71.34it/s]


dev:  61%|██████    | 1637/2703 [00:30<00:14, 71.53it/s]


dev:  61%|██████    | 1645/2703 [00:30<00:14, 70.79it/s]


dev:  61%|██████    | 1653/2703 [00:31<00:14, 71.84it/s]


dev:  61%|██████▏   | 1661/2703 [00:31<00:14, 70.82it/s]


dev:  62%|██████▏   | 1669/2703 [00:31<00:14, 71.78it/s]


dev:  62%|██████▏   | 1677/2703 [00:31<00:14, 69.17it/s]


dev:  62%|██████▏   | 1685/2703 [00:31<00:14, 70.44it/s]


dev:  63%|██████▎   | 1693/2703 [00:31<00:14, 70.49it/s]


dev:  63%|██████▎   | 1701/2703 [00:31<00:14, 69.85it/s]


dev:  63%|██████▎   | 1709/2703 [00:31<00:14, 70.56it/s]


dev:  64%|██████▎   | 1717/2703 [00:32<00:14, 70.30it/s]


dev:  64%|██████▍   | 1725/2703 [00:32<00:13, 70.10it/s]


dev:  64%|██████▍   | 1733/2703 [00:32<00:13, 69.60it/s]


dev:  64%|██████▍   | 1741/2703 [00:32<00:13, 70.90it/s]


dev:  65%|██████▍   | 1749/2703 [00:32<00:13, 70.34it/s]


dev:  65%|██████▌   | 1757/2703 [00:32<00:13, 70.04it/s]


dev:  65%|██████▌   | 1765/2703 [00:32<00:13, 71.11it/s]


dev:  66%|██████▌   | 1773/2703 [00:32<00:13, 70.78it/s]


dev:  66%|██████▌   | 1781/2703 [00:32<00:13, 70.06it/s]


dev:  66%|██████▌   | 1789/2703 [00:33<00:12, 70.90it/s]


dev:  66%|██████▋   | 1797/2703 [00:33<00:12, 72.01it/s]


dev:  67%|██████▋   | 1805/2703 [00:33<00:12, 72.53it/s]


dev:  67%|██████▋   | 1813/2703 [00:33<00:12, 70.03it/s]


dev:  67%|██████▋   | 1821/2703 [00:33<00:12, 71.57it/s]


dev:  68%|██████▊   | 1829/2703 [00:33<00:12, 72.07it/s]


dev:  68%|██████▊   | 1837/2703 [00:33<00:12, 71.56it/s]


dev:  68%|██████▊   | 1845/2703 [00:33<00:12, 71.15it/s]


dev:  69%|██████▊   | 1853/2703 [00:33<00:11, 72.10it/s]


dev:  69%|██████▉   | 1861/2703 [00:34<00:12, 69.13it/s]


dev:  69%|██████▉   | 1869/2703 [00:34<00:11, 70.51it/s]


dev:  69%|██████▉   | 1877/2703 [00:34<00:11, 71.22it/s]


dev:  70%|██████▉   | 1885/2703 [00:34<00:11, 71.17it/s]


dev:  70%|███████   | 1893/2703 [00:34<00:11, 71.06it/s]


dev:  70%|███████   | 1901/2703 [00:34<00:11, 70.56it/s]


dev:  71%|███████   | 1909/2703 [00:34<00:11, 71.68it/s]


dev:  71%|███████   | 1917/2703 [00:34<00:10, 71.69it/s]


dev:  71%|███████   | 1925/2703 [00:34<00:11, 70.52it/s]


dev:  72%|███████▏  | 1933/2703 [00:35<00:10, 71.60it/s]


dev:  72%|███████▏  | 1941/2703 [00:35<00:10, 72.14it/s]


dev:  72%|███████▏  | 1949/2703 [00:35<00:10, 70.10it/s]


dev:  72%|███████▏  | 1957/2703 [00:35<00:10, 69.84it/s]


dev:  73%|███████▎  | 1965/2703 [00:35<00:10, 71.31it/s]


dev:  73%|███████▎  | 1973/2703 [00:35<00:10, 71.25it/s]


dev:  73%|███████▎  | 1981/2703 [00:35<00:10, 71.91it/s]


dev:  74%|███████▎  | 1989/2703 [00:35<00:09, 72.94it/s]


dev:  74%|███████▍  | 1997/2703 [00:35<00:09, 71.23it/s]


dev:  74%|███████▍  | 2005/2703 [00:36<00:09, 72.22it/s]


dev:  74%|███████▍  | 2013/2703 [00:36<00:09, 73.31it/s]


dev:  75%|███████▍  | 2021/2703 [00:36<00:09, 72.76it/s]


dev:  75%|███████▌  | 2029/2703 [00:36<00:09, 71.88it/s]


dev:  75%|███████▌  | 2037/2703 [00:36<00:09, 71.18it/s]


dev:  76%|███████▌  | 2045/2703 [00:36<00:09, 70.37it/s]


dev:  76%|███████▌  | 2053/2703 [00:36<00:09, 70.44it/s]


dev:  76%|███████▌  | 2061/2703 [00:36<00:09, 69.57it/s]


dev:  77%|███████▋  | 2069/2703 [00:36<00:08, 70.78it/s]


dev:  77%|███████▋  | 2077/2703 [00:37<00:09, 69.15it/s]


dev:  77%|███████▋  | 2085/2703 [00:37<00:08, 70.78it/s]


dev:  77%|███████▋  | 2093/2703 [00:37<00:08, 71.74it/s]


dev:  78%|███████▊  | 2101/2703 [00:37<00:08, 72.44it/s]


dev:  78%|███████▊  | 2109/2703 [00:37<00:08, 72.91it/s]


dev:  78%|███████▊  | 2117/2703 [00:37<00:08, 72.37it/s]


dev:  79%|███████▊  | 2125/2703 [00:37<00:07, 73.05it/s]


dev:  79%|███████▉  | 2133/2703 [00:37<00:07, 73.29it/s]


dev:  79%|███████▉  | 2141/2703 [00:37<00:07, 72.42it/s]


dev:  80%|███████▉  | 2149/2703 [00:38<00:07, 72.93it/s]


dev:  80%|███████▉  | 2157/2703 [00:38<00:07, 73.12it/s]


dev:  80%|████████  | 2165/2703 [00:38<00:07, 72.84it/s]


dev:  80%|████████  | 2173/2703 [00:38<00:07, 71.48it/s]


dev:  81%|████████  | 2181/2703 [00:38<00:07, 72.90it/s]


dev:  81%|████████  | 2189/2703 [00:38<00:07, 70.80it/s]


dev:  81%|████████▏ | 2197/2703 [00:38<00:07, 71.96it/s]


dev:  82%|████████▏ | 2205/2703 [00:38<00:06, 73.03it/s]


dev:  82%|████████▏ | 2213/2703 [00:38<00:06, 71.82it/s]


dev:  82%|████████▏ | 2221/2703 [00:39<00:06, 71.03it/s]


dev:  82%|████████▏ | 2229/2703 [00:39<00:06, 71.34it/s]


dev:  83%|████████▎ | 2237/2703 [00:39<00:06, 72.58it/s]


dev:  83%|████████▎ | 2245/2703 [00:39<00:06, 70.90it/s]


dev:  83%|████████▎ | 2253/2703 [00:39<00:06, 72.33it/s]


dev:  84%|████████▎ | 2261/2703 [00:39<00:06, 73.48it/s]


dev:  84%|████████▍ | 2269/2703 [00:39<00:05, 73.45it/s]


dev:  84%|████████▍ | 2277/2703 [00:39<00:05, 73.36it/s]


dev:  85%|████████▍ | 2285/2703 [00:39<00:05, 73.86it/s]


dev:  85%|████████▍ | 2293/2703 [00:40<00:05, 72.59it/s]


dev:  85%|████████▌ | 2301/2703 [00:40<00:05, 71.34it/s]


dev:  85%|████████▌ | 2309/2703 [00:40<00:05, 73.10it/s]


dev:  86%|████████▌ | 2317/2703 [00:40<00:05, 74.18it/s]


dev:  86%|████████▌ | 2325/2703 [00:40<00:05, 73.39it/s]


dev:  86%|████████▋ | 2333/2703 [00:40<00:05, 72.45it/s]


dev:  87%|████████▋ | 2341/2703 [00:40<00:04, 73.30it/s]


dev:  87%|████████▋ | 2349/2703 [00:40<00:04, 71.55it/s]


dev:  87%|████████▋ | 2357/2703 [00:40<00:04, 71.06it/s]


dev:  87%|████████▋ | 2365/2703 [00:41<00:04, 72.32it/s]


dev:  88%|████████▊ | 2373/2703 [00:41<00:04, 71.47it/s]


dev:  88%|████████▊ | 2381/2703 [00:41<00:04, 72.04it/s]


dev:  88%|████████▊ | 2389/2703 [00:41<00:04, 71.45it/s]


dev:  89%|████████▊ | 2397/2703 [00:41<00:04, 72.81it/s]


dev:  89%|████████▉ | 2406/2703 [00:41<00:03, 75.32it/s]


dev:  89%|████████▉ | 2414/2703 [00:41<00:03, 75.20it/s]


dev:  90%|████████▉ | 2422/2703 [00:41<00:03, 75.45it/s]


dev:  90%|████████▉ | 2430/2703 [00:41<00:03, 75.70it/s]


dev:  90%|█████████ | 2438/2703 [00:42<00:03, 75.88it/s]


dev:  90%|█████████ | 2446/2703 [00:42<00:03, 74.56it/s]


dev:  91%|█████████ | 2455/2703 [00:42<00:03, 76.25it/s]


dev:  91%|█████████ | 2463/2703 [00:42<00:03, 76.59it/s]


dev:  91%|█████████▏| 2471/2703 [00:42<00:03, 74.64it/s]


dev:  92%|█████████▏| 2479/2703 [00:42<00:02, 75.23it/s]


dev:  92%|█████████▏| 2487/2703 [00:42<00:02, 76.07it/s]


dev:  92%|█████████▏| 2495/2703 [00:42<00:02, 75.87it/s]


dev:  93%|█████████▎| 2503/2703 [00:42<00:02, 75.42it/s]


dev:  93%|█████████▎| 2511/2703 [00:42<00:02, 75.50it/s]


dev:  93%|█████████▎| 2519/2703 [00:43<00:02, 75.84it/s]


dev:  93%|█████████▎| 2527/2703 [00:43<00:02, 76.38it/s]


dev:  94%|█████████▍| 2535/2703 [00:43<00:02, 75.59it/s]


dev:  94%|█████████▍| 2543/2703 [00:43<00:02, 76.08it/s]


dev:  94%|█████████▍| 2551/2703 [00:43<00:02, 75.22it/s]


dev:  95%|█████████▍| 2559/2703 [00:43<00:01, 76.58it/s]


dev:  95%|█████████▌| 2568/2703 [00:43<00:01, 78.95it/s]


dev:  95%|█████████▌| 2576/2703 [00:43<00:01, 77.88it/s]


dev:  96%|█████████▌| 2585/2703 [00:43<00:01, 79.00it/s]


dev:  96%|█████████▌| 2593/2703 [00:44<00:01, 75.31it/s]


dev:  96%|█████████▌| 2601/2703 [00:44<00:01, 76.26it/s]


dev:  97%|█████████▋| 2609/2703 [00:44<00:01, 76.03it/s]


dev:  97%|█████████▋| 2617/2703 [00:44<00:01, 72.94it/s]


dev:  97%|█████████▋| 2625/2703 [00:44<00:01, 74.50it/s]


dev:  97%|█████████▋| 2633/2703 [00:44<00:00, 75.58it/s]


dev:  98%|█████████▊| 2641/2703 [00:44<00:00, 75.69it/s]


dev:  98%|█████████▊| 2649/2703 [00:44<00:00, 75.02it/s]


dev:  98%|█████████▊| 2658/2703 [00:44<00:00, 77.69it/s]


dev:  99%|█████████▊| 2666/2703 [00:45<00:00, 76.92it/s]


dev:  99%|█████████▉| 2676/2703 [00:45<00:00, 81.77it/s]


dev:  99%|█████████▉| 2685/2703 [00:45<00:00, 81.57it/s]


dev: 100%|█████████▉| 2694/2703 [00:45<00:00, 82.44it/s]
dev: 100%|██████████| 2703/2703 [00:45<00:00, 59.27it/s]
dev loss: 0.7754477858543396
dev per: 0.19571800208116547

                                                            

                                                          
[Runner] - Save the checkpoint to:

overall:  50%|████▉     | 1999/4000 [10:25<08:06,  4.11it/s]

train:  57%|█████▋    | 1999/3515 [10:22<06:08,  4.11it/s]
                                                            

                                                          
1. result/downstream/grid/dev-best.ckpt

overall:  50%|████▉     | 1999/4000 [10:25<08:06,  4.11it/s]

train:  57%|█████▋    | 1999/3515 [10:22<06:08,  4.11it/s]
overall:  50%|█████     | 2000/4000 [10:25<7:21:33, 13.25s/it]

train:  57%|█████▋    | 2000/3515 [10:22<5:34:28, 13.25s/it]
overall:  50%|█████     | 2001/4000 [10:25<5:22:16,  9.67s/it]

train:  57%|█████▋    | 2001/3515 [10:23<4:04:05,  9.67s/it]
overall:  50%|█████     | 2002/4000 [10:25<3:51:30,  6.95s/it]

train:  57%|█████▋    | 2002/3515 [10:23<2:55:18,  6.95s/it]
overall:  50%|█████     | 2003/4000 [10:26<2:46:48,  5.01s/it]

train:  57%|█████▋    | 2003/3515 [10:23<2:06:17,  5.01s/it]
overall:  50%|█████     | 2004/4000 [10:26<2:00:38,  3.63s/it]

train:  57%|█████▋    | 2004/3515 [10:23<1:31:19,  3.63s/it]
overall:  50%|█████     | 2005/4000 [10:26<1:27:35,  2.63s/it]

train:  57%|█████▋    | 2005/3515 [10:24<1:06:17,  2.63s/it]
overall:  50%|█████     | 2006/4000 [10:26<1:04:15,  1.93s/it]

train:  57%|█████▋    | 2006/3515 [10:24<48:38,  1.93s/it]  
overall:  50%|█████     | 2007/4000 [10:27<47:18,  1.42s/it]  

train:  57%|█████▋    | 2007/3515 [10:24<35:47,  1.42s/it]
overall:  50%|█████     | 2008/4000 [10:27<35:47,  1.08s/it]

train:  57%|█████▋    | 2008/3515 [10:24<27:04,  1.08s/it]
overall:  50%|█████     | 2009/4000 [10:27<27:47,  1.19it/s]

train:  57%|█████▋    | 2009/3515 [10:25<21:01,  1.19it/s]
overall:  50%|█████     | 2010/4000 [10:27<22:11,  1.49it/s]

train:  57%|█████▋    | 2010/3515 [10:25<16:47,  1.49it/s]
overall:  50%|█████     | 2011/4000 [10:28<17:55,  1.85it/s]

train:  57%|█████▋    | 2011/3515 [10:25<13:33,  1.85it/s]
overall:  50%|█████     | 2012/4000 [10:28<16:37,  1.99it/s]

train:  57%|█████▋    | 2012/3515 [10:25<12:34,  1.99it/s]
overall:  50%|█████     | 2013/4000 [10:28<14:15,  2.32it/s]

train:  57%|█████▋    | 2013/3515 [10:26<10:46,  2.32it/s]
overall:  50%|█████     | 2014/4000 [10:29<12:35,  2.63it/s]

train:  57%|█████▋    | 2014/3515 [10:26<09:30,  2.63it/s]
overall:  50%|█████     | 2015/4000 [10:29<11:25,  2.89it/s]

train:  57%|█████▋    | 2015/3515 [10:26<08:38,  2.89it/s]
overall:  50%|█████     | 2016/4000 [10:29<10:41,  3.09it/s]

train:  57%|█████▋    | 2016/3515 [10:27<08:05,  3.09it/s]
overall:  50%|█████     | 2017/4000 [10:29<10:05,  3.28it/s]

train:  57%|█████▋    | 2017/3515 [10:27<07:37,  3.28it/s]
overall:  50%|█████     | 2018/4000 [10:30<09:41,  3.41it/s]

train:  57%|█████▋    | 2018/3515 [10:27<07:19,  3.41it/s]
overall:  50%|█████     | 2019/4000 [10:30<09:24,  3.51it/s]

train:  57%|█████▋    | 2019/3515 [10:27<07:06,  3.51it/s]train loss: 0.5881908535957336
train per: 0.17050814871118464

overall:  50%|█████     | 2020/4000 [10:30<09:20,  3.53it/s]

train:  57%|█████▋    | 2020/3515 [10:28<07:03,  3.53it/s]
overall:  51%|█████     | 2021/4000 [10:31<09:09,  3.60it/s]

train:  57%|█████▋    | 2021/3515 [10:28<06:54,  3.60it/s]
overall:  51%|█████     | 2022/4000 [10:31<08:59,  3.66it/s]

train:  58%|█████▊    | 2022/3515 [10:28<06:47,  3.66it/s]
overall:  51%|█████     | 2023/4000 [10:31<08:57,  3.68it/s]

train:  58%|█████▊    | 2023/3515 [10:28<06:45,  3.68it/s]
overall:  51%|█████     | 2024/4000 [10:31<08:36,  3.83it/s]

train:  58%|█████▊    | 2024/3515 [10:29<06:29,  3.83it/s]
overall:  51%|█████     | 2025/4000 [10:32<10:39,  3.09it/s]

train:  58%|█████▊    | 2025/3515 [10:29<08:02,  3.09it/s]
overall:  51%|█████     | 2027/4000 [10:32<08:19,  3.95it/s]

train:  58%|█████▊    | 2027/3515 [10:29<06:16,  3.95it/s]
overall:  51%|█████     | 2028/4000 [10:32<08:34,  3.83it/s]

train:  58%|█████▊    | 2028/3515 [10:30<06:27,  3.83it/s]
overall:  51%|█████     | 2029/4000 [10:33<08:21,  3.93it/s]

train:  58%|█████▊    | 2029/3515 [10:30<06:17,  3.93it/s]
overall:  51%|█████     | 2030/4000 [10:33<08:30,  3.86it/s]

train:  58%|█████▊    | 2030/3515 [10:30<06:24,  3.86it/s]
overall:  51%|█████     | 2031/4000 [10:33<08:12,  4.00it/s]

train:  58%|█████▊    | 2031/3515 [10:30<06:10,  4.00it/s]
overall:  51%|█████     | 2032/4000 [10:33<08:19,  3.94it/s]

train:  58%|█████▊    | 2032/3515 [10:31<06:16,  3.94it/s]
overall:  51%|█████     | 2033/4000 [10:34<08:21,  3.92it/s]

train:  58%|█████▊    | 2033/3515 [10:31<06:17,  3.92it/s]
overall:  51%|█████     | 2034/4000 [10:34<08:29,  3.86it/s]

train:  58%|█████▊    | 2034/3515 [10:31<06:24,  3.86it/s]
overall:  51%|█████     | 2035/4000 [10:34<08:19,  3.94it/s]

train:  58%|█████▊    | 2035/3515 [10:32<06:16,  3.94it/s]
overall:  51%|█████     | 2036/4000 [10:34<07:12,  4.54it/s]

train:  58%|█████▊    | 2036/3515 [10:32<05:25,  4.54it/s]
overall:  51%|█████     | 2037/4000 [10:35<07:46,  4.21it/s]

train:  58%|█████▊    | 2037/3515 [10:32<05:51,  4.21it/s]
overall:  51%|█████     | 2039/4000 [10:35<06:57,  4.70it/s]

train:  58%|█████▊    | 2039/3515 [10:32<05:13,  4.70it/s]train loss: 0.5876681804656982
train per: 0.1690801653521341

overall:  51%|█████     | 2040/4000 [10:35<07:16,  4.49it/s]

train:  58%|█████▊    | 2040/3515 [10:33<05:28,  4.49it/s]
overall:  51%|█████     | 2041/4000 [10:35<07:33,  4.32it/s]

train:  58%|█████▊    | 2041/3515 [10:33<05:41,  4.32it/s]
overall:  51%|█████     | 2042/4000 [10:36<07:49,  4.17it/s]

train:  58%|█████▊    | 2042/3515 [10:33<05:53,  4.17it/s]
overall:  51%|█████     | 2043/4000 [10:36<08:02,  4.06it/s]

train:  58%|█████▊    | 2043/3515 [10:33<06:02,  4.06it/s]
overall:  51%|█████     | 2044/4000 [10:36<08:21,  3.90it/s]

train:  58%|█████▊    | 2044/3515 [10:34<06:16,  3.90it/s]
overall:  51%|█████     | 2045/4000 [10:36<08:11,  3.98it/s]

train:  58%|█████▊    | 2045/3515 [10:34<06:09,  3.98it/s]
overall:  51%|█████     | 2046/4000 [10:37<08:19,  3.91it/s]

train:  58%|█████▊    | 2046/3515 [10:34<06:15,  3.91it/s]
overall:  51%|█████     | 2047/4000 [10:37<08:00,  4.07it/s]

train:  58%|█████▊    | 2047/3515 [10:34<06:00,  4.07it/s]
overall:  51%|█████     | 2048/4000 [10:37<08:12,  3.96it/s]

train:  58%|█████▊    | 2048/3515 [10:35<06:10,  3.96it/s]
overall:  51%|█████     | 2049/4000 [10:38<10:09,  3.20it/s]

train:  58%|█████▊    | 2049/3515 [10:35<07:37,  3.20it/s]
overall:  51%|█████▏    | 2050/4000 [10:38<09:13,  3.52it/s]

train:  58%|█████▊    | 2050/3515 [10:35<06:55,  3.52it/s]
overall:  51%|█████▏    | 2051/4000 [10:38<08:59,  3.61it/s]

train:  58%|█████▊    | 2051/3515 [10:36<06:45,  3.61it/s]
overall:  51%|█████▏    | 2052/4000 [10:38<08:51,  3.66it/s]

train:  58%|█████▊    | 2052/3515 [10:36<06:39,  3.66it/s]
overall:  51%|█████▏    | 2053/4000 [10:39<08:52,  3.66it/s]

train:  58%|█████▊    | 2053/3515 [10:36<06:39,  3.66it/s]
overall:  51%|█████▏    | 2054/4000 [10:39<07:31,  4.31it/s]

train:  58%|█████▊    | 2054/3515 [10:36<05:38,  4.31it/s]
overall:  51%|█████▏    | 2055/4000 [10:39<07:32,  4.30it/s]

train:  58%|█████▊    | 2055/3515 [10:36<05:39,  4.30it/s]
overall:  51%|█████▏    | 2056/4000 [10:39<07:47,  4.16it/s]

train:  58%|█████▊    | 2056/3515 [10:37<05:50,  4.16it/s]
overall:  51%|█████▏    | 2057/4000 [10:40<07:57,  4.07it/s]

train:  59%|█████▊    | 2057/3515 [10:37<05:58,  4.07it/s]
overall:  51%|█████▏    | 2058/4000 [10:40<07:32,  4.30it/s]

train:  59%|█████▊    | 2058/3515 [10:37<05:39,  4.30it/s]
overall:  51%|█████▏    | 2059/4000 [10:40<07:37,  4.24it/s]

train:  59%|█████▊    | 2059/3515 [10:37<05:43,  4.24it/s]train loss: 0.5917518138885498
train per: 0.170830867406016

overall:  52%|█████▏    | 2060/4000 [10:40<08:55,  3.62it/s]

train:  59%|█████▊    | 2060/3515 [10:38<06:41,  3.62it/s]
overall:  52%|█████▏    | 2061/4000 [10:41<09:02,  3.57it/s]

train:  59%|█████▊    | 2061/3515 [10:38<06:46,  3.57it/s]
overall:  52%|█████▏    | 2062/4000 [10:41<08:57,  3.61it/s]

train:  59%|█████▊    | 2062/3515 [10:38<06:42,  3.61it/s]
overall:  52%|█████▏    | 2063/4000 [10:41<08:47,  3.67it/s]

train:  59%|█████▊    | 2063/3515 [10:39<06:35,  3.67it/s]
overall:  52%|█████▏    | 2064/4000 [10:41<08:45,  3.68it/s]

train:  59%|█████▊    | 2064/3515 [10:39<06:34,  3.68it/s]
overall:  52%|█████▏    | 2065/4000 [10:42<08:27,  3.82it/s]

train:  59%|█████▊    | 2065/3515 [10:39<06:19,  3.82it/s]
overall:  52%|█████▏    | 2066/4000 [10:42<08:30,  3.79it/s]

train:  59%|█████▉    | 2066/3515 [10:39<06:22,  3.79it/s]
overall:  52%|█████▏    | 2067/4000 [10:42<08:29,  3.79it/s]

train:  59%|█████▉    | 2067/3515 [10:40<06:21,  3.79it/s]
overall:  52%|█████▏    | 2068/4000 [10:43<08:17,  3.88it/s]

train:  59%|█████▉    | 2068/3515 [10:40<06:12,  3.88it/s]
overall:  52%|█████▏    | 2069/4000 [10:43<08:26,  3.81it/s]

train:  59%|█████▉    | 2069/3515 [10:40<06:19,  3.81it/s]
overall:  52%|█████▏    | 2070/4000 [10:43<08:35,  3.75it/s]

train:  59%|█████▉    | 2070/3515 [10:40<06:25,  3.75it/s]
overall:  52%|█████▏    | 2071/4000 [10:43<08:35,  3.74it/s]

train:  59%|█████▉    | 2071/3515 [10:41<06:25,  3.74it/s]
overall:  52%|█████▏    | 2072/4000 [10:44<08:39,  3.71it/s]

train:  59%|█████▉    | 2072/3515 [10:41<06:28,  3.71it/s]
overall:  52%|█████▏    | 2073/4000 [10:44<10:21,  3.10it/s]

train:  59%|█████▉    | 2073/3515 [10:41<07:44,  3.10it/s]
overall:  52%|█████▏    | 2074/4000 [10:44<09:45,  3.29it/s]

train:  59%|█████▉    | 2074/3515 [10:42<07:18,  3.29it/s]
overall:  52%|█████▏    | 2075/4000 [10:45<09:08,  3.51it/s]

train:  59%|█████▉    | 2075/3515 [10:42<06:50,  3.51it/s]
overall:  52%|█████▏    | 2076/4000 [10:45<09:07,  3.52it/s]

train:  59%|█████▉    | 2076/3515 [10:42<06:49,  3.52it/s]
overall:  52%|█████▏    | 2077/4000 [10:45<08:50,  3.63it/s]

train:  59%|█████▉    | 2077/3515 [10:42<06:36,  3.63it/s]
overall:  52%|█████▏    | 2078/4000 [10:45<08:38,  3.70it/s]

train:  59%|█████▉    | 2078/3515 [10:43<06:27,  3.70it/s]
overall:  52%|█████▏    | 2079/4000 [10:46<08:21,  3.83it/s]

train:  59%|█████▉    | 2079/3515 [10:43<06:14,  3.83it/s]train loss: 0.5918160080909729
train per: 0.1692955260358913

overall:  52%|█████▏    | 2080/4000 [10:46<08:34,  3.73it/s]

train:  59%|█████▉    | 2080/3515 [10:43<06:24,  3.73it/s]
overall:  52%|█████▏    | 2081/4000 [10:46<08:28,  3.77it/s]

train:  59%|█████▉    | 2081/3515 [10:44<06:20,  3.77it/s]
overall:  52%|█████▏    | 2082/4000 [10:46<08:27,  3.78it/s]

train:  59%|█████▉    | 2082/3515 [10:44<06:18,  3.78it/s]
overall:  52%|█████▏    | 2083/4000 [10:47<07:57,  4.01it/s]

train:  59%|█████▉    | 2083/3515 [10:44<05:56,  4.01it/s]
overall:  52%|█████▏    | 2084/4000 [10:47<08:21,  3.82it/s]

train:  59%|█████▉    | 2084/3515 [10:44<06:14,  3.82it/s]
overall:  52%|█████▏    | 2085/4000 [10:47<08:30,  3.75it/s]

train:  59%|█████▉    | 2085/3515 [10:45<06:21,  3.75it/s]
overall:  52%|█████▏    | 2086/4000 [10:47<08:02,  3.96it/s]

train:  59%|█████▉    | 2086/3515 [10:45<06:00,  3.96it/s]
overall:  52%|█████▏    | 2087/4000 [10:48<08:20,  3.82it/s]

train:  59%|█████▉    | 2087/3515 [10:45<06:13,  3.82it/s]
overall:  52%|█████▏    | 2089/4000 [10:48<05:46,  5.51it/s]

train:  59%|█████▉    | 2089/3515 [10:45<04:18,  5.51it/s]
overall:  52%|█████▏    | 2090/4000 [10:48<06:24,  4.97it/s]

train:  59%|█████▉    | 2090/3515 [10:45<04:46,  4.97it/s]
overall:  52%|█████▏    | 2091/4000 [10:48<06:41,  4.76it/s]

train:  59%|█████▉    | 2091/3515 [10:46<04:59,  4.76it/s]
overall:  52%|█████▏    | 2092/4000 [10:49<07:12,  4.41it/s]

train:  60%|█████▉    | 2092/3515 [10:46<05:22,  4.41it/s]
overall:  52%|█████▏    | 2093/4000 [10:49<07:36,  4.18it/s]

train:  60%|█████▉    | 2093/3515 [10:46<05:40,  4.18it/s]
overall:  52%|█████▏    | 2094/4000 [10:49<07:51,  4.05it/s]

train:  60%|█████▉    | 2094/3515 [10:47<05:51,  4.05it/s]
overall:  52%|█████▏    | 2095/4000 [10:49<08:01,  3.95it/s]

train:  60%|█████▉    | 2095/3515 [10:47<05:59,  3.95it/s]
overall:  52%|█████▏    | 2096/4000 [10:50<08:08,  3.90it/s]

train:  60%|█████▉    | 2096/3515 [10:47<06:04,  3.90it/s]
overall:  52%|█████▏    | 2097/4000 [10:50<10:11,  3.11it/s]

train:  60%|█████▉    | 2097/3515 [10:48<07:35,  3.11it/s]
overall:  52%|█████▏    | 2098/4000 [10:50<09:24,  3.37it/s]

train:  60%|█████▉    | 2098/3515 [10:48<07:00,  3.37it/s]
overall:  52%|█████▏    | 2099/4000 [10:51<09:01,  3.51it/s]

train:  60%|█████▉    | 2099/3515 [10:48<06:43,  3.51it/s]train loss: 0.5601674318313599
train per: 0.16281462939785843

overall:  52%|█████▎    | 2100/4000 [10:51<07:43,  4.10it/s]

train:  60%|█████▉    | 2100/3515 [10:48<05:45,  4.09it/s]
overall:  53%|█████▎    | 2101/4000 [10:51<06:27,  4.89it/s]

train:  60%|█████▉    | 2101/3515 [10:48<04:48,  4.90it/s]
overall:  53%|█████▎    | 2102/4000 [10:51<05:29,  5.77it/s]

train:  60%|█████▉    | 2102/3515 [10:48<04:04,  5.77it/s]
overall:  53%|█████▎    | 2103/4000 [10:51<06:08,  5.15it/s]

train:  60%|█████▉    | 2103/3515 [10:49<04:34,  5.15it/s]
overall:  53%|█████▎    | 2104/4000 [10:52<06:45,  4.67it/s]

train:  60%|█████▉    | 2104/3515 [10:49<05:01,  4.67it/s]
overall:  53%|█████▎    | 2105/4000 [10:52<07:21,  4.30it/s]

train:  60%|█████▉    | 2105/3515 [10:49<05:28,  4.30it/s]
overall:  53%|█████▎    | 2106/4000 [10:52<06:24,  4.92it/s]

train:  60%|█████▉    | 2106/3515 [10:49<04:46,  4.92it/s]
overall:  53%|█████▎    | 2107/4000 [10:52<07:04,  4.46it/s]

train:  60%|█████▉    | 2107/3515 [10:50<05:15,  4.46it/s]
overall:  53%|█████▎    | 2108/4000 [10:53<08:22,  3.77it/s]

train:  60%|█████▉    | 2108/3515 [10:50<06:13,  3.77it/s]
overall:  53%|█████▎    | 2109/4000 [10:53<08:22,  3.76it/s]

train:  60%|██████    | 2109/3515 [10:50<06:13,  3.76it/s]
overall:  53%|█████▎    | 2110/4000 [10:53<08:15,  3.81it/s]

train:  60%|██████    | 2110/3515 [10:50<06:08,  3.81it/s]
overall:  53%|█████▎    | 2111/4000 [10:53<08:11,  3.85it/s]

train:  60%|██████    | 2111/3515 [10:51<06:05,  3.85it/s]
overall:  53%|█████▎    | 2112/4000 [10:54<07:53,  3.99it/s]

train:  60%|██████    | 2112/3515 [10:51<05:52,  3.99it/s]
overall:  53%|█████▎    | 2113/4000 [10:54<07:24,  4.24it/s]

train:  60%|██████    | 2113/3515 [10:51<05:30,  4.24it/s]
overall:  53%|█████▎    | 2115/4000 [10:54<06:26,  4.88it/s]

train:  60%|██████    | 2115/3515 [10:51<04:46,  4.88it/s]
overall:  53%|█████▎    | 2116/4000 [10:54<06:59,  4.49it/s]

train:  60%|██████    | 2116/3515 [10:52<05:11,  4.49it/s]
overall:  53%|█████▎    | 2117/4000 [10:55<07:19,  4.29it/s]

train:  60%|██████    | 2117/3515 [10:52<05:26,  4.29it/s]
overall:  53%|█████▎    | 2118/4000 [10:55<07:31,  4.17it/s]

train:  60%|██████    | 2118/3515 [10:52<05:35,  4.17it/s]
overall:  53%|█████▎    | 2119/4000 [10:55<07:37,  4.11it/s]

train:  60%|██████    | 2119/3515 [10:53<05:39,  4.11it/s]train loss: 0.5668290257453918
train per: 0.16387587152817465

overall:  53%|█████▎    | 2120/4000 [10:55<07:58,  3.93it/s]

train:  60%|██████    | 2120/3515 [10:53<05:55,  3.93it/s]
overall:  53%|█████▎    | 2121/4000 [10:56<09:07,  3.43it/s]

train:  60%|██████    | 2121/3515 [10:53<06:45,  3.44it/s]
overall:  53%|█████▎    | 2122/4000 [10:56<08:51,  3.53it/s]

train:  60%|██████    | 2122/3515 [10:53<06:34,  3.53it/s]
overall:  53%|█████▎    | 2123/4000 [10:56<08:17,  3.77it/s]

train:  60%|██████    | 2123/3515 [10:54<06:09,  3.77it/s]
overall:  53%|█████▎    | 2124/4000 [10:57<08:26,  3.70it/s]

train:  60%|██████    | 2124/3515 [10:54<06:15,  3.70it/s]
overall:  53%|█████▎    | 2125/4000 [10:57<08:27,  3.69it/s]

train:  60%|██████    | 2125/3515 [10:54<06:16,  3.69it/s]
overall:  53%|█████▎    | 2126/4000 [10:57<08:10,  3.82it/s]

train:  60%|██████    | 2126/3515 [10:54<06:03,  3.82it/s]
overall:  53%|█████▎    | 2127/4000 [10:57<08:13,  3.79it/s]

train:  61%|██████    | 2127/3515 [10:55<06:05,  3.79it/s]
overall:  53%|█████▎    | 2128/4000 [10:58<08:22,  3.73it/s]

train:  61%|██████    | 2128/3515 [10:55<06:11,  3.73it/s]
overall:  53%|█████▎    | 2129/4000 [10:58<08:23,  3.71it/s]

train:  61%|██████    | 2129/3515 [10:55<06:13,  3.72it/s]
overall:  53%|█████▎    | 2130/4000 [10:58<08:15,  3.78it/s]

train:  61%|██████    | 2130/3515 [10:56<06:06,  3.78it/s]
overall:  53%|█████▎    | 2131/4000 [10:58<08:13,  3.78it/s]

train:  61%|██████    | 2131/3515 [10:56<06:05,  3.78it/s]
overall:  53%|█████▎    | 2132/4000 [10:59<09:24,  3.31it/s]

train:  61%|██████    | 2132/3515 [10:56<06:57,  3.31it/s]
overall:  53%|█████▎    | 2133/4000 [10:59<08:55,  3.49it/s]

train:  61%|██████    | 2133/3515 [10:56<06:36,  3.49it/s]
overall:  53%|█████▎    | 2134/4000 [10:59<08:23,  3.70it/s]

train:  61%|██████    | 2134/3515 [10:57<06:12,  3.70it/s]
overall:  53%|█████▎    | 2135/4000 [11:00<08:23,  3.71it/s]

train:  61%|██████    | 2135/3515 [10:57<06:12,  3.71it/s]
overall:  53%|█████▎    | 2136/4000 [11:00<08:31,  3.65it/s]

train:  61%|██████    | 2136/3515 [10:57<06:18,  3.65it/s]
overall:  53%|█████▎    | 2137/4000 [11:00<08:27,  3.67it/s]

train:  61%|██████    | 2137/3515 [10:58<06:15,  3.67it/s]
overall:  53%|█████▎    | 2138/4000 [11:00<08:24,  3.69it/s]

train:  61%|██████    | 2138/3515 [10:58<06:13,  3.69it/s]
overall:  53%|█████▎    | 2139/4000 [11:01<07:06,  4.37it/s]

train:  61%|██████    | 2139/3515 [10:58<05:15,  4.37it/s]train loss: 0.5917158722877502
train per: 0.17047172664045962

overall:  54%|█████▎    | 2140/4000 [11:01<07:05,  4.37it/s]

train:  61%|██████    | 2140/3515 [10:58<05:14,  4.37it/s]
overall:  54%|█████▎    | 2141/4000 [11:01<07:26,  4.16it/s]

train:  61%|██████    | 2141/3515 [10:58<05:30,  4.16it/s]
overall:  54%|█████▎    | 2143/4000 [11:01<06:03,  5.11it/s]

train:  61%|██████    | 2143/3515 [10:59<04:28,  5.11it/s]
overall:  54%|█████▎    | 2144/4000 [11:02<06:24,  4.83it/s]

train:  61%|██████    | 2144/3515 [10:59<04:43,  4.83it/s]
overall:  54%|█████▎    | 2145/4000 [11:02<08:18,  3.72it/s]

train:  61%|██████    | 2145/3515 [10:59<06:08,  3.72it/s]
overall:  54%|█████▎    | 2146/4000 [11:02<07:28,  4.13it/s]

train:  61%|██████    | 2146/3515 [11:00<05:31,  4.13it/s]
overall:  54%|█████▎    | 2147/4000 [11:02<07:38,  4.04it/s]

train:  61%|██████    | 2147/3515 [11:00<05:38,  4.04it/s]
overall:  54%|█████▎    | 2148/4000 [11:03<07:57,  3.87it/s]

train:  61%|██████    | 2148/3515 [11:00<05:52,  3.88it/s]
overall:  54%|█████▎    | 2149/4000 [11:03<08:04,  3.82it/s]

train:  61%|██████    | 2149/3515 [11:00<05:57,  3.82it/s]
overall:  54%|█████▍    | 2151/4000 [11:03<06:45,  4.56it/s]

train:  61%|██████    | 2151/3515 [11:01<04:58,  4.56it/s]
overall:  54%|█████▍    | 2152/4000 [11:04<06:49,  4.52it/s]

train:  61%|██████    | 2152/3515 [11:01<05:01,  4.52it/s]
overall:  54%|█████▍    | 2153/4000 [11:04<07:10,  4.29it/s]

train:  61%|██████▏   | 2153/3515 [11:01<05:17,  4.29it/s]
overall:  54%|█████▍    | 2154/4000 [11:04<07:28,  4.12it/s]

train:  61%|██████▏   | 2154/3515 [11:01<05:30,  4.12it/s]
overall:  54%|█████▍    | 2155/4000 [11:04<06:38,  4.63it/s]

train:  61%|██████▏   | 2155/3515 [11:02<04:53,  4.63it/s]
overall:  54%|█████▍    | 2156/4000 [11:05<07:33,  4.06it/s]

train:  61%|██████▏   | 2156/3515 [11:02<05:34,  4.06it/s]
overall:  54%|█████▍    | 2157/4000 [11:05<07:35,  4.04it/s]

train:  61%|██████▏   | 2157/3515 [11:02<05:35,  4.04it/s]
overall:  54%|█████▍    | 2158/4000 [11:05<07:41,  3.99it/s]

train:  61%|██████▏   | 2158/3515 [11:02<05:39,  3.99it/s]
overall:  54%|█████▍    | 2159/4000 [11:05<07:55,  3.87it/s]

train:  61%|██████▏   | 2159/3515 [11:03<05:50,  3.87it/s]train loss: 0.5805287957191467
train per: 0.17104913678618858

overall:  54%|█████▍    | 2160/4000 [11:06<08:07,  3.77it/s]

train:  61%|██████▏   | 2160/3515 [11:03<05:59,  3.77it/s]
overall:  54%|█████▍    | 2161/4000 [11:06<08:10,  3.75it/s]

train:  61%|██████▏   | 2161/3515 [11:03<06:01,  3.75it/s]
overall:  54%|█████▍    | 2162/4000 [11:06<08:14,  3.72it/s]

train:  62%|██████▏   | 2162/3515 [11:04<06:03,  3.72it/s]
overall:  54%|█████▍    | 2163/4000 [11:06<07:47,  3.93it/s]

train:  62%|██████▏   | 2163/3515 [11:04<05:44,  3.93it/s]
overall:  54%|█████▍    | 2164/4000 [11:07<06:50,  4.47it/s]

train:  62%|██████▏   | 2164/3515 [11:04<05:01,  4.48it/s]
overall:  54%|█████▍    | 2165/4000 [11:07<05:43,  5.34it/s]

train:  62%|██████▏   | 2165/3515 [11:04<04:12,  5.34it/s]
overall:  54%|█████▍    | 2166/4000 [11:07<06:27,  4.74it/s]

train:  62%|██████▏   | 2166/3515 [11:04<04:44,  4.74it/s]
overall:  54%|█████▍    | 2167/4000 [11:07<06:51,  4.46it/s]

train:  62%|██████▏   | 2167/3515 [11:05<05:02,  4.46it/s]
overall:  54%|█████▍    | 2169/4000 [11:08<07:39,  3.98it/s]

train:  62%|██████▏   | 2169/3515 [11:05<05:37,  3.98it/s]
overall:  54%|█████▍    | 2170/4000 [11:08<07:46,  3.92it/s]

train:  62%|██████▏   | 2170/3515 [11:05<05:42,  3.92it/s]
overall:  54%|█████▍    | 2171/4000 [11:08<07:50,  3.89it/s]

train:  62%|██████▏   | 2171/3515 [11:06<05:45,  3.89it/s]
overall:  54%|█████▍    | 2172/4000 [11:09<07:53,  3.86it/s]

train:  62%|██████▏   | 2172/3515 [11:06<05:48,  3.86it/s]
overall:  54%|█████▍    | 2173/4000 [11:09<07:45,  3.93it/s]

train:  62%|██████▏   | 2173/3515 [11:06<05:41,  3.93it/s]
overall:  54%|█████▍    | 2174/4000 [11:09<07:52,  3.87it/s]

train:  62%|██████▏   | 2174/3515 [11:06<05:46,  3.87it/s]
overall:  54%|█████▍    | 2175/4000 [11:09<07:53,  3.86it/s]

train:  62%|██████▏   | 2175/3515 [11:07<05:47,  3.86it/s]
overall:  54%|█████▍    | 2176/4000 [11:10<07:53,  3.85it/s]

train:  62%|██████▏   | 2176/3515 [11:07<05:47,  3.85it/s]
overall:  54%|█████▍    | 2177/4000 [11:10<07:58,  3.81it/s]

train:  62%|██████▏   | 2177/3515 [11:07<05:51,  3.81it/s]
overall:  54%|█████▍    | 2178/4000 [11:10<07:48,  3.89it/s]

train:  62%|██████▏   | 2178/3515 [11:07<05:43,  3.89it/s]
overall:  54%|█████▍    | 2179/4000 [11:10<07:51,  3.86it/s]

train:  62%|██████▏   | 2179/3515 [11:08<05:46,  3.86it/s]train loss: 0.6004005074501038
train per: 0.17374816397382828

overall:  55%|█████▍    | 2180/4000 [11:11<07:42,  3.93it/s]

train:  62%|██████▏   | 2180/3515 [11:08<05:39,  3.93it/s]
overall:  55%|█████▍    | 2181/4000 [11:11<07:55,  3.82it/s]

train:  62%|██████▏   | 2181/3515 [11:08<05:48,  3.82it/s]
overall:  55%|█████▍    | 2182/4000 [11:11<07:56,  3.82it/s]

train:  62%|██████▏   | 2182/3515 [11:08<05:48,  3.82it/s]
overall:  55%|█████▍    | 2183/4000 [11:11<07:51,  3.86it/s]

train:  62%|██████▏   | 2183/3515 [11:09<05:45,  3.86it/s]
overall:  55%|█████▍    | 2184/4000 [11:12<07:58,  3.79it/s]

train:  62%|██████▏   | 2184/3515 [11:09<05:50,  3.79it/s]
overall:  55%|█████▍    | 2185/4000 [11:12<07:28,  4.05it/s]

train:  62%|██████▏   | 2185/3515 [11:09<05:28,  4.05it/s]
overall:  55%|█████▍    | 2186/4000 [11:12<06:37,  4.56it/s]

train:  62%|██████▏   | 2186/3515 [11:09<04:51,  4.56it/s]
overall:  55%|█████▍    | 2187/4000 [11:12<07:00,  4.31it/s]

train:  62%|██████▏   | 2187/3515 [11:10<05:08,  4.31it/s]
overall:  55%|█████▍    | 2188/4000 [11:13<07:14,  4.17it/s]

train:  62%|██████▏   | 2188/3515 [11:10<05:18,  4.17it/s]
overall:  55%|█████▍    | 2189/4000 [11:13<06:55,  4.36it/s]

train:  62%|██████▏   | 2189/3515 [11:10<05:03,  4.36it/s]
overall:  55%|█████▍    | 2190/4000 [11:13<06:02,  4.99it/s]

train:  62%|██████▏   | 2190/3515 [11:10<04:25,  4.99it/s]
overall:  55%|█████▍    | 2191/4000 [11:13<06:33,  4.59it/s]

train:  62%|██████▏   | 2191/3515 [11:11<04:48,  4.59it/s]
overall:  55%|█████▍    | 2192/4000 [11:13<07:07,  4.23it/s]

train:  62%|██████▏   | 2192/3515 [11:11<05:13,  4.23it/s]
overall:  55%|█████▍    | 2193/4000 [11:14<09:21,  3.22it/s]

train:  62%|██████▏   | 2193/3515 [11:11<06:50,  3.22it/s]
overall:  55%|█████▍    | 2194/4000 [11:14<08:55,  3.37it/s]

train:  62%|██████▏   | 2194/3515 [11:12<06:31,  3.37it/s]
overall:  55%|█████▍    | 2195/4000 [11:14<08:31,  3.53it/s]

train:  62%|██████▏   | 2195/3515 [11:12<06:14,  3.53it/s]
overall:  55%|█████▍    | 2196/4000 [11:15<08:31,  3.53it/s]

train:  62%|██████▏   | 2196/3515 [11:12<06:13,  3.53it/s]
overall:  55%|█████▍    | 2197/4000 [11:15<08:00,  3.75it/s]

train:  63%|██████▎   | 2197/3515 [11:12<05:51,  3.75it/s]
overall:  55%|█████▍    | 2198/4000 [11:15<08:03,  3.73it/s]

train:  63%|██████▎   | 2198/3515 [11:13<05:53,  3.73it/s]
overall:  55%|█████▍    | 2199/4000 [11:15<08:04,  3.72it/s]

train:  63%|██████▎   | 2199/3515 [11:13<05:53,  3.72it/s]train loss: 0.5732216835021973
train per: 0.16647229597280463

overall:  55%|█████▌    | 2200/4000 [11:16<08:02,  3.73it/s]

train:  63%|██████▎   | 2200/3515 [11:13<05:52,  3.73it/s]
overall:  55%|█████▌    | 2201/4000 [11:16<08:08,  3.68it/s]

train:  63%|██████▎   | 2201/3515 [11:13<05:56,  3.68it/s]
overall:  55%|█████▌    | 2203/4000 [11:16<06:49,  4.38it/s]

train:  63%|██████▎   | 2203/3515 [11:14<04:59,  4.38it/s]
overall:  55%|█████▌    | 2204/4000 [11:17<08:05,  3.70it/s]

train:  63%|██████▎   | 2204/3515 [11:14<05:54,  3.70it/s]
overall:  55%|█████▌    | 2205/4000 [11:17<07:45,  3.85it/s]

train:  63%|██████▎   | 2205/3515 [11:14<05:39,  3.85it/s]
overall:  55%|█████▌    | 2206/4000 [11:17<07:46,  3.84it/s]

train:  63%|██████▎   | 2206/3515 [11:15<05:40,  3.84it/s]
overall:  55%|█████▌    | 2207/4000 [11:18<07:46,  3.85it/s]

train:  63%|██████▎   | 2207/3515 [11:15<05:40,  3.85it/s]
overall:  55%|█████▌    | 2208/4000 [11:18<07:54,  3.77it/s]

train:  63%|██████▎   | 2208/3515 [11:15<05:46,  3.77it/s]
overall:  55%|█████▌    | 2209/4000 [11:18<07:48,  3.82it/s]

train:  63%|██████▎   | 2209/3515 [11:15<05:41,  3.82it/s]
overall:  55%|█████▌    | 2210/4000 [11:18<07:47,  3.83it/s]

train:  63%|██████▎   | 2210/3515 [11:16<05:40,  3.83it/s]
overall:  55%|█████▌    | 2211/4000 [11:19<08:02,  3.71it/s]

train:  63%|██████▎   | 2211/3515 [11:16<05:51,  3.71it/s]
overall:  55%|█████▌    | 2212/4000 [11:19<07:59,  3.73it/s]

train:  63%|██████▎   | 2212/3515 [11:16<05:49,  3.73it/s]
overall:  55%|█████▌    | 2213/4000 [11:19<08:02,  3.70it/s]

train:  63%|██████▎   | 2213/3515 [11:17<05:51,  3.70it/s]
overall:  55%|█████▌    | 2214/4000 [11:19<07:56,  3.75it/s]

train:  63%|██████▎   | 2214/3515 [11:17<05:46,  3.75it/s]
overall:  55%|█████▌    | 2215/4000 [11:20<07:56,  3.75it/s]

train:  63%|██████▎   | 2215/3515 [11:17<05:46,  3.75it/s]
overall:  55%|█████▌    | 2216/4000 [11:20<08:01,  3.70it/s]

train:  63%|██████▎   | 2216/3515 [11:17<05:50,  3.70it/s]
overall:  55%|█████▌    | 2217/4000 [11:20<10:01,  2.97it/s]

train:  63%|██████▎   | 2217/3515 [11:18<07:17,  2.97it/s]
overall:  55%|█████▌    | 2218/4000 [11:21<09:24,  3.16it/s]

train:  63%|██████▎   | 2218/3515 [11:18<06:50,  3.16it/s]
overall:  55%|█████▌    | 2219/4000 [11:21<08:43,  3.40it/s]

train:  63%|██████▎   | 2219/3515 [11:18<06:21,  3.40it/s]train loss: 0.5871105194091797
train per: 0.1689644571617755

overall:  56%|█████▌    | 2220/4000 [11:21<08:26,  3.51it/s]

train:  63%|██████▎   | 2220/3515 [11:19<06:10,  3.50it/s]
overall:  56%|█████▌    | 2221/4000 [11:21<08:19,  3.56it/s]

train:  63%|██████▎   | 2221/3515 [11:19<06:02,  3.57it/s]
overall:  56%|█████▌    | 2222/4000 [11:22<08:00,  3.70it/s]

train:  63%|██████▎   | 2222/3515 [11:19<05:48,  3.71it/s]
overall:  56%|█████▌    | 2223/4000 [11:22<08:01,  3.69it/s]

train:  63%|██████▎   | 2223/3515 [11:19<05:49,  3.70it/s]
overall:  56%|█████▌    | 2224/4000 [11:22<08:00,  3.70it/s]

train:  63%|██████▎   | 2224/3515 [11:20<05:48,  3.70it/s]
overall:  56%|█████▌    | 2225/4000 [11:23<08:00,  3.69it/s]

train:  63%|██████▎   | 2225/3515 [11:20<05:49,  3.69it/s]
overall:  56%|█████▌    | 2226/4000 [11:23<08:01,  3.68it/s]

train:  63%|██████▎   | 2226/3515 [11:20<05:49,  3.69it/s]
overall:  56%|█████▌    | 2227/4000 [11:23<08:07,  3.63it/s]

train:  63%|██████▎   | 2227/3515 [11:20<05:54,  3.63it/s]
overall:  56%|█████▌    | 2228/4000 [11:23<08:40,  3.40it/s]

train:  63%|██████▎   | 2228/3515 [11:21<06:18,  3.40it/s]
overall:  56%|█████▌    | 2229/4000 [11:24<08:30,  3.47it/s]

train:  63%|██████▎   | 2229/3515 [11:21<06:10,  3.47it/s]
overall:  56%|█████▌    | 2230/4000 [11:24<08:11,  3.60it/s]

train:  63%|██████▎   | 2230/3515 [11:21<05:56,  3.60it/s]
overall:  56%|█████▌    | 2231/4000 [11:24<08:08,  3.62it/s]

train:  63%|██████▎   | 2231/3515 [11:22<05:54,  3.62it/s]
overall:  56%|█████▌    | 2232/4000 [11:25<08:05,  3.64it/s]

train:  63%|██████▎   | 2232/3515 [11:22<05:52,  3.64it/s]
overall:  56%|█████▌    | 2233/4000 [11:25<07:57,  3.70it/s]

train:  64%|██████▎   | 2233/3515 [11:22<05:46,  3.70it/s]
overall:  56%|█████▌    | 2234/4000 [11:25<07:45,  3.79it/s]

train:  64%|██████▎   | 2234/3515 [11:22<05:37,  3.79it/s]
overall:  56%|█████▌    | 2235/4000 [11:25<07:35,  3.88it/s]

train:  64%|██████▎   | 2235/3515 [11:23<05:30,  3.88it/s]
overall:  56%|█████▌    | 2236/4000 [11:26<07:36,  3.86it/s]

train:  64%|██████▎   | 2236/3515 [11:23<05:31,  3.86it/s]
overall:  56%|█████▌    | 2237/4000 [11:26<07:46,  3.78it/s]

train:  64%|██████▎   | 2237/3515 [11:23<05:38,  3.78it/s]
overall:  56%|█████▌    | 2238/4000 [11:26<07:50,  3.74it/s]

train:  64%|██████▎   | 2238/3515 [11:23<05:41,  3.74it/s]
overall:  56%|█████▌    | 2239/4000 [11:26<07:33,  3.88it/s]

train:  64%|██████▎   | 2239/3515 [11:24<05:28,  3.88it/s]train loss: 0.5581371784210205
train per: 0.16223003515821197

overall:  56%|█████▌    | 2240/4000 [11:27<07:52,  3.73it/s]

train:  64%|██████▎   | 2240/3515 [11:24<05:43,  3.71it/s]
overall:  56%|█████▌    | 2241/4000 [11:27<08:33,  3.42it/s]

train:  64%|██████▍   | 2241/3515 [11:24<06:11,  3.43it/s]
overall:  56%|█████▌    | 2242/4000 [11:27<08:17,  3.54it/s]

train:  64%|██████▍   | 2242/3515 [11:25<05:59,  3.54it/s]
overall:  56%|█████▌    | 2243/4000 [11:27<08:06,  3.61it/s]

train:  64%|██████▍   | 2243/3515 [11:25<05:52,  3.61it/s]
overall:  56%|█████▌    | 2244/4000 [11:28<07:59,  3.66it/s]

train:  64%|██████▍   | 2244/3515 [11:25<05:46,  3.66it/s]
overall:  56%|█████▌    | 2245/4000 [11:28<07:51,  3.72it/s]

train:  64%|██████▍   | 2245/3515 [11:25<05:41,  3.72it/s]
overall:  56%|█████▌    | 2246/4000 [11:28<07:56,  3.68it/s]

train:  64%|██████▍   | 2246/3515 [11:26<05:44,  3.68it/s]
overall:  56%|█████▌    | 2247/4000 [11:29<08:05,  3.61it/s]

train:  64%|██████▍   | 2247/3515 [11:26<05:51,  3.61it/s]
overall:  56%|█████▌    | 2248/4000 [11:29<08:07,  3.59it/s]

train:  64%|██████▍   | 2248/3515 [11:26<05:52,  3.59it/s]
overall:  56%|█████▌    | 2249/4000 [11:29<08:05,  3.61it/s]

train:  64%|██████▍   | 2249/3515 [11:26<05:50,  3.61it/s]
overall:  56%|█████▋    | 2250/4000 [11:29<07:50,  3.72it/s]

train:  64%|██████▍   | 2250/3515 [11:27<05:39,  3.72it/s]
overall:  56%|█████▋    | 2251/4000 [11:30<07:57,  3.66it/s]

train:  64%|██████▍   | 2251/3515 [11:27<05:44,  3.66it/s]
overall:  56%|█████▋    | 2252/4000 [11:30<09:00,  3.23it/s]

train:  64%|██████▍   | 2252/3515 [11:27<06:30,  3.23it/s]
overall:  56%|█████▋    | 2253/4000 [11:30<07:49,  3.72it/s]

train:  64%|██████▍   | 2253/3515 [11:28<05:38,  3.73it/s]
overall:  56%|█████▋    | 2254/4000 [11:30<07:37,  3.81it/s]

train:  64%|██████▍   | 2254/3515 [11:28<05:30,  3.81it/s]
overall:  56%|█████▋    | 2255/4000 [11:31<07:43,  3.77it/s]

train:  64%|██████▍   | 2255/3515 [11:28<05:34,  3.77it/s]
overall:  56%|█████▋    | 2256/4000 [11:31<07:42,  3.77it/s]

train:  64%|██████▍   | 2256/3515 [11:28<05:34,  3.77it/s]
overall:  56%|█████▋    | 2257/4000 [11:31<07:13,  4.02it/s]

train:  64%|██████▍   | 2257/3515 [11:29<05:12,  4.02it/s]
overall:  56%|█████▋    | 2258/4000 [11:31<07:30,  3.87it/s]

train:  64%|██████▍   | 2258/3515 [11:29<05:24,  3.87it/s]
overall:  56%|█████▋    | 2259/4000 [11:32<06:50,  4.24it/s]

train:  64%|██████▍   | 2259/3515 [11:29<04:56,  4.24it/s]train loss: 0.5588110089302063
train per: 0.1633720930232558

overall:  56%|█████▋    | 2260/4000 [11:32<07:05,  4.09it/s]

train:  64%|██████▍   | 2260/3515 [11:29<05:07,  4.09it/s]
overall:  57%|█████▋    | 2261/4000 [11:32<07:26,  3.90it/s]

train:  64%|██████▍   | 2261/3515 [11:30<05:21,  3.90it/s]
overall:  57%|█████▋    | 2262/4000 [11:32<07:31,  3.85it/s]

train:  64%|██████▍   | 2262/3515 [11:30<05:25,  3.85it/s]
overall:  57%|█████▋    | 2263/4000 [11:33<07:32,  3.84it/s]

train:  64%|██████▍   | 2263/3515 [11:30<05:26,  3.84it/s]
overall:  57%|█████▋    | 2264/4000 [11:33<07:34,  3.82it/s]

train:  64%|██████▍   | 2264/3515 [11:30<05:27,  3.82it/s]
overall:  57%|█████▋    | 2265/4000 [11:33<09:23,  3.08it/s]

train:  64%|██████▍   | 2265/3515 [11:31<06:46,  3.08it/s]
overall:  57%|█████▋    | 2266/4000 [11:34<09:05,  3.18it/s]

train:  64%|██████▍   | 2266/3515 [11:31<06:32,  3.18it/s]
overall:  57%|█████▋    | 2267/4000 [11:34<08:24,  3.44it/s]

train:  64%|██████▍   | 2267/3515 [11:31<06:03,  3.44it/s]
overall:  57%|█████▋    | 2268/4000 [11:34<08:01,  3.59it/s]

train:  65%|██████▍   | 2268/3515 [11:32<05:47,  3.59it/s]
overall:  57%|█████▋    | 2269/4000 [11:35<08:20,  3.46it/s]

train:  65%|██████▍   | 2269/3515 [11:32<06:00,  3.46it/s]
overall:  57%|█████▋    | 2270/4000 [11:35<08:09,  3.54it/s]

train:  65%|██████▍   | 2270/3515 [11:32<05:51,  3.54it/s]
overall:  57%|█████▋    | 2271/4000 [11:35<06:38,  4.34it/s]

train:  65%|██████▍   | 2271/3515 [11:32<04:46,  4.34it/s]
overall:  57%|█████▋    | 2272/4000 [11:35<06:47,  4.24it/s]

train:  65%|██████▍   | 2272/3515 [11:33<04:53,  4.24it/s]
overall:  57%|█████▋    | 2273/4000 [11:35<07:10,  4.01it/s]

train:  65%|██████▍   | 2273/3515 [11:33<05:09,  4.01it/s]
overall:  57%|█████▋    | 2274/4000 [11:36<07:25,  3.87it/s]

train:  65%|██████▍   | 2274/3515 [11:33<05:20,  3.87it/s]
overall:  57%|█████▋    | 2275/4000 [11:36<07:26,  3.87it/s]

train:  65%|██████▍   | 2275/3515 [11:33<05:20,  3.87it/s]
overall:  57%|█████▋    | 2276/4000 [11:36<07:53,  3.64it/s]

train:  65%|██████▍   | 2276/3515 [11:34<05:40,  3.64it/s]
overall:  57%|█████▋    | 2277/4000 [11:37<07:59,  3.59it/s]

train:  65%|██████▍   | 2277/3515 [11:34<05:44,  3.59it/s]
overall:  57%|█████▋    | 2278/4000 [11:37<07:54,  3.63it/s]

train:  65%|██████▍   | 2278/3515 [11:34<05:40,  3.63it/s]
overall:  57%|█████▋    | 2279/4000 [11:37<07:47,  3.68it/s]

train:  65%|██████▍   | 2279/3515 [11:35<05:35,  3.68it/s]train loss: 0.5997000932693481
train per: 0.16764428995582592

overall:  57%|█████▋    | 2280/4000 [11:37<06:55,  4.14it/s]

train:  65%|██████▍   | 2280/3515 [11:35<04:58,  4.14it/s]
overall:  57%|█████▋    | 2281/4000 [11:38<06:56,  4.13it/s]

train:  65%|██████▍   | 2281/3515 [11:35<04:58,  4.13it/s]
overall:  57%|█████▋    | 2282/4000 [11:38<07:05,  4.04it/s]

train:  65%|██████▍   | 2282/3515 [11:35<05:05,  4.04it/s]
overall:  57%|█████▋    | 2283/4000 [11:38<07:15,  3.94it/s]

train:  65%|██████▍   | 2283/3515 [11:35<05:12,  3.94it/s]
overall:  57%|█████▋    | 2284/4000 [11:38<06:12,  4.61it/s]

train:  65%|██████▍   | 2284/3515 [11:36<04:27,  4.61it/s]
overall:  57%|█████▋    | 2285/4000 [11:38<06:39,  4.29it/s]

train:  65%|██████▌   | 2285/3515 [11:36<04:46,  4.29it/s]
overall:  57%|█████▋    | 2286/4000 [11:39<07:00,  4.07it/s]

train:  65%|██████▌   | 2286/3515 [11:36<05:01,  4.07it/s]
overall:  57%|█████▋    | 2287/4000 [11:39<07:01,  4.07it/s]

train:  65%|██████▌   | 2287/3515 [11:36<05:01,  4.07it/s]
overall:  57%|█████▋    | 2288/4000 [11:39<07:09,  3.98it/s]

train:  65%|██████▌   | 2288/3515 [11:37<05:07,  3.98it/s]
overall:  57%|█████▋    | 2289/4000 [11:40<09:06,  3.13it/s]

train:  65%|██████▌   | 2289/3515 [11:37<06:31,  3.13it/s]
overall:  57%|█████▋    | 2290/4000 [11:40<08:40,  3.29it/s]

train:  65%|██████▌   | 2290/3515 [11:37<06:12,  3.29it/s]
overall:  57%|█████▋    | 2291/4000 [11:40<08:27,  3.37it/s]

train:  65%|██████▌   | 2291/3515 [11:38<06:03,  3.37it/s]
overall:  57%|█████▋    | 2292/4000 [11:41<08:18,  3.42it/s]

train:  65%|██████▌   | 2292/3515 [11:38<05:57,  3.42it/s]
overall:  57%|█████▋    | 2293/4000 [11:41<08:26,  3.37it/s]

train:  65%|██████▌   | 2293/3515 [11:38<06:02,  3.37it/s]
overall:  57%|█████▋    | 2294/4000 [11:41<08:15,  3.44it/s]

train:  65%|██████▌   | 2294/3515 [11:39<05:54,  3.44it/s]
overall:  57%|█████▋    | 2295/4000 [11:41<08:06,  3.51it/s]

train:  65%|██████▌   | 2295/3515 [11:39<05:47,  3.51it/s]
overall:  57%|█████▋    | 2296/4000 [11:42<07:37,  3.73it/s]

train:  65%|██████▌   | 2296/3515 [11:39<05:27,  3.73it/s]
overall:  57%|█████▋    | 2297/4000 [11:42<07:37,  3.72it/s]

train:  65%|██████▌   | 2297/3515 [11:39<05:27,  3.72it/s]
overall:  57%|█████▋    | 2298/4000 [11:42<07:09,  3.97it/s]

train:  65%|██████▌   | 2298/3515 [11:40<05:06,  3.97it/s]
overall:  57%|█████▋    | 2299/4000 [11:42<07:19,  3.87it/s]

train:  65%|██████▌   | 2299/3515 [11:40<05:14,  3.87it/s]train loss: 0.5561060905456543
train per: 0.16278905477323036

overall:  57%|█████▊    | 2300/4000 [11:43<08:11,  3.46it/s]

train:  65%|██████▌   | 2300/3515 [11:40<05:51,  3.46it/s]
overall:  58%|█████▊    | 2301/4000 [11:43<08:01,  3.53it/s]

train:  65%|██████▌   | 2301/3515 [11:40<05:43,  3.53it/s]
overall:  58%|█████▊    | 2302/4000 [11:43<06:56,  4.07it/s]

train:  65%|██████▌   | 2302/3515 [11:41<04:57,  4.08it/s]
overall:  58%|█████▊    | 2303/4000 [11:43<07:11,  3.93it/s]

train:  66%|██████▌   | 2303/3515 [11:41<05:08,  3.93it/s]
overall:  58%|█████▊    | 2304/4000 [11:44<07:15,  3.90it/s]

train:  66%|██████▌   | 2304/3515 [11:41<05:10,  3.90it/s]
overall:  58%|█████▊    | 2305/4000 [11:44<06:58,  4.05it/s]

train:  66%|██████▌   | 2305/3515 [11:41<04:58,  4.05it/s]
overall:  58%|█████▊    | 2306/4000 [11:44<06:56,  4.07it/s]

train:  66%|██████▌   | 2306/3515 [11:42<04:57,  4.07it/s]
overall:  58%|█████▊    | 2307/4000 [11:44<07:08,  3.95it/s]

train:  66%|██████▌   | 2307/3515 [11:42<05:06,  3.95it/s]
overall:  58%|█████▊    | 2308/4000 [11:45<07:15,  3.88it/s]

train:  66%|██████▌   | 2308/3515 [11:42<05:10,  3.88it/s]
overall:  58%|█████▊    | 2309/4000 [11:45<07:02,  4.00it/s]

train:  66%|██████▌   | 2309/3515 [11:42<05:01,  4.00it/s]
overall:  58%|█████▊    | 2310/4000 [11:45<07:12,  3.91it/s]

train:  66%|██████▌   | 2310/3515 [11:43<05:08,  3.91it/s]
overall:  58%|█████▊    | 2311/4000 [11:46<07:15,  3.87it/s]

train:  66%|██████▌   | 2311/3515 [11:43<05:10,  3.87it/s]
overall:  58%|█████▊    | 2312/4000 [11:46<07:23,  3.81it/s]

train:  66%|██████▌   | 2312/3515 [11:43<05:16,  3.81it/s]
overall:  58%|█████▊    | 2313/4000 [11:46<09:05,  3.09it/s]

train:  66%|██████▌   | 2313/3515 [11:44<06:28,  3.09it/s]
overall:  58%|█████▊    | 2314/4000 [11:47<08:34,  3.28it/s]

train:  66%|██████▌   | 2314/3515 [11:44<06:06,  3.28it/s]
overall:  58%|█████▊    | 2315/4000 [11:47<08:10,  3.44it/s]

train:  66%|██████▌   | 2315/3515 [11:44<05:49,  3.44it/s]
overall:  58%|█████▊    | 2316/4000 [11:47<07:57,  3.52it/s]

train:  66%|██████▌   | 2316/3515 [11:44<05:40,  3.52it/s]
overall:  58%|█████▊    | 2317/4000 [11:47<08:05,  3.46it/s]

train:  66%|██████▌   | 2317/3515 [11:45<05:45,  3.46it/s]
overall:  58%|█████▊    | 2318/4000 [11:48<07:54,  3.54it/s]

train:  66%|██████▌   | 2318/3515 [11:45<05:37,  3.54it/s]
overall:  58%|█████▊    | 2319/4000 [11:48<07:50,  3.57it/s]

train:  66%|██████▌   | 2319/3515 [11:45<05:35,  3.57it/s]train loss: 0.565602719783783
train per: 0.16539251755936737

overall:  58%|█████▊    | 2320/4000 [11:48<07:33,  3.70it/s]

train:  66%|██████▌   | 2320/3515 [11:46<05:22,  3.70it/s]
overall:  58%|█████▊    | 2321/4000 [11:48<07:24,  3.78it/s]

train:  66%|██████▌   | 2321/3515 [11:46<05:15,  3.78it/s]
overall:  58%|█████▊    | 2322/4000 [11:49<07:28,  3.74it/s]

train:  66%|██████▌   | 2322/3515 [11:46<05:18,  3.74it/s]
overall:  58%|█████▊    | 2323/4000 [11:49<07:30,  3.73it/s]

train:  66%|██████▌   | 2323/3515 [11:46<05:19,  3.73it/s]
overall:  58%|█████▊    | 2324/4000 [11:49<06:42,  4.16it/s]

train:  66%|██████▌   | 2324/3515 [11:46<04:46,  4.16it/s]
overall:  58%|█████▊    | 2325/4000 [11:49<07:10,  3.89it/s]

train:  66%|██████▌   | 2325/3515 [11:47<05:06,  3.89it/s]
overall:  58%|█████▊    | 2326/4000 [11:50<07:11,  3.88it/s]

train:  66%|██████▌   | 2326/3515 [11:47<05:06,  3.88it/s]
overall:  58%|█████▊    | 2327/4000 [11:50<07:19,  3.81it/s]

train:  66%|██████▌   | 2327/3515 [11:47<05:11,  3.81it/s]
overall:  58%|█████▊    | 2328/4000 [11:50<07:44,  3.60it/s]

train:  66%|██████▌   | 2328/3515 [11:48<05:29,  3.60it/s]
overall:  58%|█████▊    | 2329/4000 [11:51<07:36,  3.66it/s]

train:  66%|██████▋   | 2329/3515 [11:48<05:24,  3.66it/s]
overall:  58%|█████▊    | 2330/4000 [11:51<07:05,  3.93it/s]

train:  66%|██████▋   | 2330/3515 [11:48<05:01,  3.93it/s]
overall:  58%|█████▊    | 2332/4000 [11:51<05:56,  4.67it/s]

train:  66%|██████▋   | 2332/3515 [11:48<04:13,  4.67it/s]
overall:  58%|█████▊    | 2333/4000 [11:51<06:20,  4.38it/s]

train:  66%|██████▋   | 2333/3515 [11:49<04:30,  4.38it/s]
overall:  58%|█████▊    | 2334/4000 [11:51<05:29,  5.05it/s]

train:  66%|██████▋   | 2334/3515 [11:49<03:53,  5.05it/s]
overall:  58%|█████▊    | 2335/4000 [11:52<05:50,  4.75it/s]

train:  66%|██████▋   | 2335/3515 [11:49<04:08,  4.75it/s]
overall:  58%|█████▊    | 2336/4000 [11:52<06:19,  4.39it/s]

train:  66%|██████▋   | 2336/3515 [11:49<04:28,  4.39it/s]
overall:  58%|█████▊    | 2337/4000 [11:52<07:48,  3.55it/s]

train:  66%|██████▋   | 2337/3515 [11:50<05:31,  3.55it/s]
overall:  58%|█████▊    | 2338/4000 [11:53<07:26,  3.72it/s]

train:  67%|██████▋   | 2338/3515 [11:50<05:16,  3.72it/s]
overall:  58%|█████▊    | 2339/4000 [11:53<07:18,  3.79it/s]

train:  67%|██████▋   | 2339/3515 [11:50<05:10,  3.79it/s]train loss: 0.5865007638931274
train per: 0.16369386113345724

overall:  58%|█████▊    | 2340/4000 [11:53<06:43,  4.11it/s]

train:  67%|██████▋   | 2340/3515 [11:50<04:46,  4.11it/s]
overall:  59%|█████▊    | 2342/4000 [11:53<05:55,  4.66it/s]

train:  67%|██████▋   | 2342/3515 [11:51<04:11,  4.66it/s]
overall:  59%|█████▊    | 2343/4000 [11:54<06:14,  4.43it/s]

train:  67%|██████▋   | 2343/3515 [11:51<04:24,  4.43it/s]
overall:  59%|█████▊    | 2344/4000 [11:54<06:27,  4.28it/s]

train:  67%|██████▋   | 2344/3515 [11:51<04:33,  4.28it/s]
overall:  59%|█████▊    | 2345/4000 [11:54<06:46,  4.07it/s]

train:  67%|██████▋   | 2345/3515 [11:52<04:47,  4.07it/s]
overall:  59%|█████▊    | 2346/4000 [11:54<06:57,  3.96it/s]

train:  67%|██████▋   | 2346/3515 [11:52<04:55,  3.96it/s]
overall:  59%|█████▊    | 2347/4000 [11:55<06:46,  4.07it/s]

train:  67%|██████▋   | 2347/3515 [11:52<04:47,  4.07it/s]
overall:  59%|█████▊    | 2348/4000 [11:55<07:16,  3.79it/s]

train:  67%|██████▋   | 2348/3515 [11:52<05:08,  3.79it/s]
overall:  59%|█████▊    | 2349/4000 [11:55<07:22,  3.73it/s]

train:  67%|██████▋   | 2349/3515 [11:53<05:12,  3.73it/s]
overall:  59%|█████▉    | 2350/4000 [11:56<07:16,  3.78it/s]

train:  67%|██████▋   | 2350/3515 [11:53<05:08,  3.78it/s]
overall:  59%|█████▉    | 2351/4000 [11:56<06:47,  4.05it/s]

train:  67%|██████▋   | 2351/3515 [11:53<04:47,  4.05it/s]
overall:  59%|█████▉    | 2352/4000 [11:56<06:54,  3.98it/s]

train:  67%|██████▋   | 2352/3515 [11:53<04:52,  3.98it/s]
overall:  59%|█████▉    | 2353/4000 [11:56<06:57,  3.95it/s]

train:  67%|██████▋   | 2353/3515 [11:54<04:54,  3.95it/s]
overall:  59%|█████▉    | 2354/4000 [11:57<06:52,  3.99it/s]

train:  67%|██████▋   | 2354/3515 [11:54<04:50,  3.99it/s]
overall:  59%|█████▉    | 2355/4000 [11:57<06:56,  3.95it/s]

train:  67%|██████▋   | 2355/3515 [11:54<04:53,  3.95it/s]
overall:  59%|█████▉    | 2356/4000 [11:57<07:06,  3.86it/s]

train:  67%|██████▋   | 2356/3515 [11:54<05:00,  3.86it/s]
overall:  59%|█████▉    | 2357/4000 [11:57<06:59,  3.92it/s]

train:  67%|██████▋   | 2357/3515 [11:55<04:55,  3.92it/s]
overall:  59%|█████▉    | 2358/4000 [11:58<07:07,  3.84it/s]

train:  67%|██████▋   | 2358/3515 [11:55<05:01,  3.84it/s]
overall:  59%|█████▉    | 2359/4000 [11:58<07:11,  3.81it/s]

train:  67%|██████▋   | 2359/3515 [11:55<05:03,  3.81it/s]train loss: 0.573129415512085
train per: 0.16865286260489598

overall:  59%|█████▉    | 2360/4000 [11:58<07:24,  3.69it/s]

train:  67%|██████▋   | 2360/3515 [11:56<05:13,  3.69it/s]
overall:  59%|█████▉    | 2361/4000 [11:59<08:29,  3.22it/s]

train:  67%|██████▋   | 2361/3515 [11:56<05:58,  3.22it/s]
overall:  59%|█████▉    | 2362/4000 [11:59<08:01,  3.40it/s]

train:  67%|██████▋   | 2362/3515 [11:56<05:39,  3.40it/s]
overall:  59%|█████▉    | 2363/4000 [11:59<07:43,  3.53it/s]

train:  67%|██████▋   | 2363/3515 [11:56<05:26,  3.53it/s]
overall:  59%|█████▉    | 2364/4000 [11:59<06:15,  4.35it/s]

train:  67%|██████▋   | 2364/3515 [11:57<04:24,  4.35it/s]
overall:  59%|█████▉    | 2365/4000 [11:59<06:33,  4.15it/s]

train:  67%|██████▋   | 2365/3515 [11:57<04:37,  4.15it/s]
overall:  59%|█████▉    | 2366/4000 [12:00<05:47,  4.71it/s]

train:  67%|██████▋   | 2366/3515 [11:57<04:04,  4.71it/s]
overall:  59%|█████▉    | 2367/4000 [12:00<05:44,  4.74it/s]

train:  67%|██████▋   | 2367/3515 [11:57<04:02,  4.74it/s]
overall:  59%|█████▉    | 2368/4000 [12:00<06:09,  4.42it/s]

train:  67%|██████▋   | 2368/3515 [11:57<04:19,  4.42it/s]
overall:  59%|█████▉    | 2369/4000 [12:00<06:24,  4.24it/s]

train:  67%|██████▋   | 2369/3515 [11:58<04:30,  4.24it/s]
overall:  59%|█████▉    | 2370/4000 [12:01<06:43,  4.04it/s]

train:  67%|██████▋   | 2370/3515 [11:58<04:43,  4.04it/s]
overall:  59%|█████▉    | 2371/4000 [12:01<06:46,  4.01it/s]

train:  67%|██████▋   | 2371/3515 [11:58<04:45,  4.01it/s]
overall:  59%|█████▉    | 2372/4000 [12:01<07:14,  3.75it/s]

train:  67%|██████▋   | 2372/3515 [11:59<05:04,  3.75it/s]
overall:  59%|█████▉    | 2373/4000 [12:01<07:08,  3.80it/s]

train:  68%|██████▊   | 2373/3515 [11:59<05:00,  3.80it/s]
overall:  59%|█████▉    | 2374/4000 [12:02<06:57,  3.90it/s]

train:  68%|██████▊   | 2374/3515 [11:59<04:52,  3.90it/s]
overall:  59%|█████▉    | 2375/4000 [12:02<07:03,  3.84it/s]

train:  68%|██████▊   | 2375/3515 [11:59<04:57,  3.84it/s]
overall:  59%|█████▉    | 2376/4000 [12:02<05:58,  4.53it/s]

train:  68%|██████▊   | 2376/3515 [11:59<04:11,  4.53it/s]
overall:  59%|█████▉    | 2377/4000 [12:02<06:07,  4.42it/s]

train:  68%|██████▊   | 2377/3515 [12:00<04:17,  4.42it/s]
overall:  59%|█████▉    | 2378/4000 [12:03<06:21,  4.25it/s]

train:  68%|██████▊   | 2378/3515 [12:00<04:27,  4.25it/s]
overall:  59%|█████▉    | 2379/4000 [12:03<06:14,  4.33it/s]

train:  68%|██████▊   | 2379/3515 [12:00<04:22,  4.33it/s]train loss: 0.563374400138855
train per: 0.16319384650708915

overall:  60%|█████▉    | 2380/4000 [12:03<06:39,  4.05it/s]

train:  68%|██████▊   | 2380/3515 [12:00<04:40,  4.05it/s]
overall:  60%|█████▉    | 2381/4000 [12:03<05:54,  4.57it/s]

train:  68%|██████▊   | 2381/3515 [12:01<04:07,  4.57it/s]
overall:  60%|█████▉    | 2382/4000 [12:03<06:19,  4.26it/s]

train:  68%|██████▊   | 2382/3515 [12:01<04:25,  4.26it/s]
overall:  60%|█████▉    | 2383/4000 [12:04<06:37,  4.07it/s]

train:  68%|██████▊   | 2383/3515 [12:01<04:37,  4.07it/s]
overall:  60%|█████▉    | 2384/4000 [12:04<06:54,  3.90it/s]

train:  68%|██████▊   | 2384/3515 [12:01<04:50,  3.90it/s]
overall:  60%|█████▉    | 2385/4000 [12:04<08:02,  3.35it/s]

train:  68%|██████▊   | 2385/3515 [12:02<05:37,  3.35it/s]
overall:  60%|█████▉    | 2386/4000 [12:05<07:41,  3.50it/s]

train:  68%|██████▊   | 2386/3515 [12:02<05:22,  3.50it/s]
overall:  60%|█████▉    | 2387/4000 [12:05<07:06,  3.79it/s]

train:  68%|██████▊   | 2387/3515 [12:02<04:57,  3.79it/s]
overall:  60%|█████▉    | 2388/4000 [12:05<07:15,  3.70it/s]

train:  68%|██████▊   | 2388/3515 [12:03<05:04,  3.70it/s]
overall:  60%|█████▉    | 2389/4000 [12:05<07:28,  3.59it/s]

train:  68%|██████▊   | 2389/3515 [12:03<05:13,  3.59it/s]
overall:  60%|█████▉    | 2390/4000 [12:06<07:26,  3.61it/s]

train:  68%|██████▊   | 2390/3515 [12:03<05:11,  3.61it/s]
overall:  60%|█████▉    | 2391/4000 [12:06<07:17,  3.68it/s]

train:  68%|██████▊   | 2391/3515 [12:03<05:05,  3.68it/s]
overall:  60%|█████▉    | 2392/4000 [12:06<07:11,  3.72it/s]

train:  68%|██████▊   | 2392/3515 [12:04<05:01,  3.72it/s]
overall:  60%|█████▉    | 2393/4000 [12:07<07:07,  3.76it/s]

train:  68%|██████▊   | 2393/3515 [12:04<04:58,  3.76it/s]
overall:  60%|█████▉    | 2395/4000 [12:07<04:59,  5.36it/s]

train:  68%|██████▊   | 2395/3515 [12:04<03:29,  5.36it/s]
overall:  60%|█████▉    | 2396/4000 [12:07<05:55,  4.51it/s]

train:  68%|██████▊   | 2396/3515 [12:04<04:08,  4.51it/s]
overall:  60%|█████▉    | 2397/4000 [12:07<06:17,  4.25it/s]

train:  68%|██████▊   | 2397/3515 [12:05<04:23,  4.25it/s]
overall:  60%|█████▉    | 2398/4000 [12:08<06:30,  4.10it/s]

train:  68%|██████▊   | 2398/3515 [12:05<04:32,  4.10it/s]
overall:  60%|█████▉    | 2399/4000 [12:08<06:31,  4.09it/s]

train:  68%|██████▊   | 2399/3515 [12:05<04:32,  4.09it/s]train loss: 0.5608921647071838
train per: 0.16265107621319772

overall:  60%|██████    | 2400/4000 [12:08<06:21,  4.19it/s]

train:  68%|██████▊   | 2400/3515 [12:05<04:26,  4.19it/s]
overall:  60%|██████    | 2401/4000 [12:08<06:36,  4.03it/s]

train:  68%|██████▊   | 2401/3515 [12:06<04:36,  4.03it/s]
overall:  60%|██████    | 2402/4000 [12:09<06:30,  4.09it/s]

train:  68%|██████▊   | 2402/3515 [12:06<04:31,  4.09it/s]
overall:  60%|██████    | 2403/4000 [12:09<06:45,  3.94it/s]

train:  68%|██████▊   | 2403/3515 [12:06<04:42,  3.94it/s]
overall:  60%|██████    | 2404/4000 [12:09<06:52,  3.87it/s]

train:  68%|██████▊   | 2404/3515 [12:06<04:47,  3.87it/s]
overall:  60%|██████    | 2405/4000 [12:09<06:54,  3.85it/s]

train:  68%|██████▊   | 2405/3515 [12:07<04:48,  3.85it/s]
overall:  60%|██████    | 2406/4000 [12:10<06:25,  4.14it/s]

train:  68%|██████▊   | 2406/3515 [12:07<04:27,  4.14it/s]
overall:  60%|██████    | 2407/4000 [12:10<05:54,  4.49it/s]

train:  68%|██████▊   | 2407/3515 [12:07<04:06,  4.49it/s]
overall:  60%|██████    | 2408/4000 [12:10<05:31,  4.80it/s]

train:  69%|██████▊   | 2408/3515 [12:07<03:50,  4.80it/s]
overall:  60%|██████    | 2409/4000 [12:10<07:13,  3.67it/s]

train:  69%|██████▊   | 2409/3515 [12:08<05:01,  3.67it/s]
overall:  60%|██████    | 2410/4000 [12:11<07:11,  3.69it/s]

train:  69%|██████▊   | 2410/3515 [12:08<04:59,  3.69it/s]
overall:  60%|██████    | 2411/4000 [12:11<07:05,  3.73it/s]

train:  69%|██████▊   | 2411/3515 [12:08<04:55,  3.73it/s]
overall:  60%|██████    | 2412/4000 [12:11<07:00,  3.78it/s]

train:  69%|██████▊   | 2412/3515 [12:08<04:52,  3.78it/s]
overall:  60%|██████    | 2413/4000 [12:11<07:07,  3.71it/s]

train:  69%|██████▊   | 2413/3515 [12:09<04:56,  3.71it/s]
overall:  60%|██████    | 2414/4000 [12:12<06:40,  3.96it/s]

train:  69%|██████▊   | 2414/3515 [12:09<04:37,  3.96it/s]
overall:  60%|██████    | 2415/4000 [12:12<06:35,  4.01it/s]

train:  69%|██████▊   | 2415/3515 [12:09<04:34,  4.01it/s]
overall:  60%|██████    | 2416/4000 [12:12<08:11,  3.22it/s]

train:  69%|██████▊   | 2416/3515 [12:10<05:40,  3.22it/s]
overall:  60%|██████    | 2417/4000 [12:13<07:44,  3.41it/s]

train:  69%|██████▉   | 2417/3515 [12:10<05:22,  3.41it/s]
overall:  60%|██████    | 2418/4000 [12:13<07:26,  3.54it/s]

train:  69%|██████▉   | 2418/3515 [12:10<05:09,  3.54it/s]
overall:  60%|██████    | 2419/4000 [12:13<07:26,  3.54it/s]

train:  69%|██████▉   | 2419/3515 [12:10<05:09,  3.54it/s]train loss: 0.5425689816474915
train per: 0.15679902508378185

overall:  60%|██████    | 2420/4000 [12:13<07:21,  3.58it/s]

train:  69%|██████▉   | 2420/3515 [12:11<05:05,  3.58it/s]
overall:  61%|██████    | 2421/4000 [12:14<07:01,  3.75it/s]

train:  69%|██████▉   | 2421/3515 [12:11<04:51,  3.75it/s]
overall:  61%|██████    | 2422/4000 [12:14<07:04,  3.71it/s]

train:  69%|██████▉   | 2422/3515 [12:11<04:54,  3.71it/s]
overall:  61%|██████    | 2423/4000 [12:14<07:03,  3.73it/s]

train:  69%|██████▉   | 2423/3515 [12:12<04:52,  3.73it/s]
overall:  61%|██████    | 2424/4000 [12:14<06:49,  3.85it/s]

train:  69%|██████▉   | 2424/3515 [12:12<04:43,  3.85it/s]
overall:  61%|██████    | 2425/4000 [12:15<06:51,  3.83it/s]

train:  69%|██████▉   | 2425/3515 [12:12<04:44,  3.83it/s]
overall:  61%|██████    | 2426/4000 [12:15<06:27,  4.06it/s]

train:  69%|██████▉   | 2426/3515 [12:12<04:28,  4.06it/s]
overall:  61%|██████    | 2428/4000 [12:15<05:31,  4.75it/s]

train:  69%|██████▉   | 2428/3515 [12:13<03:48,  4.75it/s]
overall:  61%|██████    | 2429/4000 [12:15<05:52,  4.46it/s]

train:  69%|██████▉   | 2429/3515 [12:13<04:03,  4.46it/s]
overall:  61%|██████    | 2430/4000 [12:16<06:05,  4.30it/s]

train:  69%|██████▉   | 2430/3515 [12:13<04:12,  4.30it/s]
overall:  61%|██████    | 2431/4000 [12:16<06:04,  4.30it/s]

train:  69%|██████▉   | 2431/3515 [12:13<04:11,  4.30it/s]
overall:  61%|██████    | 2432/4000 [12:16<06:18,  4.14it/s]

train:  69%|██████▉   | 2432/3515 [12:14<04:21,  4.14it/s]
overall:  61%|██████    | 2433/4000 [12:17<06:55,  3.77it/s]

train:  69%|██████▉   | 2433/3515 [12:14<04:46,  3.77it/s]
overall:  61%|██████    | 2435/4000 [12:17<04:56,  5.29it/s]

train:  69%|██████▉   | 2435/3515 [12:14<03:24,  5.29it/s]
overall:  61%|██████    | 2436/4000 [12:17<04:51,  5.36it/s]

train:  69%|██████▉   | 2436/3515 [12:14<03:21,  5.36it/s]
overall:  61%|██████    | 2437/4000 [12:17<05:36,  4.65it/s]

train:  69%|██████▉   | 2437/3515 [12:15<03:51,  4.65it/s]
overall:  61%|██████    | 2438/4000 [12:17<06:00,  4.33it/s]

train:  69%|██████▉   | 2438/3515 [12:15<04:08,  4.33it/s]
overall:  61%|██████    | 2439/4000 [12:18<06:17,  4.13it/s]

train:  69%|██████▉   | 2439/3515 [12:15<04:20,  4.13it/s]train loss: 0.5729182958602905
train per: 0.1635442762226337

overall:  61%|██████    | 2440/4000 [12:18<07:02,  3.69it/s]

train:  69%|██████▉   | 2440/3515 [12:15<04:51,  3.69it/s]
overall:  61%|██████    | 2441/4000 [12:18<06:57,  3.73it/s]

train:  69%|██████▉   | 2441/3515 [12:16<04:47,  3.74it/s]
overall:  61%|██████    | 2442/4000 [12:19<06:59,  3.72it/s]

train:  69%|██████▉   | 2442/3515 [12:16<04:48,  3.72it/s]
overall:  61%|██████    | 2443/4000 [12:19<06:35,  3.94it/s]

train:  70%|██████▉   | 2443/3515 [12:16<04:32,  3.94it/s]
overall:  61%|██████    | 2444/4000 [12:19<06:33,  3.96it/s]

train:  70%|██████▉   | 2444/3515 [12:16<04:30,  3.96it/s]
overall:  61%|██████    | 2445/4000 [12:19<06:38,  3.90it/s]

train:  70%|██████▉   | 2445/3515 [12:17<04:34,  3.90it/s]
overall:  61%|██████    | 2446/4000 [12:20<06:45,  3.83it/s]

train:  70%|██████▉   | 2446/3515 [12:17<04:38,  3.83it/s]
overall:  61%|██████    | 2447/4000 [12:20<06:44,  3.84it/s]

train:  70%|██████▉   | 2447/3515 [12:17<04:38,  3.84it/s]
overall:  61%|██████    | 2448/4000 [12:20<06:47,  3.81it/s]

train:  70%|██████▉   | 2448/3515 [12:18<04:40,  3.81it/s]
overall:  61%|██████    | 2449/4000 [12:20<06:50,  3.78it/s]

train:  70%|██████▉   | 2449/3515 [12:18<04:41,  3.78it/s]
overall:  61%|██████▏   | 2450/4000 [12:21<06:55,  3.73it/s]

train:  70%|██████▉   | 2450/3515 [12:18<04:45,  3.73it/s]
overall:  61%|██████▏   | 2451/4000 [12:21<06:47,  3.80it/s]

train:  70%|██████▉   | 2451/3515 [12:18<04:40,  3.80it/s]
overall:  61%|██████▏   | 2452/4000 [12:21<06:50,  3.77it/s]

train:  70%|██████▉   | 2452/3515 [12:19<04:42,  3.77it/s]
overall:  61%|██████▏   | 2453/4000 [12:21<06:38,  3.88it/s]

train:  70%|██████▉   | 2453/3515 [12:19<04:33,  3.88it/s]
overall:  61%|██████▏   | 2454/4000 [12:22<06:44,  3.83it/s]

train:  70%|██████▉   | 2454/3515 [12:19<04:37,  3.83it/s]
overall:  61%|██████▏   | 2455/4000 [12:22<06:36,  3.90it/s]

train:  70%|██████▉   | 2455/3515 [12:19<04:31,  3.90it/s]
overall:  61%|██████▏   | 2456/4000 [12:22<06:36,  3.89it/s]

train:  70%|██████▉   | 2456/3515 [12:20<04:32,  3.89it/s]
overall:  61%|██████▏   | 2457/4000 [12:23<08:12,  3.14it/s]

train:  70%|██████▉   | 2457/3515 [12:20<05:37,  3.14it/s]
overall:  61%|██████▏   | 2458/4000 [12:23<07:47,  3.30it/s]

train:  70%|██████▉   | 2458/3515 [12:20<05:20,  3.30it/s]
overall:  61%|██████▏   | 2459/4000 [12:23<07:17,  3.52it/s]

train:  70%|██████▉   | 2459/3515 [12:21<05:00,  3.52it/s]train loss: 0.5500720143318176
train per: 0.15735626558530844

overall:  62%|██████▏   | 2460/4000 [12:23<07:08,  3.59it/s]

train:  70%|██████▉   | 2460/3515 [12:21<04:53,  3.59it/s]
overall:  62%|██████▏   | 2461/4000 [12:24<07:36,  3.37it/s]

train:  70%|███████   | 2461/3515 [12:21<05:12,  3.37it/s]
overall:  62%|██████▏   | 2462/4000 [12:24<07:22,  3.47it/s]

train:  70%|███████   | 2462/3515 [12:21<05:03,  3.47it/s]
overall:  62%|██████▏   | 2463/4000 [12:24<07:07,  3.60it/s]

train:  70%|███████   | 2463/3515 [12:22<04:52,  3.60it/s]
overall:  62%|██████▏   | 2464/4000 [12:25<07:41,  3.33it/s]

train:  70%|███████   | 2464/3515 [12:22<05:15,  3.33it/s]
overall:  62%|██████▏   | 2465/4000 [12:25<06:21,  4.03it/s]

train:  70%|███████   | 2465/3515 [12:22<04:20,  4.03it/s]
overall:  62%|██████▏   | 2466/4000 [12:25<05:26,  4.69it/s]

train:  70%|███████   | 2466/3515 [12:22<03:43,  4.70it/s]
overall:  62%|██████▏   | 2467/4000 [12:25<05:55,  4.31it/s]

train:  70%|███████   | 2467/3515 [12:23<04:03,  4.31it/s]
overall:  62%|██████▏   | 2468/4000 [12:25<06:08,  4.16it/s]

train:  70%|███████   | 2468/3515 [12:23<04:11,  4.16it/s]
overall:  62%|██████▏   | 2469/4000 [12:26<06:18,  4.05it/s]

train:  70%|███████   | 2469/3515 [12:23<04:18,  4.05it/s]
overall:  62%|██████▏   | 2470/4000 [12:26<06:27,  3.95it/s]

train:  70%|███████   | 2470/3515 [12:23<04:24,  3.95it/s]
overall:  62%|██████▏   | 2471/4000 [12:26<06:36,  3.85it/s]

train:  70%|███████   | 2471/3515 [12:24<04:31,  3.85it/s]
overall:  62%|██████▏   | 2472/4000 [12:27<06:38,  3.83it/s]

train:  70%|███████   | 2472/3515 [12:24<04:32,  3.83it/s]
overall:  62%|██████▏   | 2473/4000 [12:27<06:30,  3.92it/s]

train:  70%|███████   | 2473/3515 [12:24<04:26,  3.92it/s]
overall:  62%|██████▏   | 2474/4000 [12:27<06:24,  3.97it/s]

train:  70%|███████   | 2474/3515 [12:24<04:22,  3.97it/s]
overall:  62%|██████▏   | 2475/4000 [12:27<06:20,  4.00it/s]

train:  70%|███████   | 2475/3515 [12:25<04:19,  4.00it/s]
overall:  62%|██████▏   | 2476/4000 [12:28<06:18,  4.03it/s]

train:  70%|███████   | 2476/3515 [12:25<04:17,  4.03it/s]
overall:  62%|██████▏   | 2477/4000 [12:28<06:17,  4.03it/s]

train:  70%|███████   | 2477/3515 [12:25<04:17,  4.03it/s]
overall:  62%|██████▏   | 2478/4000 [12:28<06:08,  4.13it/s]

train:  70%|███████   | 2478/3515 [12:25<04:11,  4.13it/s]
overall:  62%|██████▏   | 2479/4000 [12:28<06:03,  4.18it/s]

train:  71%|███████   | 2479/3515 [12:26<04:07,  4.18it/s]train loss: 0.5571133494377136
train per: 0.16453173168916765

overall:  62%|██████▏   | 2480/4000 [12:28<06:09,  4.11it/s]

train:  71%|███████   | 2480/3515 [12:26<04:12,  4.11it/s]
overall:  62%|██████▏   | 2481/4000 [12:29<07:24,  3.42it/s]

train:  71%|███████   | 2481/3515 [12:26<05:02,  3.42it/s]
overall:  62%|██████▏   | 2482/4000 [12:29<07:17,  3.47it/s]

train:  71%|███████   | 2482/3515 [12:27<04:57,  3.47it/s]
overall:  62%|██████▏   | 2483/4000 [12:29<06:52,  3.68it/s]

train:  71%|███████   | 2483/3515 [12:27<04:40,  3.68it/s]
overall:  62%|██████▏   | 2484/4000 [12:30<06:51,  3.68it/s]

train:  71%|███████   | 2484/3515 [12:27<04:39,  3.69it/s]
overall:  62%|██████▏   | 2485/4000 [12:30<07:07,  3.55it/s]

train:  71%|███████   | 2485/3515 [12:27<04:50,  3.55it/s]
overall:  62%|██████▏   | 2486/4000 [12:30<07:03,  3.57it/s]

train:  71%|███████   | 2486/3515 [12:28<04:47,  3.57it/s]
overall:  62%|██████▏   | 2487/4000 [12:31<06:56,  3.63it/s]

train:  71%|███████   | 2487/3515 [12:28<04:43,  3.63it/s]
overall:  62%|██████▏   | 2488/4000 [12:31<07:22,  3.42it/s]

train:  71%|███████   | 2488/3515 [12:28<05:00,  3.42it/s]
overall:  62%|██████▏   | 2489/4000 [12:31<06:43,  3.74it/s]

train:  71%|███████   | 2489/3515 [12:28<04:34,  3.74it/s]
overall:  62%|██████▏   | 2490/4000 [12:31<06:38,  3.79it/s]

train:  71%|███████   | 2490/3515 [12:29<04:30,  3.79it/s]
overall:  62%|██████▏   | 2491/4000 [12:32<06:42,  3.75it/s]

train:  71%|███████   | 2491/3515 [12:29<04:33,  3.75it/s]
overall:  62%|██████▏   | 2492/4000 [12:32<06:42,  3.75it/s]

train:  71%|███████   | 2492/3515 [12:29<04:33,  3.75it/s]
overall:  62%|██████▏   | 2493/4000 [12:32<06:21,  3.95it/s]

train:  71%|███████   | 2493/3515 [12:29<04:18,  3.95it/s]
overall:  62%|██████▏   | 2494/4000 [12:32<05:12,  4.82it/s]

train:  71%|███████   | 2494/3515 [12:30<03:31,  4.82it/s]
overall:  62%|██████▏   | 2495/4000 [12:32<05:40,  4.42it/s]

train:  71%|███████   | 2495/3515 [12:30<03:50,  4.42it/s]
overall:  62%|██████▏   | 2496/4000 [12:33<06:00,  4.18it/s]

train:  71%|███████   | 2496/3515 [12:30<04:03,  4.18it/s]
overall:  62%|██████▏   | 2497/4000 [12:33<05:56,  4.21it/s]

train:  71%|███████   | 2497/3515 [12:30<04:01,  4.21it/s]
overall:  62%|██████▏   | 2498/4000 [12:33<05:58,  4.19it/s]

train:  71%|███████   | 2498/3515 [12:31<04:03,  4.19it/s]
overall:  62%|██████▏   | 2499/4000 [12:33<06:17,  3.98it/s]

train:  71%|███████   | 2499/3515 [12:31<04:15,  3.98it/s]train loss: 0.553031325340271
train per: 0.16120545947072307

overall:  62%|██████▎   | 2500/4000 [12:34<06:33,  3.81it/s]

train:  71%|███████   | 2500/3515 [12:31<04:26,  3.81it/s]
overall:  63%|██████▎   | 2501/4000 [12:34<05:37,  4.45it/s]

train:  71%|███████   | 2501/3515 [12:31<03:47,  4.45it/s]
overall:  63%|██████▎   | 2502/4000 [12:34<05:45,  4.33it/s]

train:  71%|███████   | 2502/3515 [12:32<03:53,  4.33it/s]
overall:  63%|██████▎   | 2503/4000 [12:34<06:06,  4.09it/s]

train:  71%|███████   | 2503/3515 [12:32<04:07,  4.09it/s]
overall:  63%|██████▎   | 2504/4000 [12:35<06:12,  4.01it/s]

train:  71%|███████   | 2504/3515 [12:32<04:11,  4.01it/s]
overall:  63%|██████▎   | 2505/4000 [12:35<07:38,  3.26it/s]

train:  71%|███████▏  | 2505/3515 [12:33<05:09,  3.26it/s]
overall:  63%|██████▎   | 2506/4000 [12:35<06:53,  3.62it/s]

train:  71%|███████▏  | 2506/3515 [12:33<04:38,  3.62it/s]
overall:  63%|██████▎   | 2507/4000 [12:36<06:50,  3.63it/s]

train:  71%|███████▏  | 2507/3515 [12:33<04:37,  3.63it/s]
overall:  63%|██████▎   | 2508/4000 [12:36<06:41,  3.71it/s]

train:  71%|███████▏  | 2508/3515 [12:33<04:31,  3.71it/s]
overall:  63%|██████▎   | 2509/4000 [12:36<05:47,  4.29it/s]

train:  71%|███████▏  | 2509/3515 [12:33<03:54,  4.29it/s]
overall:  63%|██████▎   | 2510/4000 [12:36<05:12,  4.77it/s]

train:  71%|███████▏  | 2510/3515 [12:34<03:30,  4.77it/s]
overall:  63%|██████▎   | 2511/4000 [12:36<05:38,  4.39it/s]

train:  71%|███████▏  | 2511/3515 [12:34<03:48,  4.39it/s]
overall:  63%|██████▎   | 2512/4000 [12:37<05:57,  4.16it/s]

train:  71%|███████▏  | 2512/3515 [12:34<04:01,  4.16it/s]
overall:  63%|██████▎   | 2513/4000 [12:37<05:57,  4.16it/s]

train:  71%|███████▏  | 2513/3515 [12:34<04:00,  4.16it/s]
overall:  63%|██████▎   | 2514/4000 [12:37<06:06,  4.05it/s]

train:  72%|███████▏  | 2514/3515 [12:35<04:07,  4.05it/s]
overall:  63%|██████▎   | 2515/4000 [12:37<06:25,  3.85it/s]

train:  72%|███████▏  | 2515/3515 [12:35<04:19,  3.85it/s]
overall:  63%|██████▎   | 2516/4000 [12:38<06:15,  3.95it/s]

train:  72%|███████▏  | 2516/3515 [12:35<04:12,  3.95it/s]
overall:  63%|██████▎   | 2517/4000 [12:38<06:19,  3.91it/s]

train:  72%|███████▏  | 2517/3515 [12:35<04:15,  3.91it/s]
overall:  63%|██████▎   | 2518/4000 [12:38<06:24,  3.85it/s]

train:  72%|███████▏  | 2518/3515 [12:36<04:18,  3.85it/s]
overall:  63%|██████▎   | 2519/4000 [12:39<06:31,  3.78it/s]

train:  72%|███████▏  | 2519/3515 [12:36<04:23,  3.78it/s]train loss: 0.6462507843971252
train per: 0.17492140390661237

overall:  63%|██████▎   | 2520/4000 [12:39<05:35,  4.41it/s]

train:  72%|███████▏  | 2520/3515 [12:36<03:46,  4.40it/s]
overall:  63%|██████▎   | 2521/4000 [12:39<05:11,  4.75it/s]

train:  72%|███████▏  | 2521/3515 [12:36<03:29,  4.75it/s]
overall:  63%|██████▎   | 2522/4000 [12:39<05:33,  4.44it/s]

train:  72%|███████▏  | 2522/3515 [12:36<03:43,  4.44it/s]
overall:  63%|██████▎   | 2523/4000 [12:39<05:54,  4.16it/s]

train:  72%|███████▏  | 2523/3515 [12:37<03:58,  4.16it/s]
overall:  63%|██████▎   | 2524/4000 [12:40<06:12,  3.96it/s]

train:  72%|███████▏  | 2524/3515 [12:37<04:10,  3.96it/s]
overall:  63%|██████▎   | 2525/4000 [12:40<06:15,  3.92it/s]

train:  72%|███████▏  | 2525/3515 [12:37<04:12,  3.92it/s]
overall:  63%|██████▎   | 2526/4000 [12:40<06:21,  3.87it/s]

train:  72%|███████▏  | 2526/3515 [12:38<04:15,  3.87it/s]
overall:  63%|██████▎   | 2527/4000 [12:40<06:19,  3.88it/s]

train:  72%|███████▏  | 2527/3515 [12:38<04:14,  3.88it/s]
overall:  63%|██████▎   | 2529/4000 [12:41<06:25,  3.81it/s]

train:  72%|███████▏  | 2529/3515 [12:38<04:18,  3.81it/s]
overall:  63%|██████▎   | 2530/4000 [12:41<06:24,  3.82it/s]

train:  72%|███████▏  | 2530/3515 [12:39<04:17,  3.82it/s]
overall:  63%|██████▎   | 2531/4000 [12:42<06:27,  3.79it/s]

train:  72%|███████▏  | 2531/3515 [12:39<04:19,  3.79it/s]
overall:  63%|██████▎   | 2532/4000 [12:42<05:25,  4.51it/s]

train:  72%|███████▏  | 2532/3515 [12:39<03:38,  4.51it/s]
overall:  63%|██████▎   | 2533/4000 [12:42<06:07,  3.99it/s]

train:  72%|███████▏  | 2533/3515 [12:39<04:05,  3.99it/s]
overall:  63%|██████▎   | 2534/4000 [12:42<06:13,  3.93it/s]

train:  72%|███████▏  | 2534/3515 [12:40<04:09,  3.93it/s]
overall:  63%|██████▎   | 2535/4000 [12:42<06:20,  3.85it/s]

train:  72%|███████▏  | 2535/3515 [12:40<04:14,  3.85it/s]
overall:  63%|██████▎   | 2536/4000 [12:43<06:49,  3.58it/s]

train:  72%|███████▏  | 2536/3515 [12:40<04:33,  3.57it/s]
overall:  63%|██████▎   | 2537/4000 [12:43<06:42,  3.64it/s]

train:  72%|███████▏  | 2537/3515 [12:40<04:28,  3.64it/s]
overall:  63%|██████▎   | 2538/4000 [12:43<06:31,  3.73it/s]

train:  72%|███████▏  | 2538/3515 [12:41<04:21,  3.73it/s]
overall:  63%|██████▎   | 2539/4000 [12:44<06:36,  3.69it/s]

train:  72%|███████▏  | 2539/3515 [12:41<04:24,  3.69it/s]train loss: 0.5783843994140625
train per: 0.16876305061151436

overall:  64%|██████▎   | 2540/4000 [12:44<06:41,  3.63it/s]

train:  72%|███████▏  | 2540/3515 [12:41<04:28,  3.63it/s]
overall:  64%|██████▎   | 2541/4000 [12:44<05:40,  4.29it/s]

train:  72%|███████▏  | 2541/3515 [12:41<03:47,  4.29it/s]
overall:  64%|██████▎   | 2542/4000 [12:44<05:42,  4.25it/s]

train:  72%|███████▏  | 2542/3515 [12:42<03:48,  4.25it/s]
overall:  64%|██████▎   | 2543/4000 [12:45<05:55,  4.10it/s]

train:  72%|███████▏  | 2543/3515 [12:42<03:57,  4.10it/s]
overall:  64%|██████▎   | 2544/4000 [12:45<06:07,  3.97it/s]

train:  72%|███████▏  | 2544/3515 [12:42<04:04,  3.97it/s]
overall:  64%|██████▎   | 2545/4000 [12:45<06:11,  3.92it/s]

train:  72%|███████▏  | 2545/3515 [12:42<04:07,  3.92it/s]
overall:  64%|██████▎   | 2546/4000 [12:45<06:15,  3.87it/s]

train:  72%|███████▏  | 2546/3515 [12:43<04:10,  3.87it/s]
overall:  64%|██████▎   | 2547/4000 [12:46<06:25,  3.77it/s]

train:  72%|███████▏  | 2547/3515 [12:43<04:16,  3.77it/s]
overall:  64%|██████▎   | 2548/4000 [12:46<06:22,  3.80it/s]

train:  72%|███████▏  | 2548/3515 [12:43<04:14,  3.80it/s]
overall:  64%|██████▎   | 2549/4000 [12:46<06:22,  3.80it/s]

train:  73%|███████▎  | 2549/3515 [12:44<04:14,  3.80it/s]
overall:  64%|██████▍   | 2550/4000 [12:46<06:29,  3.73it/s]

train:  73%|███████▎  | 2550/3515 [12:44<04:18,  3.73it/s]
overall:  64%|██████▍   | 2551/4000 [12:47<06:27,  3.74it/s]

train:  73%|███████▎  | 2551/3515 [12:44<04:17,  3.74it/s]
overall:  64%|██████▍   | 2552/4000 [12:47<06:12,  3.88it/s]

train:  73%|███████▎  | 2552/3515 [12:44<04:07,  3.88it/s]
overall:  64%|██████▍   | 2553/4000 [12:47<07:34,  3.18it/s]

train:  73%|███████▎  | 2553/3515 [12:45<05:02,  3.18it/s]
overall:  64%|██████▍   | 2554/4000 [12:48<07:17,  3.31it/s]

train:  73%|███████▎  | 2554/3515 [12:45<04:50,  3.31it/s]
overall:  64%|██████▍   | 2555/4000 [12:48<06:47,  3.54it/s]

train:  73%|███████▎  | 2555/3515 [12:45<04:31,  3.54it/s]
overall:  64%|██████▍   | 2556/4000 [12:48<06:49,  3.52it/s]

train:  73%|███████▎  | 2556/3515 [12:46<04:32,  3.52it/s]
overall:  64%|██████▍   | 2557/4000 [12:48<07:05,  3.39it/s]

train:  73%|███████▎  | 2557/3515 [12:46<04:42,  3.39it/s]
overall:  64%|██████▍   | 2558/4000 [12:49<06:53,  3.49it/s]

train:  73%|███████▎  | 2558/3515 [12:46<04:34,  3.49it/s]
overall:  64%|██████▍   | 2559/4000 [12:49<06:38,  3.62it/s]

train:  73%|███████▎  | 2559/3515 [12:46<04:24,  3.62it/s]train loss: 0.5624681115150452
train per: 0.16466474421561728

overall:  64%|██████▍   | 2560/4000 [12:49<07:09,  3.35it/s]

train:  73%|███████▎  | 2560/3515 [12:47<04:45,  3.35it/s]
overall:  64%|██████▍   | 2561/4000 [12:50<06:54,  3.47it/s]

train:  73%|███████▎  | 2561/3515 [12:47<04:35,  3.47it/s]
overall:  64%|██████▍   | 2562/4000 [12:50<06:36,  3.62it/s]

train:  73%|███████▎  | 2562/3515 [12:47<04:22,  3.62it/s]
overall:  64%|██████▍   | 2563/4000 [12:50<06:35,  3.63it/s]

train:  73%|███████▎  | 2563/3515 [12:48<04:22,  3.63it/s]
overall:  64%|██████▍   | 2564/4000 [12:50<06:30,  3.68it/s]

train:  73%|███████▎  | 2564/3515 [12:48<04:18,  3.68it/s]
overall:  64%|██████▍   | 2565/4000 [12:51<06:23,  3.75it/s]

train:  73%|███████▎  | 2565/3515 [12:48<04:13,  3.75it/s]
overall:  64%|██████▍   | 2566/4000 [12:51<06:21,  3.75it/s]

train:  73%|███████▎  | 2566/3515 [12:48<04:12,  3.75it/s]
overall:  64%|██████▍   | 2567/4000 [12:51<06:24,  3.72it/s]

train:  73%|███████▎  | 2567/3515 [12:49<04:14,  3.72it/s]
overall:  64%|██████▍   | 2568/4000 [12:51<06:14,  3.82it/s]

train:  73%|███████▎  | 2568/3515 [12:49<04:07,  3.82it/s]
overall:  64%|██████▍   | 2569/4000 [12:52<06:08,  3.88it/s]

train:  73%|███████▎  | 2569/3515 [12:49<04:03,  3.88it/s]
overall:  64%|██████▍   | 2570/4000 [12:52<06:14,  3.81it/s]

train:  73%|███████▎  | 2570/3515 [12:49<04:07,  3.81it/s]
overall:  64%|██████▍   | 2571/4000 [12:52<05:19,  4.47it/s]

train:  73%|███████▎  | 2571/3515 [12:49<03:31,  4.47it/s]
overall:  64%|██████▍   | 2572/4000 [12:52<05:40,  4.19it/s]

train:  73%|███████▎  | 2572/3515 [12:50<03:45,  4.19it/s]
overall:  64%|██████▍   | 2573/4000 [12:53<05:56,  4.00it/s]

train:  73%|███████▎  | 2573/3515 [12:50<03:55,  4.00it/s]
overall:  64%|██████▍   | 2574/4000 [12:53<06:05,  3.91it/s]

train:  73%|███████▎  | 2574/3515 [12:50<04:00,  3.91it/s]
overall:  64%|██████▍   | 2575/4000 [12:53<05:10,  4.59it/s]

train:  73%|███████▎  | 2575/3515 [12:50<03:24,  4.59it/s]
overall:  64%|██████▍   | 2576/4000 [12:53<05:32,  4.28it/s]

train:  73%|███████▎  | 2576/3515 [12:51<03:39,  4.28it/s]
overall:  64%|██████▍   | 2577/4000 [12:54<06:36,  3.59it/s]

train:  73%|███████▎  | 2577/3515 [12:51<04:21,  3.59it/s]
overall:  64%|██████▍   | 2578/4000 [12:54<06:30,  3.64it/s]

train:  73%|███████▎  | 2578/3515 [12:51<04:17,  3.64it/s]
overall:  64%|██████▍   | 2579/4000 [12:54<06:27,  3.67it/s]

train:  73%|███████▎  | 2579/3515 [12:52<04:15,  3.67it/s]train loss: 0.556399941444397
train per: 0.16209373097991478

overall:  64%|██████▍   | 2580/4000 [12:54<05:43,  4.14it/s]

train:  73%|███████▎  | 2580/3515 [12:52<03:46,  4.14it/s]
overall:  65%|██████▍   | 2581/4000 [12:55<05:36,  4.21it/s]

train:  73%|███████▎  | 2581/3515 [12:52<03:41,  4.21it/s]
overall:  65%|██████▍   | 2582/4000 [12:55<05:51,  4.03it/s]

train:  73%|███████▎  | 2582/3515 [12:52<03:51,  4.03it/s]
overall:  65%|██████▍   | 2583/4000 [12:55<06:02,  3.91it/s]

train:  73%|███████▎  | 2583/3515 [12:53<03:58,  3.91it/s]
overall:  65%|██████▍   | 2584/4000 [12:56<06:34,  3.59it/s]

train:  74%|███████▎  | 2584/3515 [12:53<04:19,  3.59it/s]
overall:  65%|██████▍   | 2586/4000 [12:56<05:00,  4.71it/s]

train:  74%|███████▎  | 2586/3515 [12:53<03:17,  4.71it/s]
overall:  65%|██████▍   | 2587/4000 [12:56<05:22,  4.39it/s]

train:  74%|███████▎  | 2587/3515 [12:53<03:31,  4.39it/s]
overall:  65%|██████▍   | 2588/4000 [12:56<05:27,  4.31it/s]

train:  74%|███████▎  | 2588/3515 [12:54<03:34,  4.31it/s]
overall:  65%|██████▍   | 2589/4000 [12:57<05:39,  4.16it/s]

train:  74%|███████▎  | 2589/3515 [12:54<03:42,  4.16it/s]
overall:  65%|██████▍   | 2590/4000 [12:57<05:52,  4.00it/s]

train:  74%|███████▎  | 2590/3515 [12:54<03:51,  4.00it/s]
overall:  65%|██████▍   | 2591/4000 [12:57<05:47,  4.06it/s]

train:  74%|███████▎  | 2591/3515 [12:54<03:47,  4.06it/s]
overall:  65%|██████▍   | 2592/4000 [12:57<06:00,  3.91it/s]

train:  74%|███████▎  | 2592/3515 [12:55<03:56,  3.91it/s]
overall:  65%|██████▍   | 2593/4000 [12:58<06:05,  3.85it/s]

train:  74%|███████▍  | 2593/3515 [12:55<03:59,  3.85it/s]
overall:  65%|██████▍   | 2594/4000 [12:58<05:31,  4.25it/s]

train:  74%|███████▍  | 2594/3515 [12:55<03:36,  4.25it/s]
overall:  65%|██████▍   | 2595/4000 [12:58<05:39,  4.14it/s]

train:  74%|███████▍  | 2595/3515 [12:55<03:42,  4.14it/s]
overall:  65%|██████▍   | 2596/4000 [12:58<05:53,  3.97it/s]

train:  74%|███████▍  | 2596/3515 [12:56<03:51,  3.97it/s]
overall:  65%|██████▍   | 2597/4000 [12:59<06:05,  3.84it/s]

train:  74%|███████▍  | 2597/3515 [12:56<03:59,  3.84it/s]
overall:  65%|██████▍   | 2598/4000 [12:59<06:06,  3.83it/s]

train:  74%|███████▍  | 2598/3515 [12:56<03:59,  3.83it/s]train loss: 0.5967673659324646
train per: 0.17107574526100522

overall:  65%|██████▌   | 2600/4000 [12:59<04:58,  4.69it/s]

train:  74%|███████▍  | 2600/3515 [12:57<03:15,  4.69it/s]
overall:  65%|██████▌   | 2601/4000 [13:00<05:34,  4.19it/s]

train:  74%|███████▍  | 2601/3515 [12:57<03:38,  4.19it/s]
overall:  65%|██████▌   | 2602/4000 [13:00<05:35,  4.17it/s]

train:  74%|███████▍  | 2602/3515 [12:57<03:38,  4.17it/s]
overall:  65%|██████▌   | 2603/4000 [13:00<05:47,  4.02it/s]

train:  74%|███████▍  | 2603/3515 [12:57<03:46,  4.02it/s]
overall:  65%|██████▌   | 2605/4000 [13:00<05:16,  4.41it/s]

train:  74%|███████▍  | 2605/3515 [12:58<03:26,  4.41it/s]
overall:  65%|██████▌   | 2606/4000 [13:01<05:09,  4.50it/s]

train:  74%|███████▍  | 2606/3515 [12:58<03:21,  4.50it/s]
overall:  65%|██████▌   | 2607/4000 [13:01<05:21,  4.33it/s]

train:  74%|███████▍  | 2607/3515 [12:58<03:29,  4.33it/s]
overall:  65%|██████▌   | 2608/4000 [13:01<05:47,  4.01it/s]

train:  74%|███████▍  | 2608/3515 [12:59<03:46,  4.01it/s]
overall:  65%|██████▌   | 2609/4000 [13:01<05:46,  4.01it/s]

train:  74%|███████▍  | 2609/3515 [12:59<03:45,  4.01it/s]
overall:  65%|██████▌   | 2610/4000 [13:02<05:18,  4.37it/s]

train:  74%|███████▍  | 2610/3515 [12:59<03:27,  4.37it/s]
overall:  65%|██████▌   | 2611/4000 [13:02<05:39,  4.09it/s]

train:  74%|███████▍  | 2611/3515 [12:59<03:41,  4.09it/s]
overall:  65%|██████▌   | 2612/4000 [13:02<05:37,  4.11it/s]

train:  74%|███████▍  | 2612/3515 [13:00<03:39,  4.11it/s]
overall:  65%|██████▌   | 2613/4000 [13:02<05:36,  4.12it/s]

train:  74%|███████▍  | 2613/3515 [13:00<03:38,  4.12it/s]
overall:  65%|██████▌   | 2614/4000 [13:03<05:43,  4.03it/s]

train:  74%|███████▍  | 2614/3515 [13:00<03:43,  4.03it/s]
overall:  65%|██████▌   | 2615/4000 [13:03<05:51,  3.94it/s]

train:  74%|███████▍  | 2615/3515 [13:00<03:48,  3.94it/s]
overall:  65%|██████▌   | 2616/4000 [13:03<05:57,  3.87it/s]

train:  74%|███████▍  | 2616/3515 [13:01<03:52,  3.87it/s]
overall:  65%|██████▌   | 2617/4000 [13:03<05:59,  3.84it/s]

train:  74%|███████▍  | 2617/3515 [13:01<03:53,  3.84it/s]
overall:  65%|██████▌   | 2618/4000 [13:04<05:49,  3.95it/s]

train:  74%|███████▍  | 2618/3515 [13:01<03:46,  3.95it/s]
overall:  65%|██████▌   | 2619/4000 [13:04<05:51,  3.93it/s]

train:  75%|███████▍  | 2619/3515 [13:01<03:47,  3.93it/s]train loss: 0.5864554643630981
train per: 0.16941286318448948

overall:  66%|██████▌   | 2620/4000 [13:04<05:47,  3.97it/s]

train:  75%|███████▍  | 2620/3515 [13:02<03:45,  3.97it/s]
overall:  66%|██████▌   | 2621/4000 [13:04<05:51,  3.93it/s]

train:  75%|███████▍  | 2621/3515 [13:02<03:47,  3.93it/s]
overall:  66%|██████▌   | 2622/4000 [13:05<05:49,  3.94it/s]

train:  75%|███████▍  | 2622/3515 [13:02<03:46,  3.94it/s]
overall:  66%|██████▌   | 2623/4000 [13:05<05:54,  3.89it/s]

train:  75%|███████▍  | 2623/3515 [13:02<03:49,  3.89it/s]
overall:  66%|██████▌   | 2624/4000 [13:05<05:48,  3.95it/s]

train:  75%|███████▍  | 2624/3515 [13:03<03:45,  3.95it/s]
overall:  66%|██████▌   | 2625/4000 [13:06<07:13,  3.17it/s]

train:  75%|███████▍  | 2625/3515 [13:03<04:40,  3.17it/s]
overall:  66%|██████▌   | 2626/4000 [13:06<06:51,  3.34it/s]

train:  75%|███████▍  | 2626/3515 [13:03<04:26,  3.34it/s]
overall:  66%|██████▌   | 2627/4000 [13:06<06:34,  3.48it/s]

train:  75%|███████▍  | 2627/3515 [13:04<04:15,  3.48it/s]
overall:  66%|██████▌   | 2628/4000 [13:06<06:35,  3.47it/s]

train:  75%|███████▍  | 2628/3515 [13:04<04:15,  3.47it/s]
overall:  66%|██████▌   | 2629/4000 [13:07<06:12,  3.68it/s]

train:  75%|███████▍  | 2629/3515 [13:04<04:00,  3.68it/s]
overall:  66%|██████▌   | 2630/4000 [13:07<06:03,  3.76it/s]

train:  75%|███████▍  | 2630/3515 [13:04<03:55,  3.76it/s]
overall:  66%|██████▌   | 2631/4000 [13:07<05:03,  4.51it/s]

train:  75%|███████▍  | 2631/3515 [13:04<03:15,  4.51it/s]
overall:  66%|██████▌   | 2632/4000 [13:07<05:33,  4.10it/s]

train:  75%|███████▍  | 2632/3515 [13:05<03:35,  4.10it/s]
overall:  66%|██████▌   | 2633/4000 [13:08<05:41,  4.01it/s]

train:  75%|███████▍  | 2633/3515 [13:05<03:40,  4.01it/s]
overall:  66%|██████▌   | 2634/4000 [13:08<05:09,  4.42it/s]

train:  75%|███████▍  | 2634/3515 [13:05<03:19,  4.42it/s]
overall:  66%|██████▌   | 2635/4000 [13:08<05:27,  4.17it/s]

train:  75%|███████▍  | 2635/3515 [13:05<03:31,  4.17it/s]
overall:  66%|██████▌   | 2636/4000 [13:08<05:40,  4.01it/s]

train:  75%|███████▍  | 2636/3515 [13:06<03:39,  4.01it/s]
overall:  66%|██████▌   | 2637/4000 [13:08<04:58,  4.56it/s]

train:  75%|███████▌  | 2637/3515 [13:06<03:12,  4.56it/s]
overall:  66%|██████▌   | 2638/4000 [13:09<05:18,  4.28it/s]

train:  75%|███████▌  | 2638/3515 [13:06<03:25,  4.28it/s]
overall:  66%|██████▌   | 2639/4000 [13:09<05:34,  4.07it/s]

train:  75%|███████▌  | 2639/3515 [13:06<03:35,  4.07it/s]train loss: 0.5867980718612671
train per: 0.17249285080366827

overall:  66%|██████▌   | 2640/4000 [13:09<05:49,  3.89it/s]

train:  75%|███████▌  | 2640/3515 [13:07<03:45,  3.88it/s]
overall:  66%|██████▌   | 2641/4000 [13:10<05:52,  3.86it/s]

train:  75%|███████▌  | 2641/3515 [13:07<03:46,  3.86it/s]
overall:  66%|██████▌   | 2642/4000 [13:10<05:53,  3.84it/s]

train:  75%|███████▌  | 2642/3515 [13:07<03:47,  3.84it/s]
overall:  66%|██████▌   | 2643/4000 [13:10<05:45,  3.92it/s]

train:  75%|███████▌  | 2643/3515 [13:07<03:42,  3.93it/s]
overall:  66%|██████▌   | 2644/4000 [13:10<05:53,  3.83it/s]

train:  75%|███████▌  | 2644/3515 [13:08<03:47,  3.83it/s]
overall:  66%|██████▌   | 2645/4000 [13:11<05:46,  3.91it/s]

train:  75%|███████▌  | 2645/3515 [13:08<03:42,  3.91it/s]
overall:  66%|██████▌   | 2646/4000 [13:11<05:48,  3.88it/s]

train:  75%|███████▌  | 2646/3515 [13:08<03:43,  3.88it/s]
overall:  66%|██████▌   | 2647/4000 [13:11<05:31,  4.08it/s]

train:  75%|███████▌  | 2647/3515 [13:08<03:32,  4.08it/s]
overall:  66%|██████▌   | 2648/4000 [13:11<05:48,  3.88it/s]

train:  75%|███████▌  | 2648/3515 [13:09<03:43,  3.88it/s]
overall:  66%|██████▌   | 2649/4000 [13:12<07:00,  3.22it/s]

train:  75%|███████▌  | 2649/3515 [13:09<04:29,  3.22it/s]
overall:  66%|██████▋   | 2650/4000 [13:12<06:19,  3.56it/s]

train:  75%|███████▌  | 2650/3515 [13:09<04:02,  3.56it/s]
overall:  66%|██████▋   | 2651/4000 [13:12<05:49,  3.86it/s]

train:  75%|███████▌  | 2651/3515 [13:10<03:43,  3.86it/s]
overall:  66%|██████▋   | 2652/4000 [13:13<05:57,  3.77it/s]

train:  75%|███████▌  | 2652/3515 [13:10<03:49,  3.77it/s]
overall:  66%|██████▋   | 2653/4000 [13:13<05:24,  4.15it/s]

train:  75%|███████▌  | 2653/3515 [13:10<03:27,  4.15it/s]
overall:  66%|██████▋   | 2654/4000 [13:13<05:45,  3.89it/s]

train:  76%|███████▌  | 2654/3515 [13:10<03:41,  3.89it/s]
overall:  66%|██████▋   | 2655/4000 [13:13<05:51,  3.82it/s]

train:  76%|███████▌  | 2655/3515 [13:11<03:44,  3.82it/s]
overall:  66%|██████▋   | 2656/4000 [13:14<06:16,  3.57it/s]

train:  76%|███████▌  | 2656/3515 [13:11<04:00,  3.57it/s]
overall:  66%|██████▋   | 2657/4000 [13:14<06:08,  3.65it/s]

train:  76%|███████▌  | 2657/3515 [13:11<03:55,  3.65it/s]
overall:  66%|██████▋   | 2658/4000 [13:14<06:01,  3.71it/s]

train:  76%|███████▌  | 2658/3515 [13:11<03:50,  3.71it/s]
overall:  66%|██████▋   | 2659/4000 [13:14<06:02,  3.70it/s]

train:  76%|███████▌  | 2659/3515 [13:12<03:51,  3.70it/s]train loss: 0.5435940027236938
train per: 0.15944711835122793

overall:  66%|██████▋   | 2660/4000 [13:15<05:59,  3.72it/s]

train:  76%|███████▌  | 2660/3515 [13:12<03:49,  3.72it/s]
overall:  67%|██████▋   | 2661/4000 [13:15<05:54,  3.78it/s]

train:  76%|███████▌  | 2661/3515 [13:12<03:45,  3.78it/s]
overall:  67%|██████▋   | 2662/4000 [13:15<05:31,  4.03it/s]

train:  76%|███████▌  | 2662/3515 [13:12<03:31,  4.03it/s]
overall:  67%|██████▋   | 2663/4000 [13:15<05:37,  3.96it/s]

train:  76%|███████▌  | 2663/3515 [13:13<03:35,  3.96it/s]
overall:  67%|██████▋   | 2664/4000 [13:16<05:35,  3.98it/s]

train:  76%|███████▌  | 2664/3515 [13:13<03:33,  3.98it/s]
overall:  67%|██████▋   | 2665/4000 [13:16<05:37,  3.96it/s]

train:  76%|███████▌  | 2665/3515 [13:13<03:34,  3.96it/s]
overall:  67%|██████▋   | 2666/4000 [13:16<05:40,  3.92it/s]

train:  76%|███████▌  | 2666/3515 [13:14<03:36,  3.92it/s]
overall:  67%|██████▋   | 2667/4000 [13:16<05:45,  3.86it/s]

train:  76%|███████▌  | 2667/3515 [13:14<03:39,  3.86it/s]
overall:  67%|██████▋   | 2668/4000 [13:17<05:33,  3.99it/s]

train:  76%|███████▌  | 2668/3515 [13:14<03:32,  3.99it/s]
overall:  67%|██████▋   | 2669/4000 [13:17<05:41,  3.90it/s]

train:  76%|███████▌  | 2669/3515 [13:14<03:36,  3.90it/s]
overall:  67%|██████▋   | 2670/4000 [13:17<05:22,  4.12it/s]

train:  76%|███████▌  | 2670/3515 [13:14<03:24,  4.12it/s]
overall:  67%|██████▋   | 2671/4000 [13:17<05:30,  4.02it/s]

train:  76%|███████▌  | 2671/3515 [13:15<03:29,  4.02it/s]
overall:  67%|██████▋   | 2672/4000 [13:18<05:34,  3.97it/s]

train:  76%|███████▌  | 2672/3515 [13:15<03:32,  3.97it/s]
overall:  67%|██████▋   | 2673/4000 [13:18<06:37,  3.34it/s]

train:  76%|███████▌  | 2673/3515 [13:15<04:11,  3.34it/s]
overall:  67%|██████▋   | 2674/4000 [13:18<06:18,  3.50it/s]

train:  76%|███████▌  | 2674/3515 [13:16<04:00,  3.50it/s]
overall:  67%|██████▋   | 2675/4000 [13:19<05:59,  3.69it/s]

train:  76%|███████▌  | 2675/3515 [13:16<03:47,  3.69it/s]
overall:  67%|██████▋   | 2676/4000 [13:19<06:02,  3.65it/s]

train:  76%|███████▌  | 2676/3515 [13:16<03:50,  3.65it/s]
overall:  67%|██████▋   | 2677/4000 [13:19<06:22,  3.46it/s]

train:  76%|███████▌  | 2677/3515 [13:17<04:01,  3.46it/s]
overall:  67%|██████▋   | 2678/4000 [13:19<06:19,  3.48it/s]

train:  76%|███████▌  | 2678/3515 [13:17<04:00,  3.48it/s]
overall:  67%|██████▋   | 2679/4000 [13:20<06:08,  3.59it/s]

train:  76%|███████▌  | 2679/3515 [13:17<03:53,  3.59it/s]train loss: 0.5415371656417847
train per: 0.15690673846383965

overall:  67%|██████▋   | 2680/4000 [13:20<06:30,  3.38it/s]

train:  76%|███████▌  | 2680/3515 [13:17<04:07,  3.37it/s]
overall:  67%|██████▋   | 2681/4000 [13:20<06:18,  3.48it/s]

train:  76%|███████▋  | 2681/3515 [13:18<03:59,  3.48it/s]
overall:  67%|██████▋   | 2682/4000 [13:21<06:07,  3.59it/s]

train:  76%|███████▋  | 2682/3515 [13:18<03:51,  3.59it/s]
overall:  67%|██████▋   | 2683/4000 [13:21<06:02,  3.63it/s]

train:  76%|███████▋  | 2683/3515 [13:18<03:49,  3.63it/s]
overall:  67%|██████▋   | 2684/4000 [13:21<06:05,  3.60it/s]

train:  76%|███████▋  | 2684/3515 [13:18<03:50,  3.60it/s]
overall:  67%|██████▋   | 2685/4000 [13:21<04:55,  4.45it/s]

train:  76%|███████▋  | 2685/3515 [13:19<03:06,  4.45it/s]
overall:  67%|██████▋   | 2686/4000 [13:21<05:02,  4.35it/s]

train:  76%|███████▋  | 2686/3515 [13:19<03:10,  4.35it/s]
overall:  67%|██████▋   | 2687/4000 [13:22<05:06,  4.29it/s]

train:  76%|███████▋  | 2687/3515 [13:19<03:13,  4.29it/s]
overall:  67%|██████▋   | 2688/4000 [13:22<05:21,  4.08it/s]

train:  76%|███████▋  | 2688/3515 [13:19<03:22,  4.08it/s]
overall:  67%|██████▋   | 2689/4000 [13:22<05:32,  3.94it/s]

train:  77%|███████▋  | 2689/3515 [13:20<03:29,  3.94it/s]
overall:  67%|██████▋   | 2690/4000 [13:22<05:35,  3.90it/s]

train:  77%|███████▋  | 2690/3515 [13:20<03:31,  3.90it/s]
overall:  67%|██████▋   | 2691/4000 [13:23<05:34,  3.91it/s]

train:  77%|███████▋  | 2691/3515 [13:20<03:30,  3.91it/s]
overall:  67%|██████▋   | 2692/4000 [13:23<05:40,  3.84it/s]

train:  77%|███████▋  | 2692/3515 [13:20<03:34,  3.84it/s]
overall:  67%|██████▋   | 2693/4000 [13:23<05:31,  3.95it/s]

train:  77%|███████▋  | 2693/3515 [13:21<03:28,  3.95it/s]
overall:  67%|██████▋   | 2694/4000 [13:24<05:39,  3.85it/s]

train:  77%|███████▋  | 2694/3515 [13:21<03:33,  3.85it/s]
overall:  67%|██████▋   | 2695/4000 [13:24<05:42,  3.81it/s]

train:  77%|███████▋  | 2695/3515 [13:21<03:35,  3.81it/s]
overall:  67%|██████▋   | 2696/4000 [13:24<05:46,  3.76it/s]

train:  77%|███████▋  | 2696/3515 [13:21<03:37,  3.76it/s]
overall:  67%|██████▋   | 2697/4000 [13:25<06:54,  3.14it/s]

train:  77%|███████▋  | 2697/3515 [13:22<04:20,  3.14it/s]
overall:  67%|██████▋   | 2698/4000 [13:25<06:37,  3.27it/s]

train:  77%|███████▋  | 2698/3515 [13:22<04:09,  3.27it/s]
overall:  67%|██████▋   | 2699/4000 [13:25<06:23,  3.39it/s]

train:  77%|███████▋  | 2699/3515 [13:22<04:00,  3.39it/s]train loss: 0.5557062029838562
train per: 0.1623716153127918

overall:  68%|██████▊   | 2700/4000 [13:25<06:17,  3.45it/s]

train:  77%|███████▋  | 2700/3515 [13:23<03:56,  3.45it/s]
overall:  68%|██████▊   | 2701/4000 [13:26<06:36,  3.28it/s]

train:  77%|███████▋  | 2701/3515 [13:23<04:08,  3.28it/s]
overall:  68%|██████▊   | 2702/4000 [13:26<06:16,  3.45it/s]

train:  77%|███████▋  | 2702/3515 [13:23<03:55,  3.45it/s]
overall:  68%|██████▊   | 2703/4000 [13:26<05:55,  3.65it/s]

train:  77%|███████▋  | 2703/3515 [13:24<03:42,  3.65it/s]
overall:  68%|██████▊   | 2704/4000 [13:26<06:15,  3.45it/s]

train:  77%|███████▋  | 2704/3515 [13:24<03:54,  3.45it/s]
overall:  68%|██████▊   | 2705/4000 [13:27<06:08,  3.51it/s]

train:  77%|███████▋  | 2705/3515 [13:24<03:50,  3.51it/s]
overall:  68%|██████▊   | 2706/4000 [13:27<05:58,  3.61it/s]

train:  77%|███████▋  | 2706/3515 [13:24<03:43,  3.61it/s]
overall:  68%|██████▊   | 2707/4000 [13:27<05:56,  3.63it/s]

train:  77%|███████▋  | 2707/3515 [13:25<03:42,  3.63it/s]
overall:  68%|██████▊   | 2708/4000 [13:28<05:49,  3.70it/s]

train:  77%|███████▋  | 2708/3515 [13:25<03:38,  3.70it/s]
overall:  68%|██████▊   | 2709/4000 [13:28<05:24,  3.98it/s]

train:  77%|███████▋  | 2709/3515 [13:25<03:22,  3.98it/s]
overall:  68%|██████▊   | 2710/4000 [13:28<05:33,  3.87it/s]

train:  77%|███████▋  | 2710/3515 [13:25<03:28,  3.87it/s]
overall:  68%|██████▊   | 2711/4000 [13:28<05:29,  3.92it/s]

train:  77%|███████▋  | 2711/3515 [13:26<03:25,  3.92it/s]
overall:  68%|██████▊   | 2712/4000 [13:29<05:33,  3.87it/s]

train:  77%|███████▋  | 2712/3515 [13:26<03:27,  3.87it/s]
overall:  68%|██████▊   | 2713/4000 [13:29<05:41,  3.77it/s]

train:  77%|███████▋  | 2713/3515 [13:26<03:32,  3.77it/s]
overall:  68%|██████▊   | 2714/4000 [13:29<05:37,  3.81it/s]

train:  77%|███████▋  | 2714/3515 [13:26<03:30,  3.81it/s]
overall:  68%|██████▊   | 2715/4000 [13:29<05:30,  3.89it/s]

train:  77%|███████▋  | 2715/3515 [13:27<03:25,  3.89it/s]
overall:  68%|██████▊   | 2716/4000 [13:30<05:36,  3.82it/s]

train:  77%|███████▋  | 2716/3515 [13:27<03:29,  3.82it/s]
overall:  68%|██████▊   | 2717/4000 [13:30<05:40,  3.77it/s]

train:  77%|███████▋  | 2717/3515 [13:27<03:31,  3.77it/s]
overall:  68%|██████▊   | 2718/4000 [13:30<05:47,  3.69it/s]

train:  77%|███████▋  | 2718/3515 [13:28<03:35,  3.69it/s]
overall:  68%|██████▊   | 2719/4000 [13:30<05:48,  3.68it/s]

train:  77%|███████▋  | 2719/3515 [13:28<03:36,  3.68it/s]train loss: 0.5590537786483765
train per: 0.1635264923486972

overall:  68%|██████▊   | 2720/4000 [13:31<05:49,  3.66it/s]

train:  77%|███████▋  | 2720/3515 [13:28<03:37,  3.66it/s]
overall:  68%|██████▊   | 2721/4000 [13:31<06:44,  3.16it/s]

train:  77%|███████▋  | 2721/3515 [13:29<04:11,  3.16it/s]
overall:  68%|██████▊   | 2722/4000 [13:31<06:20,  3.36it/s]

train:  77%|███████▋  | 2722/3515 [13:29<03:55,  3.36it/s]
overall:  68%|██████▊   | 2723/4000 [13:32<06:10,  3.45it/s]

train:  77%|███████▋  | 2723/3515 [13:29<03:49,  3.45it/s]
overall:  68%|██████▊   | 2724/4000 [13:32<06:08,  3.47it/s]

train:  77%|███████▋  | 2724/3515 [13:29<03:48,  3.47it/s]
overall:  68%|██████▊   | 2725/4000 [13:32<06:20,  3.35it/s]

train:  78%|███████▊  | 2725/3515 [13:30<03:55,  3.35it/s]
overall:  68%|██████▊   | 2726/4000 [13:33<06:12,  3.42it/s]

train:  78%|███████▊  | 2726/3515 [13:30<03:50,  3.42it/s]
overall:  68%|██████▊   | 2727/4000 [13:33<05:57,  3.56it/s]

train:  78%|███████▊  | 2727/3515 [13:30<03:41,  3.56it/s]
overall:  68%|██████▊   | 2728/4000 [13:33<06:18,  3.36it/s]

train:  78%|███████▊  | 2728/3515 [13:31<03:54,  3.36it/s]
overall:  68%|██████▊   | 2730/4000 [13:33<05:03,  4.18it/s]

train:  78%|███████▊  | 2730/3515 [13:31<03:07,  4.18it/s]
overall:  68%|██████▊   | 2731/4000 [13:34<05:05,  4.15it/s]

train:  78%|███████▊  | 2731/3515 [13:31<03:08,  4.15it/s]
overall:  68%|██████▊   | 2732/4000 [13:34<05:11,  4.07it/s]

train:  78%|███████▊  | 2732/3515 [13:31<03:12,  4.07it/s]
overall:  68%|██████▊   | 2733/4000 [13:34<05:17,  3.99it/s]

train:  78%|███████▊  | 2733/3515 [13:32<03:16,  3.99it/s]
overall:  68%|██████▊   | 2734/4000 [13:34<04:43,  4.47it/s]

train:  78%|███████▊  | 2734/3515 [13:32<02:54,  4.47it/s]
overall:  68%|██████▊   | 2735/4000 [13:35<04:56,  4.26it/s]

train:  78%|███████▊  | 2735/3515 [13:32<03:03,  4.26it/s]
overall:  68%|██████▊   | 2736/4000 [13:35<05:06,  4.13it/s]

train:  78%|███████▊  | 2736/3515 [13:32<03:08,  4.13it/s]
overall:  68%|██████▊   | 2737/4000 [13:35<05:04,  4.15it/s]

train:  78%|███████▊  | 2737/3515 [13:33<03:07,  4.15it/s]
overall:  68%|██████▊   | 2738/4000 [13:35<05:10,  4.06it/s]

train:  78%|███████▊  | 2738/3515 [13:33<03:11,  4.06it/s]
overall:  68%|██████▊   | 2739/4000 [13:36<05:18,  3.96it/s]

train:  78%|███████▊  | 2739/3515 [13:33<03:16,  3.96it/s]train loss: 0.5282116532325745
train per: 0.15555882858335174

overall:  68%|██████▊   | 2740/4000 [13:36<05:28,  3.83it/s]

train:  78%|███████▊  | 2740/3515 [13:33<03:22,  3.83it/s]
overall:  69%|██████▊   | 2741/4000 [13:36<05:29,  3.82it/s]

train:  78%|███████▊  | 2741/3515 [13:34<03:22,  3.82it/s]
overall:  69%|██████▊   | 2742/4000 [13:37<05:34,  3.76it/s]

train:  78%|███████▊  | 2742/3515 [13:34<03:25,  3.76it/s]
overall:  69%|██████▊   | 2743/4000 [13:37<05:37,  3.72it/s]

train:  78%|███████▊  | 2743/3515 [13:34<03:27,  3.72it/s]
overall:  69%|██████▊   | 2745/4000 [13:37<05:30,  3.80it/s]

train:  78%|███████▊  | 2745/3515 [13:35<03:22,  3.80it/s]
overall:  69%|██████▊   | 2746/4000 [13:38<05:29,  3.80it/s]

train:  78%|███████▊  | 2746/3515 [13:35<03:22,  3.81it/s]
overall:  69%|██████▊   | 2747/4000 [13:38<05:30,  3.79it/s]

train:  78%|███████▊  | 2747/3515 [13:35<03:22,  3.79it/s]
overall:  69%|██████▊   | 2748/4000 [13:38<05:35,  3.73it/s]

train:  78%|███████▊  | 2748/3515 [13:35<03:25,  3.73it/s]
overall:  69%|██████▊   | 2749/4000 [13:38<05:40,  3.67it/s]

train:  78%|███████▊  | 2749/3515 [13:36<03:28,  3.67it/s]
overall:  69%|██████▉   | 2750/4000 [13:39<05:38,  3.70it/s]

train:  78%|███████▊  | 2750/3515 [13:36<03:26,  3.70it/s]
overall:  69%|██████▉   | 2751/4000 [13:39<05:32,  3.75it/s]

train:  78%|███████▊  | 2751/3515 [13:36<03:23,  3.75it/s]
overall:  69%|██████▉   | 2752/4000 [13:39<05:00,  4.16it/s]

train:  78%|███████▊  | 2752/3515 [13:36<03:03,  4.16it/s]
overall:  69%|██████▉   | 2753/4000 [13:39<05:08,  4.05it/s]

train:  78%|███████▊  | 2753/3515 [13:37<03:08,  4.05it/s]
overall:  69%|██████▉   | 2754/4000 [13:40<05:13,  3.97it/s]

train:  78%|███████▊  | 2754/3515 [13:37<03:11,  3.97it/s]
overall:  69%|██████▉   | 2755/4000 [13:40<05:22,  3.86it/s]

train:  78%|███████▊  | 2755/3515 [13:37<03:16,  3.86it/s]
overall:  69%|██████▉   | 2756/4000 [13:40<05:15,  3.94it/s]

train:  78%|███████▊  | 2756/3515 [13:37<03:12,  3.94it/s]
overall:  69%|██████▉   | 2757/4000 [13:40<05:26,  3.81it/s]

train:  78%|███████▊  | 2757/3515 [13:38<03:18,  3.81it/s]
overall:  69%|██████▉   | 2758/4000 [13:41<05:31,  3.75it/s]

train:  78%|███████▊  | 2758/3515 [13:38<03:21,  3.75it/s]
overall:  69%|██████▉   | 2759/4000 [13:41<04:40,  4.42it/s]

train:  78%|███████▊  | 2759/3515 [13:38<02:51,  4.42it/s]train loss: 0.5704960227012634
train per: 0.16579890957316173

overall:  69%|██████▉   | 2760/4000 [13:41<04:56,  4.19it/s]

train:  79%|███████▊  | 2760/3515 [13:38<03:00,  4.18it/s]
overall:  69%|██████▉   | 2761/4000 [13:41<05:05,  4.05it/s]

train:  79%|███████▊  | 2761/3515 [13:39<03:06,  4.05it/s]
overall:  69%|██████▉   | 2762/4000 [13:42<05:14,  3.94it/s]

train:  79%|███████▊  | 2762/3515 [13:39<03:11,  3.94it/s]
overall:  69%|██████▉   | 2763/4000 [13:42<05:09,  4.00it/s]

train:  79%|███████▊  | 2763/3515 [13:39<03:08,  4.00it/s]
overall:  69%|██████▉   | 2764/4000 [13:42<05:17,  3.89it/s]

train:  79%|███████▊  | 2764/3515 [13:40<03:13,  3.89it/s]
overall:  69%|██████▉   | 2765/4000 [13:42<04:36,  4.47it/s]

train:  79%|███████▊  | 2765/3515 [13:40<02:47,  4.47it/s]
overall:  69%|██████▉   | 2766/4000 [13:43<04:42,  4.37it/s]

train:  79%|███████▊  | 2766/3515 [13:40<02:51,  4.37it/s]
overall:  69%|██████▉   | 2767/4000 [13:43<04:57,  4.14it/s]

train:  79%|███████▊  | 2767/3515 [13:40<03:00,  4.14it/s]
overall:  69%|██████▉   | 2768/4000 [13:43<05:05,  4.03it/s]

train:  79%|███████▊  | 2768/3515 [13:40<03:05,  4.03it/s]
overall:  69%|██████▉   | 2769/4000 [13:44<06:20,  3.24it/s]

train:  79%|███████▉  | 2769/3515 [13:41<03:50,  3.24it/s]
overall:  69%|██████▉   | 2770/4000 [13:44<06:03,  3.38it/s]

train:  79%|███████▉  | 2770/3515 [13:41<03:40,  3.38it/s]
overall:  69%|██████▉   | 2771/4000 [13:44<05:55,  3.46it/s]

train:  79%|███████▉  | 2771/3515 [13:41<03:34,  3.46it/s]
overall:  69%|██████▉   | 2772/4000 [13:44<05:03,  4.05it/s]

train:  79%|███████▉  | 2772/3515 [13:42<03:03,  4.05it/s]
overall:  69%|██████▉   | 2773/4000 [13:44<05:24,  3.78it/s]

train:  79%|███████▉  | 2773/3515 [13:42<03:16,  3.78it/s]
overall:  69%|██████▉   | 2774/4000 [13:45<05:24,  3.78it/s]

train:  79%|███████▉  | 2774/3515 [13:42<03:16,  3.78it/s]
overall:  69%|██████▉   | 2775/4000 [13:45<05:23,  3.78it/s]

train:  79%|███████▉  | 2775/3515 [13:42<03:15,  3.78it/s]
overall:  69%|██████▉   | 2776/4000 [13:45<05:39,  3.61it/s]

train:  79%|███████▉  | 2776/3515 [13:43<03:24,  3.61it/s]
overall:  69%|██████▉   | 2777/4000 [13:46<05:32,  3.67it/s]

train:  79%|███████▉  | 2777/3515 [13:43<03:20,  3.67it/s]
overall:  69%|██████▉   | 2778/4000 [13:46<05:21,  3.80it/s]

train:  79%|███████▉  | 2778/3515 [13:43<03:13,  3.80it/s]
overall:  69%|██████▉   | 2779/4000 [13:46<05:25,  3.75it/s]

train:  79%|███████▉  | 2779/3515 [13:43<03:16,  3.75it/s]train loss: 0.5567286014556885
train per: 0.1588325948757465

overall:  70%|██████▉   | 2780/4000 [13:46<05:29,  3.71it/s]

train:  79%|███████▉  | 2780/3515 [13:44<03:18,  3.70it/s]
overall:  70%|██████▉   | 2781/4000 [13:47<05:17,  3.84it/s]

train:  79%|███████▉  | 2781/3515 [13:44<03:11,  3.84it/s]
overall:  70%|██████▉   | 2782/4000 [13:47<05:19,  3.82it/s]

train:  79%|███████▉  | 2782/3515 [13:44<03:12,  3.82it/s]
overall:  70%|██████▉   | 2783/4000 [13:47<05:22,  3.77it/s]

train:  79%|███████▉  | 2783/3515 [13:45<03:14,  3.77it/s]
overall:  70%|██████▉   | 2785/4000 [13:48<04:30,  4.50it/s]

train:  79%|███████▉  | 2785/3515 [13:45<02:42,  4.50it/s]
overall:  70%|██████▉   | 2786/4000 [13:48<04:42,  4.30it/s]

train:  79%|███████▉  | 2786/3515 [13:45<02:49,  4.30it/s]
overall:  70%|██████▉   | 2787/4000 [13:48<04:51,  4.17it/s]

train:  79%|███████▉  | 2787/3515 [13:45<02:54,  4.17it/s]
overall:  70%|██████▉   | 2788/4000 [13:48<05:02,  4.01it/s]

train:  79%|███████▉  | 2788/3515 [13:46<03:01,  4.01it/s]
overall:  70%|██████▉   | 2789/4000 [13:49<05:10,  3.90it/s]

train:  79%|███████▉  | 2789/3515 [13:46<03:06,  3.90it/s]
overall:  70%|██████▉   | 2790/4000 [13:49<05:18,  3.80it/s]

train:  79%|███████▉  | 2790/3515 [13:46<03:10,  3.80it/s]
overall:  70%|██████▉   | 2792/4000 [13:49<04:22,  4.61it/s]

train:  79%|███████▉  | 2792/3515 [13:47<02:36,  4.61it/s]
overall:  70%|██████▉   | 2793/4000 [13:50<05:27,  3.69it/s]

train:  79%|███████▉  | 2793/3515 [13:47<03:15,  3.69it/s]
overall:  70%|██████▉   | 2794/4000 [13:50<05:21,  3.75it/s]

train:  79%|███████▉  | 2794/3515 [13:47<03:12,  3.75it/s]
overall:  70%|██████▉   | 2796/4000 [13:50<04:25,  4.53it/s]

train:  80%|███████▉  | 2796/3515 [13:48<02:38,  4.54it/s]
overall:  70%|██████▉   | 2797/4000 [13:51<04:50,  4.14it/s]

train:  80%|███████▉  | 2797/3515 [13:48<02:53,  4.14it/s]
overall:  70%|██████▉   | 2798/4000 [13:51<05:03,  3.96it/s]

train:  80%|███████▉  | 2798/3515 [13:48<03:00,  3.96it/s]
overall:  70%|██████▉   | 2799/4000 [13:51<05:07,  3.91it/s]

train:  80%|███████▉  | 2799/3515 [13:48<03:03,  3.91it/s]train loss: 0.5799635648727417
train per: 0.16940049354042677

overall:  70%|███████   | 2800/4000 [13:51<05:30,  3.63it/s]

train:  80%|███████▉  | 2800/3515 [13:49<03:16,  3.63it/s]
overall:  70%|███████   | 2801/4000 [13:52<05:22,  3.71it/s]

train:  80%|███████▉  | 2801/3515 [13:49<03:12,  3.71it/s]
overall:  70%|███████   | 2802/4000 [13:52<05:16,  3.78it/s]

train:  80%|███████▉  | 2802/3515 [13:49<03:08,  3.78it/s]
overall:  70%|███████   | 2804/4000 [13:52<04:24,  4.52it/s]

train:  80%|███████▉  | 2804/3515 [13:50<02:37,  4.52it/s]
overall:  70%|███████   | 2805/4000 [13:52<04:23,  4.54it/s]

train:  80%|███████▉  | 2805/3515 [13:50<02:36,  4.54it/s]
overall:  70%|███████   | 2806/4000 [13:53<04:33,  4.37it/s]

train:  80%|███████▉  | 2806/3515 [13:50<02:42,  4.37it/s]
overall:  70%|███████   | 2807/4000 [13:53<04:46,  4.16it/s]

train:  80%|███████▉  | 2807/3515 [13:50<02:50,  4.16it/s]
overall:  70%|███████   | 2808/4000 [13:53<04:53,  4.06it/s]

train:  80%|███████▉  | 2808/3515 [13:51<02:53,  4.06it/s]
overall:  70%|███████   | 2809/4000 [13:54<05:01,  3.95it/s]

train:  80%|███████▉  | 2809/3515 [13:51<02:58,  3.95it/s]
overall:  70%|███████   | 2810/4000 [13:54<04:59,  3.97it/s]

train:  80%|███████▉  | 2810/3515 [13:51<02:57,  3.97it/s]
overall:  70%|███████   | 2811/4000 [13:54<05:06,  3.88it/s]

train:  80%|███████▉  | 2811/3515 [13:51<03:01,  3.88it/s]
overall:  70%|███████   | 2812/4000 [13:54<04:16,  4.64it/s]

train:  80%|████████  | 2812/3515 [13:52<02:31,  4.64it/s]
overall:  70%|███████   | 2813/4000 [13:54<04:34,  4.32it/s]

train:  80%|████████  | 2813/3515 [13:52<02:42,  4.32it/s]
overall:  70%|███████   | 2814/4000 [13:55<04:46,  4.14it/s]

train:  80%|████████  | 2814/3515 [13:52<02:49,  4.14it/s]
overall:  70%|███████   | 2815/4000 [13:55<04:57,  3.98it/s]

train:  80%|████████  | 2815/3515 [13:52<02:55,  3.98it/s]
overall:  70%|███████   | 2816/4000 [13:55<04:54,  4.02it/s]

train:  80%|████████  | 2816/3515 [13:53<02:54,  4.02it/s]
overall:  70%|███████   | 2817/4000 [13:56<05:51,  3.36it/s]

train:  80%|████████  | 2817/3515 [13:53<03:27,  3.36it/s]
overall:  70%|███████   | 2818/4000 [13:56<05:37,  3.50it/s]

train:  80%|████████  | 2818/3515 [13:53<03:19,  3.50it/s]
overall:  70%|███████   | 2819/4000 [13:56<05:29,  3.58it/s]

train:  80%|████████  | 2819/3515 [13:53<03:14,  3.58it/s]train loss: 0.5712884664535522
train per: 0.16702955677353154

overall:  70%|███████   | 2820/4000 [13:56<05:34,  3.52it/s]

train:  80%|████████  | 2820/3515 [13:54<03:17,  3.52it/s]
overall:  71%|███████   | 2821/4000 [13:57<05:38,  3.48it/s]

train:  80%|████████  | 2821/3515 [13:54<03:19,  3.48it/s]
overall:  71%|███████   | 2822/4000 [13:57<05:35,  3.51it/s]

train:  80%|████████  | 2822/3515 [13:54<03:17,  3.52it/s]
overall:  71%|███████   | 2823/4000 [13:57<05:30,  3.56it/s]

train:  80%|████████  | 2823/3515 [13:55<03:14,  3.56it/s]
overall:  71%|███████   | 2824/4000 [13:57<04:40,  4.20it/s]

train:  80%|████████  | 2824/3515 [13:55<02:44,  4.20it/s]
overall:  71%|███████   | 2825/4000 [13:58<04:53,  4.00it/s]

train:  80%|████████  | 2825/3515 [13:55<02:52,  4.00it/s]
overall:  71%|███████   | 2827/4000 [13:58<04:03,  4.81it/s]

train:  80%|████████  | 2827/3515 [13:55<02:22,  4.81it/s]
overall:  71%|███████   | 2828/4000 [13:58<04:27,  4.37it/s]

train:  80%|████████  | 2828/3515 [13:56<02:37,  4.37it/s]
overall:  71%|███████   | 2829/4000 [13:59<04:39,  4.19it/s]

train:  80%|████████  | 2829/3515 [13:56<02:43,  4.19it/s]
overall:  71%|███████   | 2830/4000 [13:59<04:50,  4.03it/s]

train:  81%|████████  | 2830/3515 [13:56<02:50,  4.03it/s]
overall:  71%|███████   | 2831/4000 [13:59<04:08,  4.70it/s]

train:  81%|████████  | 2831/3515 [13:56<02:25,  4.70it/s]
overall:  71%|███████   | 2832/4000 [13:59<04:24,  4.41it/s]

train:  81%|████████  | 2832/3515 [13:57<02:34,  4.41it/s]
overall:  71%|███████   | 2833/4000 [13:59<04:37,  4.20it/s]

train:  81%|████████  | 2833/3515 [13:57<02:42,  4.20it/s]
overall:  71%|███████   | 2834/4000 [14:00<04:43,  4.11it/s]

train:  81%|████████  | 2834/3515 [13:57<02:45,  4.11it/s]
overall:  71%|███████   | 2835/4000 [14:00<03:54,  4.97it/s]

train:  81%|████████  | 2835/3515 [13:57<02:16,  4.97it/s]
overall:  71%|███████   | 2837/4000 [14:00<03:27,  5.60it/s]

train:  81%|████████  | 2837/3515 [13:58<02:01,  5.60it/s]
overall:  71%|███████   | 2838/4000 [14:00<03:54,  4.97it/s]

train:  81%|████████  | 2838/3515 [13:58<02:16,  4.97it/s]
overall:  71%|███████   | 2839/4000 [14:01<04:14,  4.56it/s]

train:  81%|████████  | 2839/3515 [13:58<02:28,  4.56it/s]train loss: 0.5767480134963989
train per: 0.17406675678486014

overall:  71%|███████   | 2840/4000 [14:01<04:28,  4.31it/s]

train:  81%|████████  | 2840/3515 [13:58<02:36,  4.31it/s]
overall:  71%|███████   | 2841/4000 [14:01<05:35,  3.45it/s]

train:  81%|████████  | 2841/3515 [13:59<03:15,  3.45it/s]
overall:  71%|███████   | 2842/4000 [14:02<05:26,  3.55it/s]

train:  81%|████████  | 2842/3515 [13:59<03:09,  3.55it/s]
overall:  71%|███████   | 2843/4000 [14:02<05:24,  3.57it/s]

train:  81%|████████  | 2843/3515 [13:59<03:08,  3.57it/s]
overall:  71%|███████   | 2844/4000 [14:02<05:26,  3.55it/s]

train:  81%|████████  | 2844/3515 [14:00<03:09,  3.55it/s]
overall:  71%|███████   | 2845/4000 [14:03<05:31,  3.49it/s]

train:  81%|████████  | 2845/3515 [14:00<03:12,  3.49it/s]
overall:  71%|███████   | 2846/4000 [14:03<05:33,  3.46it/s]

train:  81%|████████  | 2846/3515 [14:00<03:13,  3.46it/s]
overall:  71%|███████   | 2847/4000 [14:03<05:23,  3.57it/s]

train:  81%|████████  | 2847/3515 [14:00<03:07,  3.57it/s]
overall:  71%|███████   | 2848/4000 [14:03<05:18,  3.61it/s]

train:  81%|████████  | 2848/3515 [14:01<03:04,  3.61it/s]
overall:  71%|███████▏  | 2850/4000 [14:04<04:19,  4.43it/s]

train:  81%|████████  | 2850/3515 [14:01<02:30,  4.43it/s]
overall:  71%|███████▏  | 2851/4000 [14:04<04:31,  4.24it/s]

train:  81%|████████  | 2851/3515 [14:01<02:36,  4.24it/s]
overall:  71%|███████▏  | 2852/4000 [14:04<04:40,  4.10it/s]

train:  81%|████████  | 2852/3515 [14:02<02:41,  4.10it/s]
overall:  71%|███████▏  | 2853/4000 [14:04<04:48,  3.98it/s]

train:  81%|████████  | 2853/3515 [14:02<02:46,  3.98it/s]
overall:  71%|███████▏  | 2854/4000 [14:05<04:49,  3.96it/s]

train:  81%|████████  | 2854/3515 [14:02<02:46,  3.96it/s]
overall:  71%|███████▏  | 2855/4000 [14:05<04:58,  3.84it/s]

train:  81%|████████  | 2855/3515 [14:02<02:51,  3.84it/s]
overall:  71%|███████▏  | 2856/4000 [14:05<05:01,  3.79it/s]

train:  81%|████████▏ | 2856/3515 [14:03<02:53,  3.79it/s]
overall:  71%|███████▏  | 2857/4000 [14:06<05:02,  3.78it/s]

train:  81%|████████▏ | 2857/3515 [14:03<02:54,  3.78it/s]
overall:  71%|███████▏  | 2859/4000 [14:06<04:02,  4.71it/s]

train:  81%|████████▏ | 2859/3515 [14:03<02:19,  4.71it/s]train loss: 0.5403251647949219
train per: 0.16264350897851046

overall:  72%|███████▏  | 2860/4000 [14:06<04:22,  4.34it/s]

train:  81%|████████▏ | 2860/3515 [14:03<02:31,  4.33it/s]
overall:  72%|███████▏  | 2861/4000 [14:06<04:29,  4.22it/s]

train:  81%|████████▏ | 2861/3515 [14:04<02:34,  4.23it/s]
overall:  72%|███████▏  | 2862/4000 [14:07<04:38,  4.09it/s]

train:  81%|████████▏ | 2862/3515 [14:04<02:39,  4.09it/s]
overall:  72%|███████▏  | 2863/4000 [14:07<04:47,  3.95it/s]

train:  81%|████████▏ | 2863/3515 [14:04<02:44,  3.95it/s]
overall:  72%|███████▏  | 2864/4000 [14:07<04:38,  4.08it/s]

train:  81%|████████▏ | 2864/3515 [14:05<02:39,  4.08it/s]
overall:  72%|███████▏  | 2865/4000 [14:08<05:42,  3.32it/s]

train:  82%|████████▏ | 2865/3515 [14:05<03:16,  3.32it/s]
overall:  72%|███████▏  | 2866/4000 [14:08<04:58,  3.79it/s]

train:  82%|████████▏ | 2866/3515 [14:05<02:51,  3.79it/s]
overall:  72%|███████▏  | 2867/4000 [14:08<05:00,  3.78it/s]

train:  82%|████████▏ | 2867/3515 [14:05<02:51,  3.78it/s]
overall:  72%|███████▏  | 2868/4000 [14:08<05:06,  3.69it/s]

train:  82%|████████▏ | 2868/3515 [14:06<02:55,  3.69it/s]
overall:  72%|███████▏  | 2870/4000 [14:08<03:26,  5.47it/s]

train:  82%|████████▏ | 2870/3515 [14:06<01:57,  5.47it/s]
overall:  72%|███████▏  | 2871/4000 [14:09<03:47,  4.96it/s]

train:  82%|████████▏ | 2871/3515 [14:06<02:09,  4.96it/s]
overall:  72%|███████▏  | 2872/4000 [14:09<04:14,  4.43it/s]

train:  82%|████████▏ | 2872/3515 [14:06<02:25,  4.43it/s]
overall:  72%|███████▏  | 2873/4000 [14:09<04:29,  4.18it/s]

train:  82%|████████▏ | 2873/3515 [14:07<02:33,  4.18it/s]
overall:  72%|███████▏  | 2874/4000 [14:10<04:33,  4.11it/s]

train:  82%|████████▏ | 2874/3515 [14:07<02:35,  4.11it/s]
overall:  72%|███████▏  | 2875/4000 [14:10<04:43,  3.97it/s]

train:  82%|████████▏ | 2875/3515 [14:07<02:41,  3.97it/s]
overall:  72%|███████▏  | 2876/4000 [14:10<04:52,  3.84it/s]

train:  82%|████████▏ | 2876/3515 [14:07<02:46,  3.84it/s]
overall:  72%|███████▏  | 2877/4000 [14:10<04:57,  3.77it/s]

train:  82%|████████▏ | 2877/3515 [14:08<02:49,  3.77it/s]
overall:  72%|███████▏  | 2878/4000 [14:11<04:57,  3.78it/s]

train:  82%|████████▏ | 2878/3515 [14:08<02:48,  3.78it/s]
overall:  72%|███████▏  | 2879/4000 [14:11<04:12,  4.44it/s]

train:  82%|████████▏ | 2879/3515 [14:08<02:23,  4.44it/s]train loss: 0.5370880961418152
train per: 0.1646411316431738

overall:  72%|███████▏  | 2880/4000 [14:11<04:30,  4.14it/s]

train:  82%|████████▏ | 2880/3515 [14:08<02:33,  4.13it/s]
overall:  72%|███████▏  | 2881/4000 [14:11<04:40,  3.98it/s]

train:  82%|████████▏ | 2881/3515 [14:09<02:39,  3.98it/s]
overall:  72%|███████▏  | 2882/4000 [14:12<04:46,  3.90it/s]

train:  82%|████████▏ | 2882/3515 [14:09<02:42,  3.90it/s]
overall:  72%|███████▏  | 2883/4000 [14:12<04:51,  3.83it/s]

train:  82%|████████▏ | 2883/3515 [14:09<02:44,  3.83it/s]
overall:  72%|███████▏  | 2884/4000 [14:12<04:38,  4.01it/s]

train:  82%|████████▏ | 2884/3515 [14:09<02:37,  4.01it/s]
overall:  72%|███████▏  | 2885/4000 [14:12<04:37,  4.01it/s]

train:  82%|████████▏ | 2885/3515 [14:10<02:36,  4.01it/s]
overall:  72%|███████▏  | 2886/4000 [14:13<04:39,  3.99it/s]

train:  82%|████████▏ | 2886/3515 [14:10<02:37,  3.99it/s]
overall:  72%|███████▏  | 2887/4000 [14:13<04:34,  4.05it/s]

train:  82%|████████▏ | 2887/3515 [14:10<02:35,  4.05it/s]
overall:  72%|███████▏  | 2888/4000 [14:13<04:40,  3.96it/s]

train:  82%|████████▏ | 2888/3515 [14:10<02:38,  3.96it/s]
overall:  72%|███████▏  | 2889/4000 [14:13<05:29,  3.37it/s]

train:  82%|████████▏ | 2889/3515 [14:11<03:05,  3.37it/s]
overall:  72%|███████▏  | 2890/4000 [14:14<04:25,  4.18it/s]

train:  82%|████████▏ | 2890/3515 [14:11<02:29,  4.18it/s]
overall:  72%|███████▏  | 2891/4000 [14:14<04:26,  4.16it/s]

train:  82%|████████▏ | 2891/3515 [14:11<02:29,  4.16it/s]
overall:  72%|███████▏  | 2892/4000 [14:14<04:34,  4.04it/s]

train:  82%|████████▏ | 2892/3515 [14:11<02:34,  4.04it/s]
overall:  72%|███████▏  | 2893/4000 [14:14<04:17,  4.30it/s]

train:  82%|████████▏ | 2893/3515 [14:12<02:24,  4.30it/s]
overall:  72%|███████▏  | 2894/4000 [14:15<04:32,  4.05it/s]

train:  82%|████████▏ | 2894/3515 [14:12<02:33,  4.05it/s]
overall:  72%|███████▏  | 2895/4000 [14:15<04:36,  4.00it/s]

train:  82%|████████▏ | 2895/3515 [14:12<02:35,  4.00it/s]
overall:  72%|███████▏  | 2896/4000 [14:15<04:53,  3.76it/s]

train:  82%|████████▏ | 2896/3515 [14:13<02:44,  3.76it/s]
overall:  72%|███████▏  | 2897/4000 [14:15<04:49,  3.81it/s]

train:  82%|████████▏ | 2897/3515 [14:13<02:42,  3.81it/s]
overall:  72%|███████▏  | 2898/4000 [14:16<04:12,  4.37it/s]

train:  82%|████████▏ | 2898/3515 [14:13<02:21,  4.37it/s]
overall:  72%|███████▏  | 2899/4000 [14:16<04:24,  4.16it/s]

train:  82%|████████▏ | 2899/3515 [14:13<02:28,  4.16it/s]train loss: 0.5994340181350708
train per: 0.17088049008410341

overall:  72%|███████▎  | 2900/4000 [14:16<04:26,  4.13it/s]

train:  83%|████████▎ | 2900/3515 [14:13<02:29,  4.12it/s]
overall:  73%|███████▎  | 2901/4000 [14:16<04:37,  3.96it/s]

train:  83%|████████▎ | 2901/3515 [14:14<02:34,  3.96it/s]
overall:  73%|███████▎  | 2902/4000 [14:17<04:45,  3.84it/s]

train:  83%|████████▎ | 2902/3515 [14:14<02:39,  3.84it/s]
overall:  73%|███████▎  | 2903/4000 [14:17<04:43,  3.86it/s]

train:  83%|████████▎ | 2903/3515 [14:14<02:38,  3.87it/s]
overall:  73%|███████▎  | 2904/4000 [14:17<04:26,  4.12it/s]

train:  83%|████████▎ | 2904/3515 [14:14<02:28,  4.12it/s]
overall:  73%|███████▎  | 2905/4000 [14:17<04:34,  4.00it/s]

train:  83%|████████▎ | 2905/3515 [14:15<02:32,  4.00it/s]
overall:  73%|███████▎  | 2906/4000 [14:18<04:10,  4.38it/s]

train:  83%|████████▎ | 2906/3515 [14:15<02:19,  4.38it/s]
overall:  73%|███████▎  | 2907/4000 [14:18<04:19,  4.21it/s]

train:  83%|████████▎ | 2907/3515 [14:15<02:24,  4.21it/s]
overall:  73%|███████▎  | 2908/4000 [14:18<04:29,  4.05it/s]

train:  83%|████████▎ | 2908/3515 [14:15<02:29,  4.05it/s]
overall:  73%|███████▎  | 2909/4000 [14:18<04:28,  4.07it/s]

train:  83%|████████▎ | 2909/3515 [14:16<02:28,  4.07it/s]
overall:  73%|███████▎  | 2910/4000 [14:19<04:26,  4.09it/s]

train:  83%|████████▎ | 2910/3515 [14:16<02:27,  4.09it/s]
overall:  73%|███████▎  | 2911/4000 [14:19<04:34,  3.97it/s]

train:  83%|████████▎ | 2911/3515 [14:16<02:32,  3.97it/s]
overall:  73%|███████▎  | 2912/4000 [14:19<03:48,  4.76it/s]

train:  83%|████████▎ | 2912/3515 [14:16<02:06,  4.76it/s]
overall:  73%|███████▎  | 2913/4000 [14:19<05:02,  3.59it/s]

train:  83%|████████▎ | 2913/3515 [14:17<02:47,  3.59it/s]
overall:  73%|███████▎  | 2914/4000 [14:20<04:58,  3.64it/s]

train:  83%|████████▎ | 2914/3515 [14:17<02:45,  3.64it/s]
overall:  73%|███████▎  | 2915/4000 [14:20<04:53,  3.69it/s]

train:  83%|████████▎ | 2915/3515 [14:17<02:42,  3.69it/s]
overall:  73%|███████▎  | 2916/4000 [14:20<04:55,  3.67it/s]

train:  83%|████████▎ | 2916/3515 [14:18<02:43,  3.67it/s]
overall:  73%|███████▎  | 2917/4000 [14:20<04:58,  3.63it/s]

train:  83%|████████▎ | 2917/3515 [14:18<02:44,  3.63it/s]
overall:  73%|███████▎  | 2918/4000 [14:21<04:55,  3.67it/s]

train:  83%|████████▎ | 2918/3515 [14:18<02:42,  3.67it/s]
overall:  73%|███████▎  | 2919/4000 [14:21<04:48,  3.74it/s]

train:  83%|████████▎ | 2919/3515 [14:18<02:39,  3.74it/s]train loss: 0.5618847608566284
train per: 0.16060485475527259

overall:  73%|███████▎  | 2920/4000 [14:21<05:00,  3.60it/s]

train:  83%|████████▎ | 2920/3515 [14:19<02:45,  3.59it/s]
overall:  73%|███████▎  | 2921/4000 [14:22<04:58,  3.61it/s]

train:  83%|████████▎ | 2921/3515 [14:19<02:44,  3.61it/s]
overall:  73%|███████▎  | 2922/4000 [14:22<04:55,  3.65it/s]

train:  83%|████████▎ | 2922/3515 [14:19<02:42,  3.65it/s]
overall:  73%|███████▎  | 2923/4000 [14:22<04:46,  3.76it/s]

train:  83%|████████▎ | 2923/3515 [14:19<02:37,  3.76it/s]
overall:  73%|███████▎  | 2924/4000 [14:22<04:54,  3.66it/s]

train:  83%|████████▎ | 2924/3515 [14:20<02:41,  3.66it/s]
overall:  73%|███████▎  | 2925/4000 [14:23<04:52,  3.68it/s]

train:  83%|████████▎ | 2925/3515 [14:20<02:40,  3.68it/s]
overall:  73%|███████▎  | 2926/4000 [14:23<04:50,  3.70it/s]

train:  83%|████████▎ | 2926/3515 [14:20<02:39,  3.70it/s]
overall:  73%|███████▎  | 2927/4000 [14:23<04:48,  3.72it/s]

train:  83%|████████▎ | 2927/3515 [14:21<02:38,  3.72it/s]
overall:  73%|███████▎  | 2928/4000 [14:23<04:47,  3.73it/s]

train:  83%|████████▎ | 2928/3515 [14:21<02:37,  3.73it/s]
overall:  73%|███████▎  | 2929/4000 [14:24<04:48,  3.71it/s]

train:  83%|████████▎ | 2929/3515 [14:21<02:37,  3.71it/s]
overall:  73%|███████▎  | 2930/4000 [14:24<04:49,  3.70it/s]

train:  83%|████████▎ | 2930/3515 [14:21<02:38,  3.70it/s]
overall:  73%|███████▎  | 2931/4000 [14:24<04:49,  3.70it/s]

train:  83%|████████▎ | 2931/3515 [14:22<02:38,  3.70it/s]
overall:  73%|███████▎  | 2932/4000 [14:24<04:04,  4.36it/s]

train:  83%|████████▎ | 2932/3515 [14:22<02:13,  4.36it/s]
overall:  73%|███████▎  | 2933/4000 [14:25<04:16,  4.15it/s]

train:  83%|████████▎ | 2933/3515 [14:22<02:20,  4.15it/s]
overall:  73%|███████▎  | 2934/4000 [14:25<04:26,  4.01it/s]

train:  83%|████████▎ | 2934/3515 [14:22<02:25,  4.01it/s]
overall:  73%|███████▎  | 2935/4000 [14:25<04:35,  3.87it/s]

train:  83%|████████▎ | 2935/3515 [14:23<02:29,  3.87it/s]
overall:  73%|███████▎  | 2936/4000 [14:25<04:30,  3.93it/s]

train:  84%|████████▎ | 2936/3515 [14:23<02:27,  3.93it/s]
overall:  73%|███████▎  | 2937/4000 [14:26<04:18,  4.11it/s]

train:  84%|████████▎ | 2937/3515 [14:23<02:20,  4.11it/s]
overall:  73%|███████▎  | 2938/4000 [14:26<04:17,  4.12it/s]

train:  84%|████████▎ | 2938/3515 [14:23<02:19,  4.12it/s]
overall:  73%|███████▎  | 2939/4000 [14:26<04:29,  3.94it/s]

train:  84%|████████▎ | 2939/3515 [14:24<02:26,  3.94it/s]train loss: 0.5552114248275757
train per: 0.16262984295771182

overall:  74%|███████▎  | 2940/4000 [14:26<04:20,  4.07it/s]

train:  84%|████████▎ | 2940/3515 [14:24<02:21,  4.07it/s]
overall:  74%|███████▎  | 2941/4000 [14:27<04:44,  3.72it/s]

train:  84%|████████▎ | 2941/3515 [14:24<02:34,  3.72it/s]
overall:  74%|███████▎  | 2942/4000 [14:27<04:44,  3.71it/s]

train:  84%|████████▎ | 2942/3515 [14:24<02:34,  3.71it/s]
overall:  74%|███████▎  | 2943/4000 [14:27<04:46,  3.69it/s]

train:  84%|████████▎ | 2943/3515 [14:25<02:34,  3.69it/s]
overall:  74%|███████▎  | 2944/4000 [14:28<04:58,  3.54it/s]

train:  84%|████████▍ | 2944/3515 [14:25<02:41,  3.54it/s]
overall:  74%|███████▎  | 2945/4000 [14:28<04:51,  3.62it/s]

train:  84%|████████▍ | 2945/3515 [14:25<02:37,  3.62it/s]
overall:  74%|███████▎  | 2946/4000 [14:28<03:55,  4.47it/s]

train:  84%|████████▍ | 2946/3515 [14:25<02:07,  4.47it/s]
overall:  74%|███████▎  | 2947/4000 [14:28<04:18,  4.07it/s]

train:  84%|████████▍ | 2947/3515 [14:26<02:19,  4.07it/s]
overall:  74%|███████▎  | 2948/4000 [14:28<04:23,  3.99it/s]

train:  84%|████████▍ | 2948/3515 [14:26<02:22,  3.99it/s]
overall:  74%|███████▎  | 2949/4000 [14:29<04:18,  4.07it/s]

train:  84%|████████▍ | 2949/3515 [14:26<02:18,  4.07it/s]
overall:  74%|███████▍  | 2950/4000 [14:29<04:20,  4.03it/s]

train:  84%|████████▍ | 2950/3515 [14:26<02:20,  4.03it/s]
overall:  74%|███████▍  | 2951/4000 [14:29<04:21,  4.01it/s]

train:  84%|████████▍ | 2951/3515 [14:27<02:20,  4.01it/s]
overall:  74%|███████▍  | 2952/4000 [14:29<04:06,  4.26it/s]

train:  84%|████████▍ | 2952/3515 [14:27<02:12,  4.26it/s]
overall:  74%|███████▍  | 2954/4000 [14:30<03:09,  5.51it/s]

train:  84%|████████▍ | 2954/3515 [14:27<01:41,  5.51it/s]
overall:  74%|███████▍  | 2955/4000 [14:30<03:32,  4.91it/s]

train:  84%|████████▍ | 2955/3515 [14:27<01:54,  4.91it/s]
overall:  74%|███████▍  | 2956/4000 [14:30<03:50,  4.53it/s]

train:  84%|████████▍ | 2956/3515 [14:28<02:03,  4.53it/s]
overall:  74%|███████▍  | 2957/4000 [14:30<04:04,  4.26it/s]

train:  84%|████████▍ | 2957/3515 [14:28<02:11,  4.26it/s]
overall:  74%|███████▍  | 2958/4000 [14:31<04:03,  4.28it/s]

train:  84%|████████▍ | 2958/3515 [14:28<02:10,  4.28it/s]
overall:  74%|███████▍  | 2959/4000 [14:31<04:12,  4.13it/s]

train:  84%|████████▍ | 2959/3515 [14:28<02:14,  4.13it/s]train loss: 0.556843638420105
train per: 0.16360571601086848

overall:  74%|███████▍  | 2960/4000 [14:31<04:26,  3.90it/s]

train:  84%|████████▍ | 2960/3515 [14:29<02:22,  3.90it/s]
overall:  74%|███████▍  | 2961/4000 [14:32<05:19,  3.25it/s]

train:  84%|████████▍ | 2961/3515 [14:29<02:50,  3.25it/s]
overall:  74%|███████▍  | 2962/4000 [14:32<04:59,  3.47it/s]

train:  84%|████████▍ | 2962/3515 [14:29<02:39,  3.47it/s]
overall:  74%|███████▍  | 2963/4000 [14:32<04:44,  3.65it/s]

train:  84%|████████▍ | 2963/3515 [14:30<02:31,  3.65it/s]
overall:  74%|███████▍  | 2964/4000 [14:32<04:42,  3.67it/s]

train:  84%|████████▍ | 2964/3515 [14:30<02:30,  3.67it/s]
overall:  74%|███████▍  | 2965/4000 [14:33<04:53,  3.53it/s]

train:  84%|████████▍ | 2965/3515 [14:30<02:35,  3.53it/s]
overall:  74%|███████▍  | 2966/4000 [14:33<04:12,  4.09it/s]

train:  84%|████████▍ | 2966/3515 [14:30<02:14,  4.09it/s]
overall:  74%|███████▍  | 2967/4000 [14:33<04:22,  3.93it/s]

train:  84%|████████▍ | 2967/3515 [14:31<02:19,  3.93it/s]
overall:  74%|███████▍  | 2968/4000 [14:33<04:26,  3.87it/s]

train:  84%|████████▍ | 2968/3515 [14:31<02:21,  3.87it/s]
overall:  74%|███████▍  | 2969/4000 [14:34<04:31,  3.80it/s]

train:  84%|████████▍ | 2969/3515 [14:31<02:23,  3.80it/s]
overall:  74%|███████▍  | 2970/4000 [14:34<04:28,  3.83it/s]

train:  84%|████████▍ | 2970/3515 [14:31<02:22,  3.83it/s]
overall:  74%|███████▍  | 2971/4000 [14:34<04:24,  3.89it/s]

train:  85%|████████▍ | 2971/3515 [14:32<02:19,  3.89it/s]
overall:  74%|███████▍  | 2972/4000 [14:35<04:42,  3.64it/s]

train:  85%|████████▍ | 2972/3515 [14:32<02:29,  3.64it/s]
overall:  74%|███████▍  | 2973/4000 [14:35<04:44,  3.61it/s]

train:  85%|████████▍ | 2973/3515 [14:32<02:30,  3.61it/s]
overall:  74%|███████▍  | 2974/4000 [14:35<04:39,  3.67it/s]

train:  85%|████████▍ | 2974/3515 [14:32<02:27,  3.67it/s]
overall:  74%|███████▍  | 2975/4000 [14:35<04:37,  3.70it/s]

train:  85%|████████▍ | 2975/3515 [14:33<02:25,  3.70it/s]
overall:  74%|███████▍  | 2976/4000 [14:36<04:35,  3.72it/s]

train:  85%|████████▍ | 2976/3515 [14:33<02:24,  3.72it/s]
overall:  74%|███████▍  | 2977/4000 [14:36<04:30,  3.78it/s]

train:  85%|████████▍ | 2977/3515 [14:33<02:22,  3.79it/s]
overall:  74%|███████▍  | 2978/4000 [14:36<04:30,  3.78it/s]

train:  85%|████████▍ | 2978/3515 [14:34<02:22,  3.78it/s]
overall:  74%|███████▍  | 2979/4000 [14:36<04:33,  3.74it/s]

train:  85%|████████▍ | 2979/3515 [14:34<02:23,  3.74it/s]train loss: 0.5562354922294617
train per: 0.16430884608829044

overall:  75%|███████▍  | 2981/4000 [14:37<03:03,  5.56it/s]

train:  85%|████████▍ | 2981/3515 [14:34<01:36,  5.56it/s]
overall:  75%|███████▍  | 2982/4000 [14:37<03:16,  5.19it/s]

train:  85%|████████▍ | 2982/3515 [14:34<01:42,  5.19it/s]
overall:  75%|███████▍  | 2983/4000 [14:37<03:30,  4.82it/s]

train:  85%|████████▍ | 2983/3515 [14:34<01:50,  4.82it/s]
overall:  75%|███████▍  | 2984/4000 [14:37<03:43,  4.54it/s]

train:  85%|████████▍ | 2984/3515 [14:35<01:57,  4.54it/s]
overall:  75%|███████▍  | 2985/4000 [14:38<04:53,  3.46it/s]

train:  85%|████████▍ | 2985/3515 [14:35<02:33,  3.46it/s]
overall:  75%|███████▍  | 2986/4000 [14:38<04:40,  3.61it/s]

train:  85%|████████▍ | 2986/3515 [14:35<02:26,  3.61it/s]
overall:  75%|███████▍  | 2987/4000 [14:38<04:02,  4.17it/s]

train:  85%|████████▍ | 2987/3515 [14:36<02:06,  4.17it/s]
overall:  75%|███████▍  | 2988/4000 [14:38<04:05,  4.13it/s]

train:  85%|████████▌ | 2988/3515 [14:36<02:07,  4.13it/s]
overall:  75%|███████▍  | 2989/4000 [14:39<04:30,  3.74it/s]

train:  85%|████████▌ | 2989/3515 [14:36<02:20,  3.74it/s]
overall:  75%|███████▍  | 2990/4000 [14:39<04:30,  3.73it/s]

train:  85%|████████▌ | 2990/3515 [14:36<02:20,  3.73it/s]
overall:  75%|███████▍  | 2991/4000 [14:39<04:26,  3.78it/s]

train:  85%|████████▌ | 2991/3515 [14:37<02:18,  3.78it/s]
overall:  75%|███████▍  | 2992/4000 [14:40<04:27,  3.77it/s]

train:  85%|████████▌ | 2992/3515 [14:37<02:18,  3.77it/s]
overall:  75%|███████▍  | 2993/4000 [14:40<04:23,  3.83it/s]

train:  85%|████████▌ | 2993/3515 [14:37<02:16,  3.83it/s]
overall:  75%|███████▍  | 2994/4000 [14:40<04:22,  3.83it/s]

train:  85%|████████▌ | 2994/3515 [14:37<02:15,  3.83it/s]
overall:  75%|███████▍  | 2995/4000 [14:40<04:22,  3.82it/s]

train:  85%|████████▌ | 2995/3515 [14:38<02:15,  3.82it/s]
overall:  75%|███████▍  | 2996/4000 [14:41<04:30,  3.71it/s]

train:  85%|████████▌ | 2996/3515 [14:38<02:19,  3.71it/s]
overall:  75%|███████▍  | 2997/4000 [14:41<04:25,  3.78it/s]

train:  85%|████████▌ | 2997/3515 [14:38<02:17,  3.78it/s]
overall:  75%|███████▍  | 2998/4000 [14:41<04:22,  3.82it/s]

train:  85%|████████▌ | 2998/3515 [14:38<02:15,  3.82it/s]
overall:  75%|███████▍  | 2999/4000 [14:41<04:19,  3.85it/s]

train:  85%|████████▌ | 2999/3515 [14:39<02:14,  3.85it/s]train loss: 0.5529693365097046
train per: 0.16101608112260576
[LibriPhone] - Taking the first phoneme sequences for a deterministic behavior.



word -> phonemes:   0%|          | 0/2703 [00:00<?, ?it/s]


word -> phonemes:  13%|█▎        | 343/2703 [00:00<00:00, 3428.59it/s]


word -> phonemes:  25%|██▌       | 686/2703 [00:00<00:00, 3398.60it/s]


word -> phonemes:  38%|███▊      | 1026/2703 [00:00<00:00, 3349.35it/s]


word -> phonemes:  50%|█████     | 1362/2703 [00:00<00:00, 3216.34it/s]


word -> phonemes:  62%|██████▏   | 1685/2703 [00:00<00:00, 3217.48it/s]


word -> phonemes:  74%|███████▍  | 2008/2703 [00:00<00:00, 3174.40it/s]


word -> phonemes:  87%|████████▋ | 2344/2703 [00:00<00:00, 3232.81it/s]


word -> phonemes:  99%|█████████▉| 2671/2703 [00:00<00:00, 3243.48it/s]
word -> phonemes: 100%|██████████| 2703/2703 [00:00<00:00, 3218.45it/s]



dev:   0%|          | 0/2703 [00:00<?, ?it/s]


dev:   0%|          | 1/2703 [00:01<55:20,  1.23s/it]


dev:   0%|          | 4/2703 [00:01<12:04,  3.73it/s]


dev:   0%|          | 7/2703 [00:01<06:43,  6.68it/s]


dev:   0%|          | 10/2703 [00:01<04:40,  9.59it/s]


dev:   0%|          | 13/2703 [00:01<03:31, 12.74it/s]


dev:   1%|          | 16/2703 [00:01<02:48, 15.98it/s]


dev:   1%|          | 19/2703 [00:01<02:24, 18.63it/s]


dev:   1%|          | 22/2703 [00:02<02:13, 20.11it/s]


dev:   1%|          | 25/2703 [00:02<02:00, 22.28it/s]


dev:   1%|          | 28/2703 [00:02<01:53, 23.54it/s]


dev:   1%|          | 32/2703 [00:02<01:42, 26.15it/s]


dev:   1%|▏         | 35/2703 [00:02<01:38, 27.04it/s]


dev:   1%|▏         | 39/2703 [00:02<01:35, 27.99it/s]


dev:   2%|▏         | 43/2703 [00:02<01:30, 29.40it/s]


dev:   2%|▏         | 47/2703 [00:02<01:27, 30.30it/s]


dev:   2%|▏         | 51/2703 [00:03<01:27, 30.26it/s]


dev:   2%|▏         | 55/2703 [00:03<01:26, 30.71it/s]


dev:   2%|▏         | 59/2703 [00:03<01:21, 32.34it/s]


dev:   2%|▏         | 63/2703 [00:03<01:18, 33.62it/s]


dev:   2%|▏         | 67/2703 [00:03<01:15, 34.89it/s]


dev:   3%|▎         | 71/2703 [00:03<01:16, 34.56it/s]


dev:   3%|▎         | 75/2703 [00:03<01:16, 34.43it/s]


dev:   3%|▎         | 79/2703 [00:03<01:15, 34.68it/s]


dev:   3%|▎         | 83/2703 [00:03<01:15, 34.62it/s]


dev:   3%|▎         | 87/2703 [00:04<01:17, 33.56it/s]


dev:   3%|▎         | 91/2703 [00:04<01:17, 33.79it/s]


dev:   4%|▎         | 95/2703 [00:04<01:15, 34.35it/s]


dev:   4%|▎         | 99/2703 [00:04<01:13, 35.64it/s]


dev:   4%|▍         | 103/2703 [00:04<01:11, 36.42it/s]


dev:   4%|▍         | 107/2703 [00:04<01:13, 35.53it/s]


dev:   4%|▍         | 111/2703 [00:04<01:13, 35.10it/s]


dev:   4%|▍         | 116/2703 [00:04<01:10, 36.63it/s]


dev:   4%|▍         | 120/2703 [00:04<01:11, 36.25it/s]


dev:   5%|▍         | 124/2703 [00:05<01:10, 36.55it/s]


dev:   5%|▍         | 128/2703 [00:05<01:09, 37.25it/s]


dev:   5%|▍         | 132/2703 [00:05<01:08, 37.50it/s]


dev:   5%|▌         | 136/2703 [00:05<01:07, 38.13it/s]


dev:   5%|▌         | 141/2703 [00:05<01:04, 39.55it/s]


dev:   5%|▌         | 145/2703 [00:05<01:07, 38.13it/s]


dev:   6%|▌         | 149/2703 [00:05<01:06, 38.13it/s]


dev:   6%|▌         | 154/2703 [00:05<01:06, 38.39it/s]


dev:   6%|▌         | 158/2703 [00:05<01:06, 38.03it/s]


dev:   6%|▌         | 162/2703 [00:06<01:06, 38.17it/s]


dev:   6%|▌         | 167/2703 [00:06<01:04, 39.30it/s]


dev:   6%|▋         | 172/2703 [00:06<01:03, 39.97it/s]


dev:   7%|▋         | 177/2703 [00:06<01:01, 41.36it/s]


dev:   7%|▋         | 182/2703 [00:06<01:01, 40.99it/s]


dev:   7%|▋         | 187/2703 [00:06<00:59, 42.39it/s]


dev:   7%|▋         | 192/2703 [00:06<00:59, 42.32it/s]


dev:   7%|▋         | 197/2703 [00:06<00:58, 42.59it/s]


dev:   7%|▋         | 202/2703 [00:07<00:57, 43.37it/s]


dev:   8%|▊         | 207/2703 [00:07<00:58, 42.97it/s]


dev:   8%|▊         | 212/2703 [00:07<00:57, 43.05it/s]


dev:   8%|▊         | 217/2703 [00:07<00:57, 43.09it/s]


dev:   8%|▊         | 222/2703 [00:07<00:56, 44.17it/s]


dev:   8%|▊         | 227/2703 [00:07<00:57, 43.06it/s]


dev:   9%|▊         | 232/2703 [00:07<00:56, 43.50it/s]


dev:   9%|▉         | 237/2703 [00:07<00:54, 45.26it/s]


dev:   9%|▉         | 242/2703 [00:07<00:56, 43.94it/s]


dev:   9%|▉         | 247/2703 [00:08<00:55, 44.53it/s]


dev:   9%|▉         | 252/2703 [00:08<00:54, 45.16it/s]


dev:  10%|▉         | 257/2703 [00:08<00:54, 44.78it/s]


dev:  10%|▉         | 262/2703 [00:08<00:54, 44.71it/s]


dev:  10%|▉         | 267/2703 [00:08<00:54, 44.96it/s]


dev:  10%|█         | 272/2703 [00:08<00:53, 45.45it/s]


dev:  10%|█         | 277/2703 [00:08<00:54, 44.90it/s]


dev:  10%|█         | 282/2703 [00:08<00:53, 45.35it/s]


dev:  11%|█         | 287/2703 [00:08<00:51, 46.63it/s]


dev:  11%|█         | 292/2703 [00:09<00:50, 47.55it/s]


dev:  11%|█         | 297/2703 [00:09<00:50, 47.61it/s]


dev:  11%|█         | 302/2703 [00:09<00:51, 46.62it/s]


dev:  11%|█▏        | 308/2703 [00:09<00:49, 48.33it/s]


dev:  12%|█▏        | 313/2703 [00:09<00:51, 46.52it/s]


dev:  12%|█▏        | 318/2703 [00:09<00:50, 47.27it/s]


dev:  12%|█▏        | 323/2703 [00:09<00:50, 46.96it/s]


dev:  12%|█▏        | 328/2703 [00:09<00:50, 47.35it/s]


dev:  12%|█▏        | 333/2703 [00:09<00:49, 47.54it/s]


dev:  13%|█▎        | 338/2703 [00:09<00:49, 48.19it/s]


dev:  13%|█▎        | 344/2703 [00:10<00:48, 48.93it/s]


dev:  13%|█▎        | 349/2703 [00:10<00:49, 47.61it/s]


dev:  13%|█▎        | 355/2703 [00:10<00:47, 49.71it/s]


dev:  13%|█▎        | 361/2703 [00:10<00:47, 48.83it/s]


dev:  14%|█▎        | 367/2703 [00:10<00:46, 49.98it/s]


dev:  14%|█▍        | 373/2703 [00:10<00:46, 50.52it/s]


dev:  14%|█▍        | 379/2703 [00:10<00:45, 51.09it/s]


dev:  14%|█▍        | 385/2703 [00:10<00:45, 50.40it/s]


dev:  14%|█▍        | 391/2703 [00:11<00:45, 50.43it/s]


dev:  15%|█▍        | 397/2703 [00:11<00:45, 50.34it/s]


dev:  15%|█▍        | 403/2703 [00:11<00:46, 49.68it/s]


dev:  15%|█▌        | 409/2703 [00:11<00:45, 49.93it/s]


dev:  15%|█▌        | 414/2703 [00:11<00:45, 49.87it/s]


dev:  16%|█▌        | 420/2703 [00:11<00:45, 50.24it/s]


dev:  16%|█▌        | 426/2703 [00:11<00:45, 50.43it/s]


dev:  16%|█▌        | 432/2703 [00:11<00:43, 51.77it/s]


dev:  16%|█▌        | 438/2703 [00:11<00:43, 51.74it/s]


dev:  16%|█▋        | 444/2703 [00:12<00:43, 52.02it/s]


dev:  17%|█▋        | 450/2703 [00:12<00:43, 52.01it/s]


dev:  17%|█▋        | 456/2703 [00:12<00:43, 52.01it/s]


dev:  17%|█▋        | 462/2703 [00:12<00:42, 52.39it/s]


dev:  17%|█▋        | 468/2703 [00:12<00:43, 50.86it/s]


dev:  18%|█▊        | 474/2703 [00:12<00:43, 50.92it/s]


dev:  18%|█▊        | 480/2703 [00:12<00:43, 51.50it/s]


dev:  18%|█▊        | 486/2703 [00:12<00:42, 52.23it/s]


dev:  18%|█▊        | 492/2703 [00:12<00:41, 53.47it/s]


dev:  18%|█▊        | 498/2703 [00:13<00:40, 54.92it/s]


dev:  19%|█▊        | 504/2703 [00:13<00:39, 55.61it/s]


dev:  19%|█▉        | 510/2703 [00:13<00:40, 54.43it/s]


dev:  19%|█▉        | 516/2703 [00:13<00:39, 55.48it/s]


dev:  19%|█▉        | 522/2703 [00:13<00:40, 54.16it/s]


dev:  20%|█▉        | 528/2703 [00:13<00:39, 55.12it/s]


dev:  20%|█▉        | 534/2703 [00:13<00:39, 54.65it/s]


dev:  20%|█▉        | 540/2703 [00:13<00:39, 55.24it/s]


dev:  20%|██        | 546/2703 [00:13<00:39, 55.00it/s]


dev:  20%|██        | 552/2703 [00:14<00:38, 55.19it/s]


dev:  21%|██        | 558/2703 [00:14<00:38, 56.25it/s]


dev:  21%|██        | 564/2703 [00:14<00:38, 56.15it/s]


dev:  21%|██        | 570/2703 [00:14<00:39, 54.56it/s]


dev:  21%|██▏       | 576/2703 [00:14<00:39, 54.24it/s]


dev:  22%|██▏       | 582/2703 [00:14<00:38, 54.78it/s]


dev:  22%|██▏       | 588/2703 [00:14<00:38, 54.95it/s]


dev:  22%|██▏       | 594/2703 [00:14<00:38, 55.30it/s]


dev:  22%|██▏       | 600/2703 [00:14<00:38, 55.21it/s]


dev:  22%|██▏       | 606/2703 [00:15<00:38, 55.13it/s]


dev:  23%|██▎       | 612/2703 [00:15<00:38, 54.72it/s]


dev:  23%|██▎       | 618/2703 [00:15<00:38, 54.51it/s]


dev:  23%|██▎       | 624/2703 [00:15<00:37, 55.65it/s]


dev:  23%|██▎       | 631/2703 [00:15<00:36, 57.06it/s]


dev:  24%|██▎       | 637/2703 [00:15<00:37, 55.57it/s]


dev:  24%|██▍       | 644/2703 [00:15<00:36, 57.16it/s]


dev:  24%|██▍       | 650/2703 [00:15<00:36, 56.38it/s]


dev:  24%|██▍       | 656/2703 [00:15<00:35, 57.19it/s]


dev:  24%|██▍       | 662/2703 [00:16<00:36, 56.27it/s]


dev:  25%|██▍       | 668/2703 [00:16<00:35, 56.95it/s]


dev:  25%|██▍       | 674/2703 [00:16<00:35, 56.53it/s]


dev:  25%|██▌       | 680/2703 [00:16<00:36, 56.14it/s]


dev:  25%|██▌       | 686/2703 [00:16<00:35, 56.83it/s]


dev:  26%|██▌       | 692/2703 [00:16<00:35, 57.38it/s]


dev:  26%|██▌       | 698/2703 [00:16<00:35, 57.15it/s]


dev:  26%|██▌       | 704/2703 [00:16<00:34, 57.35it/s]


dev:  26%|██▋       | 710/2703 [00:16<00:35, 56.34it/s]


dev:  27%|██▋       | 717/2703 [00:16<00:34, 57.78it/s]


dev:  27%|██▋       | 724/2703 [00:17<00:33, 58.82it/s]


dev:  27%|██▋       | 730/2703 [00:17<00:33, 58.31it/s]


dev:  27%|██▋       | 737/2703 [00:17<00:33, 58.75it/s]


dev:  27%|██▋       | 743/2703 [00:17<00:33, 58.51it/s]


dev:  28%|██▊       | 750/2703 [00:17<00:32, 59.71it/s]


dev:  28%|██▊       | 756/2703 [00:17<00:32, 59.08it/s]


dev:  28%|██▊       | 762/2703 [00:17<00:32, 59.25it/s]


dev:  28%|██▊       | 768/2703 [00:17<00:33, 57.89it/s]


dev:  29%|██▊       | 775/2703 [00:17<00:32, 60.09it/s]


dev:  29%|██▉       | 782/2703 [00:18<00:31, 60.07it/s]


dev:  29%|██▉       | 789/2703 [00:18<00:31, 61.26it/s]


dev:  29%|██▉       | 796/2703 [00:18<00:31, 61.13it/s]


dev:  30%|██▉       | 803/2703 [00:18<00:31, 60.40it/s]


dev:  30%|██▉       | 810/2703 [00:18<00:30, 61.91it/s]


dev:  30%|███       | 817/2703 [00:18<00:30, 61.41it/s]


dev:  30%|███       | 824/2703 [00:18<00:30, 62.45it/s]


dev:  31%|███       | 831/2703 [00:18<00:29, 62.88it/s]


dev:  31%|███       | 838/2703 [00:18<00:29, 62.47it/s]


dev:  31%|███▏      | 845/2703 [00:19<00:29, 62.70it/s]


dev:  32%|███▏      | 852/2703 [00:19<00:29, 63.02it/s]


dev:  32%|███▏      | 859/2703 [00:19<00:29, 62.36it/s]


dev:  32%|███▏      | 866/2703 [00:19<00:29, 62.56it/s]


dev:  32%|███▏      | 873/2703 [00:19<00:28, 63.69it/s]


dev:  33%|███▎      | 880/2703 [00:19<00:28, 64.00it/s]


dev:  33%|███▎      | 887/2703 [00:19<00:28, 63.49it/s]


dev:  33%|███▎      | 894/2703 [00:19<00:28, 63.21it/s]


dev:  33%|███▎      | 901/2703 [00:19<00:28, 62.84it/s]


dev:  34%|███▎      | 908/2703 [00:20<00:28, 63.67it/s]


dev:  34%|███▍      | 915/2703 [00:20<00:28, 62.87it/s]


dev:  34%|███▍      | 922/2703 [00:20<00:28, 62.29it/s]


dev:  34%|███▍      | 929/2703 [00:20<00:27, 63.73it/s]


dev:  35%|███▍      | 936/2703 [00:20<00:27, 63.37it/s]


dev:  35%|███▍      | 943/2703 [00:20<00:28, 62.69it/s]


dev:  35%|███▌      | 950/2703 [00:20<00:27, 62.94it/s]


dev:  35%|███▌      | 957/2703 [00:20<00:27, 63.80it/s]


dev:  36%|███▌      | 964/2703 [00:20<00:26, 64.53it/s]


dev:  36%|███▌      | 971/2703 [00:21<00:27, 63.06it/s]


dev:  36%|███▌      | 978/2703 [00:21<00:26, 64.66it/s]


dev:  36%|███▋      | 985/2703 [00:21<00:26, 64.11it/s]


dev:  37%|███▋      | 992/2703 [00:21<00:27, 61.65it/s]


dev:  37%|███▋      | 999/2703 [00:21<00:26, 63.43it/s]


dev:  37%|███▋      | 1006/2703 [00:21<00:26, 63.40it/s]


dev:  37%|███▋      | 1013/2703 [00:21<00:26, 64.71it/s]


dev:  38%|███▊      | 1020/2703 [00:21<00:25, 64.79it/s]


dev:  38%|███▊      | 1027/2703 [00:21<00:26, 64.30it/s]


dev:  38%|███▊      | 1034/2703 [00:22<00:25, 64.85it/s]


dev:  39%|███▊      | 1041/2703 [00:22<00:26, 63.57it/s]


dev:  39%|███▉      | 1048/2703 [00:22<00:26, 63.61it/s]


dev:  39%|███▉      | 1055/2703 [00:22<00:25, 64.43it/s]


dev:  39%|███▉      | 1062/2703 [00:22<00:24, 65.67it/s]


dev:  40%|███▉      | 1069/2703 [00:22<00:24, 65.98it/s]


dev:  40%|███▉      | 1076/2703 [00:22<00:24, 66.41it/s]


dev:  40%|████      | 1083/2703 [00:22<00:24, 66.94it/s]


dev:  40%|████      | 1090/2703 [00:22<00:24, 66.28it/s]


dev:  41%|████      | 1097/2703 [00:23<00:24, 66.59it/s]


dev:  41%|████      | 1104/2703 [00:23<00:24, 66.59it/s]


dev:  41%|████      | 1111/2703 [00:23<00:23, 66.51it/s]


dev:  41%|████▏     | 1118/2703 [00:23<00:23, 67.19it/s]


dev:  42%|████▏     | 1125/2703 [00:23<00:23, 67.38it/s]


dev:  42%|████▏     | 1132/2703 [00:23<00:23, 66.38it/s]


dev:  42%|████▏     | 1139/2703 [00:23<00:23, 67.39it/s]


dev:  42%|████▏     | 1146/2703 [00:23<00:23, 66.77it/s]


dev:  43%|████▎     | 1153/2703 [00:23<00:23, 65.75it/s]


dev:  43%|████▎     | 1161/2703 [00:23<00:22, 67.93it/s]


dev:  43%|████▎     | 1168/2703 [00:24<00:23, 66.66it/s]


dev:  43%|████▎     | 1175/2703 [00:24<00:22, 66.74it/s]


dev:  44%|████▎     | 1182/2703 [00:24<00:22, 67.28it/s]


dev:  44%|████▍     | 1189/2703 [00:24<00:22, 66.86it/s]


dev:  44%|████▍     | 1197/2703 [00:24<00:22, 68.18it/s]


dev:  45%|████▍     | 1204/2703 [00:24<00:22, 67.37it/s]


dev:  45%|████▍     | 1211/2703 [00:24<00:21, 68.08it/s]


dev:  45%|████▌     | 1218/2703 [00:24<00:21, 67.51it/s]


dev:  45%|████▌     | 1226/2703 [00:24<00:21, 68.76it/s]


dev:  46%|████▌     | 1234/2703 [00:25<00:21, 68.73it/s]


dev:  46%|████▌     | 1242/2703 [00:25<00:21, 69.40it/s]


dev:  46%|████▌     | 1249/2703 [00:25<00:20, 69.41it/s]


dev:  46%|████▋     | 1256/2703 [00:25<00:21, 68.90it/s]


dev:  47%|████▋     | 1264/2703 [00:25<00:20, 69.34it/s]


dev:  47%|████▋     | 1272/2703 [00:25<00:20, 70.07it/s]


dev:  47%|████▋     | 1280/2703 [00:25<00:20, 69.98it/s]


dev:  48%|████▊     | 1287/2703 [00:25<00:20, 69.63it/s]


dev:  48%|████▊     | 1294/2703 [00:25<00:20, 69.62it/s]


dev:  48%|████▊     | 1302/2703 [00:26<00:19, 70.58it/s]


dev:  48%|████▊     | 1310/2703 [00:26<00:19, 70.77it/s]


dev:  49%|████▉     | 1318/2703 [00:26<00:19, 70.18it/s]


dev:  49%|████▉     | 1326/2703 [00:26<00:19, 70.50it/s]


dev:  49%|████▉     | 1334/2703 [00:26<00:19, 70.07it/s]


dev:  50%|████▉     | 1342/2703 [00:26<00:19, 69.94it/s]


dev:  50%|████▉     | 1350/2703 [00:26<00:19, 70.51it/s]


dev:  50%|█████     | 1358/2703 [00:26<00:19, 69.96it/s]


dev:  51%|█████     | 1366/2703 [00:26<00:19, 69.20it/s]


dev:  51%|█████     | 1374/2703 [00:27<00:19, 69.74it/s]


dev:  51%|█████     | 1382/2703 [00:27<00:18, 70.36it/s]


dev:  51%|█████▏    | 1390/2703 [00:27<00:19, 69.08it/s]


dev:  52%|█████▏    | 1398/2703 [00:27<00:18, 70.50it/s]


dev:  52%|█████▏    | 1406/2703 [00:27<00:18, 70.24it/s]


dev:  52%|█████▏    | 1414/2703 [00:27<00:18, 70.45it/s]


dev:  53%|█████▎    | 1422/2703 [00:27<00:17, 71.87it/s]


dev:  53%|█████▎    | 1430/2703 [00:27<00:17, 71.59it/s]


dev:  53%|█████▎    | 1438/2703 [00:27<00:17, 71.92it/s]


dev:  53%|█████▎    | 1446/2703 [00:28<00:17, 72.30it/s]


dev:  54%|█████▍    | 1454/2703 [00:28<00:17, 71.98it/s]


dev:  54%|█████▍    | 1462/2703 [00:28<00:17, 72.59it/s]


dev:  54%|█████▍    | 1470/2703 [00:28<00:17, 72.39it/s]


dev:  55%|█████▍    | 1478/2703 [00:28<00:16, 72.23it/s]


dev:  55%|█████▍    | 1486/2703 [00:28<00:16, 72.83it/s]


dev:  55%|█████▌    | 1494/2703 [00:28<00:16, 73.33it/s]


dev:  56%|█████▌    | 1502/2703 [00:28<00:16, 73.55it/s]


dev:  56%|█████▌    | 1510/2703 [00:28<00:16, 72.82it/s]


dev:  56%|█████▌    | 1518/2703 [00:29<00:16, 72.45it/s]


dev:  56%|█████▋    | 1526/2703 [00:29<00:16, 72.90it/s]


dev:  57%|█████▋    | 1534/2703 [00:29<00:15, 73.18it/s]


dev:  57%|█████▋    | 1542/2703 [00:29<00:16, 72.32it/s]


dev:  57%|█████▋    | 1550/2703 [00:29<00:15, 73.02it/s]


dev:  58%|█████▊    | 1558/2703 [00:29<00:15, 72.71it/s]


dev:  58%|█████▊    | 1566/2703 [00:29<00:15, 72.65it/s]


dev:  58%|█████▊    | 1574/2703 [00:29<00:15, 73.12it/s]


dev:  59%|█████▊    | 1582/2703 [00:29<00:15, 71.65it/s]


dev:  59%|█████▉    | 1590/2703 [00:30<00:15, 72.64it/s]


dev:  59%|█████▉    | 1598/2703 [00:30<00:15, 73.52it/s]


dev:  59%|█████▉    | 1606/2703 [00:30<00:15, 72.43it/s]


dev:  60%|█████▉    | 1614/2703 [00:30<00:14, 73.21it/s]


dev:  60%|██████    | 1622/2703 [00:30<00:15, 71.85it/s]


dev:  60%|██████    | 1630/2703 [00:30<00:14, 72.84it/s]


dev:  61%|██████    | 1638/2703 [00:30<00:14, 73.51it/s]


dev:  61%|██████    | 1646/2703 [00:30<00:14, 73.30it/s]


dev:  61%|██████    | 1654/2703 [00:30<00:14, 73.57it/s]


dev:  61%|██████▏   | 1662/2703 [00:31<00:14, 73.48it/s]


dev:  62%|██████▏   | 1670/2703 [00:31<00:14, 72.52it/s]


dev:  62%|██████▏   | 1678/2703 [00:31<00:14, 71.81it/s]


dev:  62%|██████▏   | 1686/2703 [00:31<00:14, 71.98it/s]


dev:  63%|██████▎   | 1694/2703 [00:31<00:14, 70.96it/s]


dev:  63%|██████▎   | 1702/2703 [00:31<00:14, 71.37it/s]


dev:  63%|██████▎   | 1710/2703 [00:31<00:14, 69.79it/s]


dev:  64%|██████▎   | 1718/2703 [00:31<00:13, 71.04it/s]


dev:  64%|██████▍   | 1726/2703 [00:31<00:13, 70.35it/s]


dev:  64%|██████▍   | 1734/2703 [00:32<00:13, 71.12it/s]


dev:  64%|██████▍   | 1742/2703 [00:32<00:13, 72.16it/s]


dev:  65%|██████▍   | 1750/2703 [00:32<00:13, 70.96it/s]


dev:  65%|██████▌   | 1758/2703 [00:32<00:13, 71.50it/s]


dev:  65%|██████▌   | 1766/2703 [00:32<00:13, 71.06it/s]


dev:  66%|██████▌   | 1774/2703 [00:32<00:13, 71.24it/s]


dev:  66%|██████▌   | 1782/2703 [00:32<00:12, 71.98it/s]


dev:  66%|██████▌   | 1790/2703 [00:32<00:12, 71.47it/s]


dev:  67%|██████▋   | 1798/2703 [00:32<00:12, 72.10it/s]


dev:  67%|██████▋   | 1806/2703 [00:33<00:12, 71.58it/s]


dev:  67%|██████▋   | 1814/2703 [00:33<00:12, 72.73it/s]


dev:  67%|██████▋   | 1822/2703 [00:33<00:12, 73.06it/s]


dev:  68%|██████▊   | 1830/2703 [00:33<00:12, 72.30it/s]


dev:  68%|██████▊   | 1838/2703 [00:33<00:11, 73.20it/s]


dev:  68%|██████▊   | 1846/2703 [00:33<00:11, 72.15it/s]


dev:  69%|██████▊   | 1854/2703 [00:33<00:11, 71.66it/s]


dev:  69%|██████▉   | 1862/2703 [00:33<00:11, 72.71it/s]


dev:  69%|██████▉   | 1870/2703 [00:33<00:11, 72.40it/s]


dev:  69%|██████▉   | 1878/2703 [00:34<00:11, 72.51it/s]


dev:  70%|██████▉   | 1886/2703 [00:34<00:11, 73.27it/s]


dev:  70%|███████   | 1894/2703 [00:34<00:11, 73.33it/s]


dev:  70%|███████   | 1902/2703 [00:34<00:10, 73.47it/s]


dev:  71%|███████   | 1910/2703 [00:34<00:10, 74.02it/s]


dev:  71%|███████   | 1918/2703 [00:34<00:10, 71.44it/s]


dev:  71%|███████▏  | 1926/2703 [00:34<00:10, 71.95it/s]


dev:  72%|███████▏  | 1934/2703 [00:34<00:10, 71.14it/s]


dev:  72%|███████▏  | 1942/2703 [00:34<00:10, 71.76it/s]


dev:  72%|███████▏  | 1950/2703 [00:35<00:10, 72.58it/s]


dev:  72%|███████▏  | 1958/2703 [00:35<00:10, 72.24it/s]


dev:  73%|███████▎  | 1966/2703 [00:35<00:10, 72.50it/s]


dev:  73%|███████▎  | 1974/2703 [00:35<00:10, 72.03it/s]


dev:  73%|███████▎  | 1982/2703 [00:35<00:09, 73.04it/s]


dev:  74%|███████▎  | 1990/2703 [00:35<00:09, 72.03it/s]


dev:  74%|███████▍  | 1998/2703 [00:35<00:09, 72.47it/s]


dev:  74%|███████▍  | 2006/2703 [00:35<00:09, 73.23it/s]


dev:  75%|███████▍  | 2014/2703 [00:35<00:09, 72.01it/s]


dev:  75%|███████▍  | 2022/2703 [00:36<00:09, 72.61it/s]


dev:  75%|███████▌  | 2030/2703 [00:36<00:09, 72.19it/s]


dev:  75%|███████▌  | 2038/2703 [00:36<00:09, 72.54it/s]


dev:  76%|███████▌  | 2046/2703 [00:36<00:09, 72.93it/s]


dev:  76%|███████▌  | 2054/2703 [00:36<00:08, 72.12it/s]


dev:  76%|███████▋  | 2062/2703 [00:36<00:08, 72.43it/s]


dev:  77%|███████▋  | 2070/2703 [00:36<00:08, 72.50it/s]


dev:  77%|███████▋  | 2078/2703 [00:36<00:08, 73.41it/s]


dev:  77%|███████▋  | 2086/2703 [00:36<00:08, 72.98it/s]


dev:  77%|███████▋  | 2094/2703 [00:37<00:08, 73.32it/s]


dev:  78%|███████▊  | 2102/2703 [00:37<00:08, 73.69it/s]


dev:  78%|███████▊  | 2110/2703 [00:37<00:08, 73.87it/s]


dev:  78%|███████▊  | 2118/2703 [00:37<00:08, 72.12it/s]


dev:  79%|███████▊  | 2126/2703 [00:37<00:07, 72.99it/s]


dev:  79%|███████▉  | 2134/2703 [00:37<00:07, 73.57it/s]


dev:  79%|███████▉  | 2142/2703 [00:37<00:07, 73.12it/s]


dev:  80%|███████▉  | 2150/2703 [00:37<00:07, 73.71it/s]


dev:  80%|███████▉  | 2158/2703 [00:37<00:07, 72.69it/s]


dev:  80%|████████  | 2166/2703 [00:37<00:07, 73.38it/s]


dev:  80%|████████  | 2174/2703 [00:38<00:07, 72.61it/s]


dev:  81%|████████  | 2182/2703 [00:38<00:07, 71.54it/s]


dev:  81%|████████  | 2190/2703 [00:38<00:07, 72.63it/s]


dev:  81%|████████▏ | 2198/2703 [00:38<00:06, 73.12it/s]


dev:  82%|████████▏ | 2206/2703 [00:38<00:06, 73.36it/s]


dev:  82%|████████▏ | 2214/2703 [00:38<00:06, 74.04it/s]


dev:  82%|████████▏ | 2222/2703 [00:38<00:06, 72.66it/s]


dev:  83%|████████▎ | 2230/2703 [00:38<00:06, 73.24it/s]


dev:  83%|████████▎ | 2238/2703 [00:38<00:06, 73.21it/s]


dev:  83%|████████▎ | 2246/2703 [00:39<00:06, 73.72it/s]


dev:  83%|████████▎ | 2254/2703 [00:39<00:06, 72.42it/s]


dev:  84%|████████▎ | 2262/2703 [00:39<00:06, 73.00it/s]


dev:  84%|████████▍ | 2270/2703 [00:39<00:05, 73.76it/s]


dev:  84%|████████▍ | 2278/2703 [00:39<00:05, 74.40it/s]


dev:  85%|████████▍ | 2286/2703 [00:39<00:05, 74.27it/s]


dev:  85%|████████▍ | 2294/2703 [00:39<00:05, 73.73it/s]


dev:  85%|████████▌ | 2302/2703 [00:39<00:05, 72.55it/s]


dev:  85%|████████▌ | 2310/2703 [00:39<00:05, 73.13it/s]


dev:  86%|████████▌ | 2318/2703 [00:40<00:05, 73.45it/s]


dev:  86%|████████▌ | 2326/2703 [00:40<00:05, 72.04it/s]


dev:  86%|████████▋ | 2334/2703 [00:40<00:05, 72.67it/s]


dev:  87%|████████▋ | 2342/2703 [00:40<00:05, 71.97it/s]


dev:  87%|████████▋ | 2350/2703 [00:40<00:04, 72.84it/s]


dev:  87%|████████▋ | 2358/2703 [00:40<00:04, 73.23it/s]


dev:  88%|████████▊ | 2366/2703 [00:40<00:04, 71.92it/s]


dev:  88%|████████▊ | 2374/2703 [00:40<00:04, 72.97it/s]


dev:  88%|████████▊ | 2382/2703 [00:40<00:04, 73.67it/s]


dev:  88%|████████▊ | 2390/2703 [00:41<00:04, 73.18it/s]


dev:  89%|████████▊ | 2398/2703 [00:41<00:04, 73.34it/s]


dev:  89%|████████▉ | 2406/2703 [00:41<00:04, 73.79it/s]


dev:  89%|████████▉ | 2415/2703 [00:41<00:03, 76.30it/s]


dev:  90%|████████▉ | 2423/2703 [00:41<00:03, 75.71it/s]


dev:  90%|████████▉ | 2431/2703 [00:41<00:03, 74.88it/s]


dev:  90%|█████████ | 2439/2703 [00:41<00:03, 75.10it/s]


dev:  91%|█████████ | 2447/2703 [00:41<00:03, 75.22it/s]


dev:  91%|█████████ | 2455/2703 [00:41<00:03, 75.17it/s]


dev:  91%|█████████ | 2463/2703 [00:42<00:03, 75.53it/s]


dev:  91%|█████████▏| 2471/2703 [00:42<00:03, 74.56it/s]


dev:  92%|█████████▏| 2479/2703 [00:42<00:03, 74.31it/s]


dev:  92%|█████████▏| 2487/2703 [00:42<00:02, 73.06it/s]


dev:  92%|█████████▏| 2495/2703 [00:42<00:02, 73.40it/s]


dev:  93%|█████████▎| 2503/2703 [00:42<00:02, 74.49it/s]


dev:  93%|█████████▎| 2511/2703 [00:42<00:02, 72.36it/s]


dev:  93%|█████████▎| 2519/2703 [00:42<00:02, 73.12it/s]


dev:  93%|█████████▎| 2527/2703 [00:42<00:02, 73.85it/s]


dev:  94%|█████████▍| 2535/2703 [00:43<00:02, 73.55it/s]


dev:  94%|█████████▍| 2543/2703 [00:43<00:02, 70.71it/s]


dev:  94%|█████████▍| 2551/2703 [00:43<00:02, 70.52it/s]


dev:  95%|█████████▍| 2559/2703 [00:43<00:01, 72.81it/s]


dev:  95%|█████████▌| 2568/2703 [00:43<00:01, 74.66it/s]


dev:  95%|█████████▌| 2576/2703 [00:43<00:01, 75.53it/s]


dev:  96%|█████████▌| 2585/2703 [00:43<00:01, 77.30it/s]


dev:  96%|█████████▌| 2593/2703 [00:43<00:01, 77.23it/s]


dev:  96%|█████████▌| 2601/2703 [00:43<00:01, 77.52it/s]


dev:  97%|█████████▋| 2609/2703 [00:43<00:01, 75.12it/s]


dev:  97%|█████████▋| 2617/2703 [00:44<00:01, 75.20it/s]


dev:  97%|█████████▋| 2626/2703 [00:44<00:01, 76.95it/s]


dev:  97%|█████████▋| 2634/2703 [00:44<00:00, 75.55it/s]


dev:  98%|█████████▊| 2642/2703 [00:44<00:00, 76.55it/s]


dev:  98%|█████████▊| 2650/2703 [00:44<00:00, 76.38it/s]


dev:  98%|█████████▊| 2659/2703 [00:44<00:00, 78.71it/s]


dev:  99%|█████████▊| 2667/2703 [00:44<00:00, 77.52it/s]


dev:  99%|█████████▉| 2676/2703 [00:44<00:00, 80.87it/s]


dev:  99%|█████████▉| 2685/2703 [00:44<00:00, 82.29it/s]


dev: 100%|█████████▉| 2694/2703 [00:45<00:00, 82.92it/s]
dev: 100%|██████████| 2703/2703 [00:45<00:00, 59.65it/s]
dev loss: 0.765929102897644
dev per: 0.1968054110301769

overall:  75%|███████▌  | 3000/4000 [15:29<4:00:37, 14.44s/it]

train:  85%|████████▌ | 3000/3515 [15:26<2:03:55, 14.44s/it]
overall:  75%|███████▌  | 3001/4000 [15:29<2:49:36, 10.19s/it]

train:  85%|████████▌ | 3001/3515 [15:27<1:27:15, 10.19s/it]
overall:  75%|███████▌  | 3002/4000 [15:29<1:59:51,  7.21s/it]

train:  85%|████████▌ | 3002/3515 [15:27<1:01:36,  7.21s/it]
overall:  75%|███████▌  | 3003/4000 [15:30<1:25:04,  5.12s/it]

train:  85%|████████▌ | 3003/3515 [15:27<43:41,  5.12s/it]  
overall:  75%|███████▌  | 3004/4000 [15:30<1:00:26,  3.64s/it]

train:  85%|████████▌ | 3004/3515 [15:27<31:00,  3.64s/it]
overall:  75%|███████▌  | 3005/4000 [15:30<43:37,  2.63s/it]  

train:  85%|████████▌ | 3005/3515 [15:28<22:21,  2.63s/it]
overall:  75%|███████▌  | 3006/4000 [15:30<31:48,  1.92s/it]

train:  86%|████████▌ | 3006/3515 [15:28<16:17,  1.92s/it]
overall:  75%|███████▌  | 3007/4000 [15:31<23:33,  1.42s/it]

train:  86%|████████▌ | 3007/3515 [15:28<12:03,  1.42s/it]
overall:  75%|███████▌  | 3008/4000 [15:31<17:35,  1.06s/it]

train:  86%|████████▌ | 3008/3515 [15:28<08:59,  1.06s/it]
overall:  75%|███████▌  | 3009/4000 [15:31<14:19,  1.15it/s]

train:  86%|████████▌ | 3009/3515 [15:29<07:19,  1.15it/s]
overall:  75%|███████▌  | 3010/4000 [15:32<11:22,  1.45it/s]

train:  86%|████████▌ | 3010/3515 [15:29<05:48,  1.45it/s]
overall:  75%|███████▌  | 3011/4000 [15:32<09:19,  1.77it/s]

train:  86%|████████▌ | 3011/3515 [15:29<04:45,  1.77it/s]
overall:  75%|███████▌  | 3012/4000 [15:32<07:54,  2.08it/s]

train:  86%|████████▌ | 3012/3515 [15:30<04:01,  2.08it/s]
overall:  75%|███████▌  | 3013/4000 [15:32<07:06,  2.32it/s]

train:  86%|████████▌ | 3013/3515 [15:30<03:36,  2.32it/s]
overall:  75%|███████▌  | 3014/4000 [15:33<06:16,  2.62it/s]

train:  86%|████████▌ | 3014/3515 [15:30<03:11,  2.62it/s]
overall:  75%|███████▌  | 3016/4000 [15:33<04:42,  3.49it/s]

train:  86%|████████▌ | 3016/3515 [15:30<02:23,  3.49it/s]
overall:  75%|███████▌  | 3017/4000 [15:33<04:38,  3.54it/s]

train:  86%|████████▌ | 3017/3515 [15:31<02:20,  3.54it/s]
overall:  75%|███████▌  | 3018/4000 [15:34<04:34,  3.58it/s]

train:  86%|████████▌ | 3018/3515 [15:31<02:18,  3.58it/s]
overall:  75%|███████▌  | 3019/4000 [15:34<04:37,  3.54it/s]

train:  86%|████████▌ | 3019/3515 [15:31<02:20,  3.54it/s]train loss: 0.5525632500648499
train per: 0.16491210768965262

overall:  76%|███████▌  | 3020/4000 [15:34<04:35,  3.55it/s]

train:  86%|████████▌ | 3020/3515 [15:32<02:19,  3.55it/s]
overall:  76%|███████▌  | 3021/4000 [15:34<04:36,  3.54it/s]

train:  86%|████████▌ | 3021/3515 [15:32<02:19,  3.54it/s]
overall:  76%|███████▌  | 3022/4000 [15:35<04:23,  3.71it/s]

train:  86%|████████▌ | 3022/3515 [15:32<02:12,  3.71it/s]
overall:  76%|███████▌  | 3023/4000 [15:35<03:35,  4.54it/s]

train:  86%|████████▌ | 3023/3515 [15:32<01:48,  4.54it/s]
overall:  76%|███████▌  | 3024/4000 [15:35<03:46,  4.31it/s]

train:  86%|████████▌ | 3024/3515 [15:32<01:53,  4.31it/s]
overall:  76%|███████▌  | 3025/4000 [15:35<03:51,  4.21it/s]

train:  86%|████████▌ | 3025/3515 [15:33<01:56,  4.21it/s]
overall:  76%|███████▌  | 3026/4000 [15:36<04:04,  3.99it/s]

train:  86%|████████▌ | 3026/3515 [15:33<02:02,  3.99it/s]
overall:  76%|███████▌  | 3027/4000 [15:36<03:56,  4.11it/s]

train:  86%|████████▌ | 3027/3515 [15:33<01:58,  4.11it/s]
overall:  76%|███████▌  | 3028/4000 [15:36<04:05,  3.97it/s]

train:  86%|████████▌ | 3028/3515 [15:33<02:02,  3.97it/s]
overall:  76%|███████▌  | 3029/4000 [15:36<04:12,  3.85it/s]

train:  86%|████████▌ | 3029/3515 [15:34<02:06,  3.85it/s]
overall:  76%|███████▌  | 3030/4000 [15:37<04:11,  3.86it/s]

train:  86%|████████▌ | 3030/3515 [15:34<02:05,  3.86it/s]
overall:  76%|███████▌  | 3031/4000 [15:37<05:25,  2.98it/s]

train:  86%|████████▌ | 3031/3515 [15:35<02:42,  2.98it/s]
overall:  76%|███████▌  | 3032/4000 [15:37<05:00,  3.22it/s]

train:  86%|████████▋ | 3032/3515 [15:35<02:30,  3.22it/s]
overall:  76%|███████▌  | 3033/4000 [15:38<05:32,  2.91it/s]

train:  86%|████████▋ | 3033/3515 [15:35<02:45,  2.91it/s]
overall:  76%|███████▌  | 3034/4000 [15:38<05:06,  3.15it/s]

train:  86%|████████▋ | 3034/3515 [15:35<02:32,  3.15it/s]
overall:  76%|███████▌  | 3035/4000 [15:38<04:48,  3.34it/s]

train:  86%|████████▋ | 3035/3515 [15:36<02:23,  3.34it/s]
overall:  76%|███████▌  | 3036/4000 [15:39<04:46,  3.37it/s]

train:  86%|████████▋ | 3036/3515 [15:36<02:22,  3.37it/s]
overall:  76%|███████▌  | 3037/4000 [15:39<04:54,  3.27it/s]

train:  86%|████████▋ | 3037/3515 [15:36<02:26,  3.27it/s]
overall:  76%|███████▌  | 3038/4000 [15:39<04:44,  3.39it/s]

train:  86%|████████▋ | 3038/3515 [15:37<02:20,  3.39it/s]
overall:  76%|███████▌  | 3039/4000 [15:39<04:35,  3.49it/s]

train:  86%|████████▋ | 3039/3515 [15:37<02:16,  3.49it/s]train loss: 0.5604037046432495
train per: 0.16122340682136138

overall:  76%|███████▌  | 3040/4000 [15:40<04:09,  3.84it/s]

train:  86%|████████▋ | 3040/3515 [15:37<02:03,  3.84it/s]
overall:  76%|███████▌  | 3041/4000 [15:40<04:08,  3.87it/s]

train:  87%|████████▋ | 3041/3515 [15:37<02:02,  3.87it/s]
overall:  76%|███████▌  | 3042/4000 [15:40<04:07,  3.87it/s]

train:  87%|████████▋ | 3042/3515 [15:38<02:02,  3.87it/s]
overall:  76%|███████▌  | 3043/4000 [15:40<04:02,  3.94it/s]

train:  87%|████████▋ | 3043/3515 [15:38<01:59,  3.94it/s]
overall:  76%|███████▌  | 3044/4000 [15:41<04:14,  3.75it/s]

train:  87%|████████▋ | 3044/3515 [15:38<02:05,  3.75it/s]
overall:  76%|███████▌  | 3045/4000 [15:41<04:13,  3.76it/s]

train:  87%|████████▋ | 3045/3515 [15:38<02:04,  3.76it/s]
overall:  76%|███████▌  | 3046/4000 [15:41<04:07,  3.85it/s]

train:  87%|████████▋ | 3046/3515 [15:39<02:01,  3.85it/s]
overall:  76%|███████▌  | 3047/4000 [15:42<04:08,  3.83it/s]

train:  87%|████████▋ | 3047/3515 [15:39<02:02,  3.83it/s]
overall:  76%|███████▌  | 3048/4000 [15:42<04:12,  3.77it/s]

train:  87%|████████▋ | 3048/3515 [15:39<02:03,  3.77it/s]
overall:  76%|███████▌  | 3049/4000 [15:42<04:01,  3.93it/s]

train:  87%|████████▋ | 3049/3515 [15:39<01:58,  3.93it/s]
overall:  76%|███████▋  | 3050/4000 [15:42<03:17,  4.80it/s]

train:  87%|████████▋ | 3050/3515 [15:39<01:36,  4.80it/s]
overall:  76%|███████▋  | 3051/4000 [15:42<03:15,  4.84it/s]

train:  87%|████████▋ | 3051/3515 [15:40<01:35,  4.84it/s]
overall:  76%|███████▋  | 3052/4000 [15:43<03:32,  4.46it/s]

train:  87%|████████▋ | 3052/3515 [15:40<01:43,  4.46it/s]
overall:  76%|███████▋  | 3053/4000 [15:43<03:45,  4.20it/s]

train:  87%|████████▋ | 3053/3515 [15:40<01:50,  4.20it/s]
overall:  76%|███████▋  | 3054/4000 [15:43<03:54,  4.03it/s]

train:  87%|████████▋ | 3054/3515 [15:40<01:54,  4.03it/s]
overall:  76%|███████▋  | 3055/4000 [15:43<03:57,  3.98it/s]

train:  87%|████████▋ | 3055/3515 [15:41<01:55,  3.99it/s]
overall:  76%|███████▋  | 3056/4000 [15:44<04:00,  3.92it/s]

train:  87%|████████▋ | 3056/3515 [15:41<01:57,  3.92it/s]
overall:  76%|███████▋  | 3057/4000 [15:44<04:50,  3.25it/s]

train:  87%|████████▋ | 3057/3515 [15:41<02:20,  3.25it/s]
overall:  76%|███████▋  | 3058/4000 [15:44<04:37,  3.39it/s]

train:  87%|████████▋ | 3058/3515 [15:42<02:14,  3.39it/s]
overall:  76%|███████▋  | 3059/4000 [15:45<04:30,  3.48it/s]

train:  87%|████████▋ | 3059/3515 [15:42<02:11,  3.48it/s]train loss: 0.5410913228988647
train per: 0.15868098085837115

overall:  76%|███████▋  | 3060/4000 [15:45<04:12,  3.72it/s]

train:  87%|████████▋ | 3060/3515 [15:42<02:02,  3.71it/s]
overall:  77%|███████▋  | 3061/4000 [15:45<04:27,  3.51it/s]

train:  87%|████████▋ | 3061/3515 [15:43<02:09,  3.51it/s]
overall:  77%|███████▋  | 3062/4000 [15:45<04:27,  3.51it/s]

train:  87%|████████▋ | 3062/3515 [15:43<02:08,  3.51it/s]
overall:  77%|███████▋  | 3063/4000 [15:46<04:19,  3.61it/s]

train:  87%|████████▋ | 3063/3515 [15:43<02:05,  3.61it/s]
overall:  77%|███████▋  | 3064/4000 [15:46<04:23,  3.55it/s]

train:  87%|████████▋ | 3064/3515 [15:43<02:07,  3.55it/s]
overall:  77%|███████▋  | 3065/4000 [15:46<04:21,  3.58it/s]

train:  87%|████████▋ | 3065/3515 [15:44<02:05,  3.58it/s]
overall:  77%|███████▋  | 3066/4000 [15:46<03:32,  4.39it/s]

train:  87%|████████▋ | 3066/3515 [15:44<01:42,  4.39it/s]
overall:  77%|███████▋  | 3067/4000 [15:47<03:46,  4.12it/s]

train:  87%|████████▋ | 3067/3515 [15:44<01:48,  4.12it/s]
overall:  77%|███████▋  | 3068/4000 [15:47<03:52,  4.02it/s]

train:  87%|████████▋ | 3068/3515 [15:44<01:51,  4.02it/s]
overall:  77%|███████▋  | 3069/4000 [15:47<03:59,  3.88it/s]

train:  87%|████████▋ | 3069/3515 [15:45<01:54,  3.88it/s]
overall:  77%|███████▋  | 3070/4000 [15:47<04:01,  3.85it/s]

train:  87%|████████▋ | 3070/3515 [15:45<01:55,  3.85it/s]
overall:  77%|███████▋  | 3071/4000 [15:48<04:02,  3.82it/s]

train:  87%|████████▋ | 3071/3515 [15:45<01:56,  3.82it/s]
overall:  77%|███████▋  | 3072/4000 [15:48<04:05,  3.78it/s]

train:  87%|████████▋ | 3072/3515 [15:45<01:57,  3.78it/s]
overall:  77%|███████▋  | 3073/4000 [15:48<04:02,  3.82it/s]

train:  87%|████████▋ | 3073/3515 [15:46<01:55,  3.82it/s]
overall:  77%|███████▋  | 3074/4000 [15:49<04:04,  3.78it/s]

train:  87%|████████▋ | 3074/3515 [15:46<01:56,  3.78it/s]
overall:  77%|███████▋  | 3075/4000 [15:49<04:00,  3.85it/s]

train:  87%|████████▋ | 3075/3515 [15:46<01:54,  3.85it/s]
overall:  77%|███████▋  | 3076/4000 [15:49<04:04,  3.77it/s]

train:  88%|████████▊ | 3076/3515 [15:46<01:56,  3.77it/s]
overall:  77%|███████▋  | 3077/4000 [15:49<03:57,  3.88it/s]

train:  88%|████████▊ | 3077/3515 [15:47<01:52,  3.88it/s]
overall:  77%|███████▋  | 3079/4000 [15:50<03:23,  4.53it/s]

train:  88%|████████▊ | 3079/3515 [15:47<01:36,  4.53it/s]train loss: 0.5524406433105469
train per: 0.16012939749292357

overall:  77%|███████▋  | 3080/4000 [15:50<03:36,  4.25it/s]

train:  88%|████████▊ | 3080/3515 [15:47<01:42,  4.25it/s]
overall:  77%|███████▋  | 3081/4000 [15:50<03:39,  4.18it/s]

train:  88%|████████▊ | 3081/3515 [15:48<01:43,  4.19it/s]
overall:  77%|███████▋  | 3082/4000 [15:50<03:31,  4.33it/s]

train:  88%|████████▊ | 3082/3515 [15:48<01:39,  4.33it/s]
overall:  77%|███████▋  | 3083/4000 [15:51<03:41,  4.14it/s]

train:  88%|████████▊ | 3083/3515 [15:48<01:44,  4.14it/s]
overall:  77%|███████▋  | 3084/4000 [15:51<03:48,  4.01it/s]

train:  88%|████████▊ | 3084/3515 [15:48<01:47,  4.01it/s]
overall:  77%|███████▋  | 3085/4000 [15:51<04:08,  3.68it/s]

train:  88%|████████▊ | 3085/3515 [15:49<01:56,  3.68it/s]
overall:  77%|███████▋  | 3086/4000 [15:52<04:09,  3.66it/s]

train:  88%|████████▊ | 3086/3515 [15:49<01:57,  3.66it/s]
overall:  77%|███████▋  | 3087/4000 [15:52<04:05,  3.72it/s]

train:  88%|████████▊ | 3087/3515 [15:49<01:55,  3.72it/s]
overall:  77%|███████▋  | 3088/4000 [15:52<04:08,  3.67it/s]

train:  88%|████████▊ | 3088/3515 [15:49<01:56,  3.67it/s]
overall:  77%|███████▋  | 3089/4000 [15:52<04:09,  3.65it/s]

train:  88%|████████▊ | 3089/3515 [15:50<01:56,  3.65it/s]
overall:  77%|███████▋  | 3090/4000 [15:53<04:00,  3.79it/s]

train:  88%|████████▊ | 3090/3515 [15:50<01:52,  3.79it/s]
overall:  77%|███████▋  | 3091/4000 [15:53<04:00,  3.78it/s]

train:  88%|████████▊ | 3091/3515 [15:50<01:52,  3.78it/s]
overall:  77%|███████▋  | 3092/4000 [15:53<04:06,  3.68it/s]

train:  88%|████████▊ | 3092/3515 [15:51<01:55,  3.68it/s]
overall:  77%|███████▋  | 3093/4000 [15:53<04:06,  3.68it/s]

train:  88%|████████▊ | 3093/3515 [15:51<01:54,  3.68it/s]
overall:  77%|███████▋  | 3094/4000 [15:54<04:06,  3.68it/s]

train:  88%|████████▊ | 3094/3515 [15:51<01:54,  3.68it/s]
overall:  77%|███████▋  | 3095/4000 [15:54<04:02,  3.73it/s]

train:  88%|████████▊ | 3095/3515 [15:51<01:52,  3.73it/s]
overall:  77%|███████▋  | 3096/4000 [15:54<03:58,  3.79it/s]

train:  88%|████████▊ | 3096/3515 [15:52<01:50,  3.79it/s]
overall:  77%|███████▋  | 3097/4000 [15:54<03:58,  3.79it/s]

train:  88%|████████▊ | 3097/3515 [15:52<01:50,  3.79it/s]
overall:  77%|███████▋  | 3098/4000 [15:55<03:58,  3.78it/s]

train:  88%|████████▊ | 3098/3515 [15:52<01:50,  3.78it/s]
overall:  77%|███████▋  | 3099/4000 [15:55<03:58,  3.78it/s]

train:  88%|████████▊ | 3099/3515 [15:52<01:50,  3.78it/s]train loss: 0.5561323761940002
train per: 0.16638946801283702

overall:  78%|███████▊  | 3100/4000 [15:55<04:03,  3.69it/s]

train:  88%|████████▊ | 3100/3515 [15:53<01:52,  3.69it/s]
overall:  78%|███████▊  | 3101/4000 [15:56<04:04,  3.68it/s]

train:  88%|████████▊ | 3101/3515 [15:53<01:52,  3.68it/s]
overall:  78%|███████▊  | 3102/4000 [15:56<03:59,  3.75it/s]

train:  88%|████████▊ | 3102/3515 [15:53<01:50,  3.75it/s]
overall:  78%|███████▊  | 3103/4000 [15:56<03:57,  3.78it/s]

train:  88%|████████▊ | 3103/3515 [15:53<01:48,  3.78it/s]
overall:  78%|███████▊  | 3104/4000 [15:56<03:42,  4.03it/s]

train:  88%|████████▊ | 3104/3515 [15:54<01:41,  4.03it/s]
overall:  78%|███████▊  | 3105/4000 [15:57<03:56,  3.79it/s]

train:  88%|████████▊ | 3105/3515 [15:54<01:48,  3.79it/s]
overall:  78%|███████▊  | 3106/4000 [15:57<03:56,  3.78it/s]

train:  88%|████████▊ | 3106/3515 [15:54<01:48,  3.78it/s]
overall:  78%|███████▊  | 3107/4000 [15:57<03:53,  3.83it/s]

train:  88%|████████▊ | 3107/3515 [15:54<01:46,  3.83it/s]
overall:  78%|███████▊  | 3108/4000 [15:57<03:59,  3.73it/s]

train:  88%|████████▊ | 3108/3515 [15:55<01:49,  3.73it/s]
overall:  78%|███████▊  | 3109/4000 [15:58<04:11,  3.55it/s]

train:  88%|████████▊ | 3109/3515 [15:55<01:54,  3.55it/s]
overall:  78%|███████▊  | 3110/4000 [15:58<04:12,  3.52it/s]

train:  88%|████████▊ | 3110/3515 [15:55<01:54,  3.52it/s]
overall:  78%|███████▊  | 3111/4000 [15:58<04:07,  3.59it/s]

train:  89%|████████▊ | 3111/3515 [15:56<01:52,  3.59it/s]
overall:  78%|███████▊  | 3112/4000 [15:59<04:11,  3.53it/s]

train:  89%|████████▊ | 3112/3515 [15:56<01:54,  3.53it/s]
overall:  78%|███████▊  | 3113/4000 [15:59<04:05,  3.62it/s]

train:  89%|████████▊ | 3113/3515 [15:56<01:51,  3.62it/s]
overall:  78%|███████▊  | 3114/4000 [15:59<03:58,  3.71it/s]

train:  89%|████████▊ | 3114/3515 [15:56<01:48,  3.71it/s]
overall:  78%|███████▊  | 3115/4000 [15:59<03:59,  3.70it/s]

train:  89%|████████▊ | 3115/3515 [15:57<01:48,  3.70it/s]
overall:  78%|███████▊  | 3116/4000 [16:00<03:40,  4.01it/s]

train:  89%|████████▊ | 3116/3515 [15:57<01:39,  4.01it/s]
overall:  78%|███████▊  | 3117/4000 [16:00<03:40,  4.00it/s]

train:  89%|████████▊ | 3117/3515 [15:57<01:39,  4.00it/s]
overall:  78%|███████▊  | 3118/4000 [16:00<03:48,  3.86it/s]

train:  89%|████████▊ | 3118/3515 [15:57<01:42,  3.86it/s]
overall:  78%|███████▊  | 3119/4000 [16:00<03:52,  3.78it/s]

train:  89%|████████▊ | 3119/3515 [15:58<01:44,  3.78it/s]train loss: 0.560653805732727
train per: 0.1623278954066875

overall:  78%|███████▊  | 3120/4000 [16:01<03:54,  3.75it/s]

train:  89%|████████▉ | 3120/3515 [15:58<01:45,  3.75it/s]
overall:  78%|███████▊  | 3121/4000 [16:01<03:45,  3.90it/s]

train:  89%|████████▉ | 3121/3515 [15:58<01:41,  3.90it/s]
overall:  78%|███████▊  | 3122/4000 [16:01<03:49,  3.82it/s]

train:  89%|████████▉ | 3122/3515 [15:58<01:42,  3.82it/s]
overall:  78%|███████▊  | 3123/4000 [16:01<03:49,  3.82it/s]

train:  89%|████████▉ | 3123/3515 [15:59<01:42,  3.82it/s]
overall:  78%|███████▊  | 3124/4000 [16:02<03:34,  4.08it/s]

train:  89%|████████▉ | 3124/3515 [15:59<01:35,  4.08it/s]
overall:  78%|███████▊  | 3125/4000 [16:02<03:38,  4.01it/s]

train:  89%|████████▉ | 3125/3515 [15:59<01:37,  4.01it/s]
overall:  78%|███████▊  | 3126/4000 [16:02<03:42,  3.93it/s]

train:  89%|████████▉ | 3126/3515 [15:59<01:38,  3.93it/s]
overall:  78%|███████▊  | 3127/4000 [16:02<03:08,  4.62it/s]

train:  89%|████████▉ | 3127/3515 [16:00<01:23,  4.62it/s]
overall:  78%|███████▊  | 3128/4000 [16:02<03:19,  4.37it/s]

train:  89%|████████▉ | 3128/3515 [16:00<01:28,  4.37it/s]
overall:  78%|███████▊  | 3129/4000 [16:03<04:11,  3.47it/s]

train:  89%|████████▉ | 3129/3515 [16:00<01:51,  3.47it/s]
overall:  78%|███████▊  | 3130/4000 [16:03<03:57,  3.67it/s]

train:  89%|████████▉ | 3130/3515 [16:01<01:44,  3.67it/s]
overall:  78%|███████▊  | 3131/4000 [16:03<03:56,  3.68it/s]

train:  89%|████████▉ | 3131/3515 [16:01<01:44,  3.68it/s]
overall:  78%|███████▊  | 3132/4000 [16:04<03:53,  3.72it/s]

train:  89%|████████▉ | 3132/3515 [16:01<01:42,  3.72it/s]
overall:  78%|███████▊  | 3133/4000 [16:04<04:01,  3.59it/s]

train:  89%|████████▉ | 3133/3515 [16:01<01:46,  3.59it/s]
overall:  78%|███████▊  | 3134/4000 [16:04<03:52,  3.72it/s]

train:  89%|████████▉ | 3134/3515 [16:02<01:42,  3.72it/s]
overall:  78%|███████▊  | 3135/4000 [16:04<03:49,  3.76it/s]

train:  89%|████████▉ | 3135/3515 [16:02<01:40,  3.76it/s]
overall:  78%|███████▊  | 3136/4000 [16:05<03:57,  3.63it/s]

train:  89%|████████▉ | 3136/3515 [16:02<01:44,  3.63it/s]
overall:  78%|███████▊  | 3137/4000 [16:05<03:56,  3.64it/s]

train:  89%|████████▉ | 3137/3515 [16:02<01:43,  3.64it/s]
overall:  78%|███████▊  | 3138/4000 [16:05<03:44,  3.84it/s]

train:  89%|████████▉ | 3138/3515 [16:03<01:38,  3.84it/s]
overall:  78%|███████▊  | 3139/4000 [16:06<03:46,  3.79it/s]

train:  89%|████████▉ | 3139/3515 [16:03<01:39,  3.79it/s]train loss: 0.5369281768798828
train per: 0.15916213095827147

overall:  78%|███████▊  | 3140/4000 [16:06<03:56,  3.63it/s]

train:  89%|████████▉ | 3140/3515 [16:03<01:43,  3.63it/s]
overall:  79%|███████▊  | 3141/4000 [16:06<04:04,  3.52it/s]

train:  89%|████████▉ | 3141/3515 [16:04<01:46,  3.52it/s]
overall:  79%|███████▊  | 3142/4000 [16:06<04:01,  3.55it/s]

train:  89%|████████▉ | 3142/3515 [16:04<01:45,  3.55it/s]
overall:  79%|███████▊  | 3143/4000 [16:07<03:24,  4.19it/s]

train:  89%|████████▉ | 3143/3515 [16:04<01:28,  4.19it/s]
overall:  79%|███████▊  | 3144/4000 [16:07<03:34,  4.00it/s]

train:  89%|████████▉ | 3144/3515 [16:04<01:32,  4.00it/s]
overall:  79%|███████▊  | 3145/4000 [16:07<03:39,  3.90it/s]

train:  89%|████████▉ | 3145/3515 [16:05<01:34,  3.90it/s]
overall:  79%|███████▊  | 3146/4000 [16:07<03:11,  4.46it/s]

train:  90%|████████▉ | 3146/3515 [16:05<01:22,  4.46it/s]
overall:  79%|███████▊  | 3147/4000 [16:08<03:23,  4.19it/s]

train:  90%|████████▉ | 3147/3515 [16:05<01:27,  4.19it/s]
overall:  79%|███████▊  | 3148/4000 [16:08<03:29,  4.07it/s]

train:  90%|████████▉ | 3148/3515 [16:05<01:30,  4.07it/s]
overall:  79%|███████▊  | 3149/4000 [16:08<03:34,  3.97it/s]

train:  90%|████████▉ | 3149/3515 [16:05<01:32,  3.97it/s]
overall:  79%|███████▉  | 3150/4000 [16:08<03:40,  3.85it/s]

train:  90%|████████▉ | 3150/3515 [16:06<01:34,  3.85it/s]
overall:  79%|███████▉  | 3151/4000 [16:09<03:42,  3.81it/s]

train:  90%|████████▉ | 3151/3515 [16:06<01:35,  3.81it/s]
overall:  79%|███████▉  | 3152/4000 [16:09<03:35,  3.93it/s]

train:  90%|████████▉ | 3152/3515 [16:06<01:32,  3.93it/s]
overall:  79%|███████▉  | 3153/4000 [16:09<04:22,  3.23it/s]

train:  90%|████████▉ | 3153/3515 [16:07<01:52,  3.23it/s]
overall:  79%|███████▉  | 3154/4000 [16:10<04:13,  3.34it/s]

train:  90%|████████▉ | 3154/3515 [16:07<01:47,  3.34it/s]
overall:  79%|███████▉  | 3155/4000 [16:10<03:56,  3.57it/s]

train:  90%|████████▉ | 3155/3515 [16:07<01:40,  3.57it/s]
overall:  79%|███████▉  | 3156/4000 [16:10<03:52,  3.63it/s]

train:  90%|████████▉ | 3156/3515 [16:07<01:38,  3.63it/s]
overall:  79%|███████▉  | 3157/4000 [16:10<04:01,  3.49it/s]

train:  90%|████████▉ | 3157/3515 [16:08<01:42,  3.49it/s]
overall:  79%|███████▉  | 3158/4000 [16:11<03:55,  3.58it/s]

train:  90%|████████▉ | 3158/3515 [16:08<01:39,  3.58it/s]
overall:  79%|███████▉  | 3159/4000 [16:11<03:53,  3.60it/s]

train:  90%|████████▉ | 3159/3515 [16:08<01:38,  3.60it/s]train loss: 0.5449303984642029
train per: 0.163292552675017

overall:  79%|███████▉  | 3160/4000 [16:11<03:34,  3.91it/s]

train:  90%|████████▉ | 3160/3515 [16:09<01:30,  3.91it/s]
overall:  79%|███████▉  | 3161/4000 [16:11<03:30,  3.99it/s]

train:  90%|████████▉ | 3161/3515 [16:09<01:28,  3.99it/s]
overall:  79%|███████▉  | 3162/4000 [16:12<03:31,  3.96it/s]

train:  90%|████████▉ | 3162/3515 [16:09<01:29,  3.96it/s]
overall:  79%|███████▉  | 3163/4000 [16:12<03:32,  3.94it/s]

train:  90%|████████▉ | 3163/3515 [16:09<01:29,  3.95it/s]
overall:  79%|███████▉  | 3164/4000 [16:12<03:39,  3.81it/s]

train:  90%|█████████ | 3164/3515 [16:10<01:32,  3.81it/s]
overall:  79%|███████▉  | 3165/4000 [16:12<03:15,  4.26it/s]

train:  90%|█████████ | 3165/3515 [16:10<01:22,  4.26it/s]
overall:  79%|███████▉  | 3166/4000 [16:13<03:25,  4.07it/s]

train:  90%|█████████ | 3166/3515 [16:10<01:25,  4.07it/s]
overall:  79%|███████▉  | 3167/4000 [16:13<03:26,  4.04it/s]

train:  90%|█████████ | 3167/3515 [16:10<01:26,  4.04it/s]
overall:  79%|███████▉  | 3168/4000 [16:13<03:28,  4.00it/s]

train:  90%|█████████ | 3168/3515 [16:10<01:26,  4.00it/s]
overall:  79%|███████▉  | 3169/4000 [16:13<03:32,  3.91it/s]

train:  90%|█████████ | 3169/3515 [16:11<01:28,  3.91it/s]
overall:  79%|███████▉  | 3170/4000 [16:14<03:34,  3.86it/s]

train:  90%|█████████ | 3170/3515 [16:11<01:29,  3.86it/s]
overall:  79%|███████▉  | 3171/4000 [16:14<03:35,  3.85it/s]

train:  90%|█████████ | 3171/3515 [16:11<01:29,  3.85it/s]
overall:  79%|███████▉  | 3172/4000 [16:14<03:36,  3.83it/s]

train:  90%|█████████ | 3172/3515 [16:12<01:29,  3.83it/s]
overall:  79%|███████▉  | 3173/4000 [16:14<03:39,  3.77it/s]

train:  90%|█████████ | 3173/3515 [16:12<01:30,  3.77it/s]
overall:  79%|███████▉  | 3174/4000 [16:15<03:40,  3.74it/s]

train:  90%|█████████ | 3174/3515 [16:12<01:31,  3.74it/s]
overall:  79%|███████▉  | 3175/4000 [16:15<03:35,  3.82it/s]

train:  90%|█████████ | 3175/3515 [16:12<01:28,  3.82it/s]
overall:  79%|███████▉  | 3176/4000 [16:15<03:37,  3.79it/s]

train:  90%|█████████ | 3176/3515 [16:13<01:29,  3.79it/s]
overall:  79%|███████▉  | 3177/4000 [16:16<04:14,  3.24it/s]

train:  90%|█████████ | 3177/3515 [16:13<01:44,  3.24it/s]
overall:  79%|███████▉  | 3178/4000 [16:16<04:01,  3.41it/s]

train:  90%|█████████ | 3178/3515 [16:13<01:38,  3.41it/s]
overall:  79%|███████▉  | 3179/4000 [16:16<03:54,  3.49it/s]

train:  90%|█████████ | 3179/3515 [16:14<01:36,  3.49it/s]train loss: 0.5406509637832642
train per: 0.1592283195643144

overall:  80%|███████▉  | 3180/4000 [16:16<03:59,  3.42it/s]

train:  90%|█████████ | 3180/3515 [16:14<01:37,  3.42it/s]
overall:  80%|███████▉  | 3181/4000 [16:17<04:07,  3.31it/s]

train:  90%|█████████ | 3181/3515 [16:14<01:40,  3.31it/s]
overall:  80%|███████▉  | 3182/4000 [16:17<03:54,  3.49it/s]

train:  91%|█████████ | 3182/3515 [16:14<01:35,  3.49it/s]
overall:  80%|███████▉  | 3183/4000 [16:17<03:44,  3.64it/s]

train:  91%|█████████ | 3183/3515 [16:15<01:31,  3.64it/s]
overall:  80%|███████▉  | 3185/4000 [16:18<03:07,  4.36it/s]

train:  91%|█████████ | 3185/3515 [16:15<01:15,  4.36it/s]
overall:  80%|███████▉  | 3186/4000 [16:18<03:15,  4.16it/s]

train:  91%|█████████ | 3186/3515 [16:15<01:18,  4.17it/s]
overall:  80%|███████▉  | 3187/4000 [16:18<02:49,  4.80it/s]

train:  91%|█████████ | 3187/3515 [16:15<01:08,  4.80it/s]
overall:  80%|███████▉  | 3188/4000 [16:18<03:09,  4.30it/s]

train:  91%|█████████ | 3188/3515 [16:16<01:16,  4.30it/s]
overall:  80%|███████▉  | 3189/4000 [16:19<03:19,  4.06it/s]

train:  91%|█████████ | 3189/3515 [16:16<01:20,  4.06it/s]
overall:  80%|███████▉  | 3190/4000 [16:19<03:20,  4.04it/s]

train:  91%|█████████ | 3190/3515 [16:16<01:20,  4.04it/s]
overall:  80%|███████▉  | 3191/4000 [16:19<03:27,  3.90it/s]

train:  91%|█████████ | 3191/3515 [16:17<01:23,  3.90it/s]
overall:  80%|███████▉  | 3192/4000 [16:19<03:33,  3.79it/s]

train:  91%|█████████ | 3192/3515 [16:17<01:25,  3.79it/s]
overall:  80%|███████▉  | 3193/4000 [16:20<03:35,  3.74it/s]

train:  91%|█████████ | 3193/3515 [16:17<01:26,  3.74it/s]
overall:  80%|███████▉  | 3194/4000 [16:20<03:32,  3.79it/s]

train:  91%|█████████ | 3194/3515 [16:17<01:24,  3.79it/s]
overall:  80%|███████▉  | 3195/4000 [16:20<03:34,  3.75it/s]

train:  91%|█████████ | 3195/3515 [16:18<01:25,  3.75it/s]
overall:  80%|███████▉  | 3196/4000 [16:21<03:34,  3.74it/s]

train:  91%|█████████ | 3196/3515 [16:18<01:25,  3.74it/s]
overall:  80%|███████▉  | 3197/4000 [16:21<03:32,  3.79it/s]

train:  91%|█████████ | 3197/3515 [16:18<01:23,  3.79it/s]
overall:  80%|███████▉  | 3198/4000 [16:21<03:32,  3.77it/s]

train:  91%|█████████ | 3198/3515 [16:18<01:24,  3.77it/s]
overall:  80%|███████▉  | 3199/4000 [16:21<03:32,  3.77it/s]

train:  91%|█████████ | 3199/3515 [16:19<01:23,  3.77it/s]train loss: 0.5660659074783325
train per: 0.16601129259931438

overall:  80%|████████  | 3200/4000 [16:22<03:34,  3.74it/s]

train:  91%|█████████ | 3200/3515 [16:19<01:24,  3.74it/s]
overall:  80%|████████  | 3201/4000 [16:22<04:07,  3.22it/s]

train:  91%|█████████ | 3201/3515 [16:19<01:37,  3.22it/s]
overall:  80%|████████  | 3202/4000 [16:22<03:58,  3.35it/s]

train:  91%|█████████ | 3202/3515 [16:20<01:33,  3.35it/s]
overall:  80%|████████  | 3203/4000 [16:22<03:11,  4.16it/s]

train:  91%|█████████ | 3203/3515 [16:20<01:15,  4.16it/s]
overall:  80%|████████  | 3204/4000 [16:23<03:25,  3.87it/s]

train:  91%|█████████ | 3204/3515 [16:20<01:20,  3.87it/s]
overall:  80%|████████  | 3205/4000 [16:23<03:42,  3.57it/s]

train:  91%|█████████ | 3205/3515 [16:20<01:26,  3.57it/s]
overall:  80%|████████  | 3206/4000 [16:23<03:45,  3.52it/s]

train:  91%|█████████ | 3206/3515 [16:21<01:27,  3.52it/s]
overall:  80%|████████  | 3207/4000 [16:24<03:40,  3.60it/s]

train:  91%|█████████ | 3207/3515 [16:21<01:25,  3.60it/s]
overall:  80%|████████  | 3208/4000 [16:24<03:29,  3.78it/s]

train:  91%|█████████▏| 3208/3515 [16:21<01:21,  3.78it/s]
overall:  80%|████████  | 3209/4000 [16:24<03:29,  3.77it/s]

train:  91%|█████████▏| 3209/3515 [16:21<01:21,  3.77it/s]
overall:  80%|████████  | 3210/4000 [16:24<03:41,  3.56it/s]

train:  91%|█████████▏| 3210/3515 [16:22<01:25,  3.56it/s]
overall:  80%|████████  | 3211/4000 [16:25<03:26,  3.82it/s]

train:  91%|█████████▏| 3211/3515 [16:22<01:19,  3.82it/s]
overall:  80%|████████  | 3212/4000 [16:25<03:28,  3.79it/s]

train:  91%|█████████▏| 3212/3515 [16:22<01:20,  3.79it/s]
overall:  80%|████████  | 3214/4000 [16:25<02:49,  4.63it/s]

train:  91%|█████████▏| 3214/3515 [16:23<01:04,  4.63it/s]
overall:  80%|████████  | 3215/4000 [16:25<03:01,  4.33it/s]

train:  91%|█████████▏| 3215/3515 [16:23<01:09,  4.33it/s]
overall:  80%|████████  | 3216/4000 [16:26<02:45,  4.75it/s]

train:  91%|█████████▏| 3216/3515 [16:23<01:02,  4.75it/s]
overall:  80%|████████  | 3217/4000 [16:26<02:28,  5.29it/s]

train:  92%|█████████▏| 3217/3515 [16:23<00:56,  5.29it/s]
overall:  80%|████████  | 3218/4000 [16:26<02:46,  4.70it/s]

train:  92%|█████████▏| 3218/3515 [16:23<01:03,  4.70it/s]
overall:  80%|████████  | 3219/4000 [16:26<02:58,  4.37it/s]

train:  92%|█████████▏| 3219/3515 [16:24<01:07,  4.37it/s]train loss: 0.5893038511276245
train per: 0.16191396810053166

overall:  80%|████████  | 3220/4000 [16:27<03:07,  4.17it/s]

train:  92%|█████████▏| 3220/3515 [16:24<01:10,  4.17it/s]
overall:  81%|████████  | 3221/4000 [16:27<03:08,  4.14it/s]

train:  92%|█████████▏| 3221/3515 [16:24<01:10,  4.14it/s]
overall:  81%|████████  | 3222/4000 [16:27<03:10,  4.09it/s]

train:  92%|█████████▏| 3222/3515 [16:24<01:11,  4.09it/s]
overall:  81%|████████  | 3223/4000 [16:27<03:15,  3.97it/s]

train:  92%|█████████▏| 3223/3515 [16:25<01:13,  3.97it/s]
overall:  81%|████████  | 3224/4000 [16:28<03:13,  4.01it/s]

train:  92%|█████████▏| 3224/3515 [16:25<01:12,  4.01it/s]
overall:  81%|████████  | 3225/4000 [16:28<03:41,  3.49it/s]

train:  92%|█████████▏| 3225/3515 [16:25<01:23,  3.49it/s]
overall:  81%|████████  | 3226/4000 [16:28<03:39,  3.52it/s]

train:  92%|█████████▏| 3226/3515 [16:26<01:22,  3.52it/s]
overall:  81%|████████  | 3227/4000 [16:28<03:33,  3.62it/s]

train:  92%|█████████▏| 3227/3515 [16:26<01:19,  3.62it/s]
overall:  81%|████████  | 3228/4000 [16:29<03:35,  3.59it/s]

train:  92%|█████████▏| 3228/3515 [16:26<01:19,  3.59it/s]
overall:  81%|████████  | 3229/4000 [16:29<03:45,  3.42it/s]

train:  92%|█████████▏| 3229/3515 [16:26<01:23,  3.42it/s]
overall:  81%|████████  | 3230/4000 [16:29<03:38,  3.52it/s]

train:  92%|█████████▏| 3230/3515 [16:27<01:20,  3.52it/s]
overall:  81%|████████  | 3231/4000 [16:30<03:36,  3.56it/s]

train:  92%|█████████▏| 3231/3515 [16:27<01:19,  3.56it/s]
overall:  81%|████████  | 3232/4000 [16:30<03:32,  3.61it/s]

train:  92%|█████████▏| 3232/3515 [16:27<01:18,  3.61it/s]
overall:  81%|████████  | 3233/4000 [16:30<03:08,  4.06it/s]

train:  92%|█████████▏| 3233/3515 [16:27<01:09,  4.06it/s]
overall:  81%|████████  | 3234/4000 [16:30<02:43,  4.70it/s]

train:  92%|█████████▏| 3234/3515 [16:28<00:59,  4.70it/s]
overall:  81%|████████  | 3235/4000 [16:30<02:52,  4.44it/s]

train:  92%|█████████▏| 3235/3515 [16:28<01:03,  4.44it/s]
overall:  81%|████████  | 3236/4000 [16:31<02:59,  4.26it/s]

train:  92%|█████████▏| 3236/3515 [16:28<01:05,  4.26it/s]
overall:  81%|████████  | 3237/4000 [16:31<03:04,  4.13it/s]

train:  92%|█████████▏| 3237/3515 [16:28<01:07,  4.13it/s]
overall:  81%|████████  | 3239/4000 [16:31<02:34,  4.92it/s]

train:  92%|█████████▏| 3239/3515 [16:29<00:56,  4.92it/s]train loss: 0.52202308177948
train per: 0.15455537370430988

overall:  81%|████████  | 3240/4000 [16:32<02:42,  4.66it/s]

train:  92%|█████████▏| 3240/3515 [16:29<00:58,  4.66it/s]
overall:  81%|████████  | 3242/4000 [16:32<02:34,  4.91it/s]

train:  92%|█████████▏| 3242/3515 [16:29<00:55,  4.91it/s]
overall:  81%|████████  | 3243/4000 [16:32<02:40,  4.71it/s]

train:  92%|█████████▏| 3243/3515 [16:30<00:57,  4.71it/s]
overall:  81%|████████  | 3244/4000 [16:32<02:22,  5.29it/s]

train:  92%|█████████▏| 3244/3515 [16:30<00:51,  5.29it/s]
overall:  81%|████████  | 3245/4000 [16:33<02:38,  4.76it/s]

train:  92%|█████████▏| 3245/3515 [16:30<00:56,  4.76it/s]
overall:  81%|████████  | 3246/4000 [16:33<02:49,  4.44it/s]

train:  92%|█████████▏| 3246/3515 [16:30<01:00,  4.44it/s]
overall:  81%|████████  | 3247/4000 [16:33<02:56,  4.27it/s]

train:  92%|█████████▏| 3247/3515 [16:30<01:02,  4.27it/s]
overall:  81%|████████  | 3248/4000 [16:33<03:00,  4.18it/s]

train:  92%|█████████▏| 3248/3515 [16:31<01:03,  4.18it/s]
overall:  81%|████████  | 3249/4000 [16:34<03:36,  3.47it/s]

train:  92%|█████████▏| 3249/3515 [16:31<01:16,  3.47it/s]
overall:  81%|████████▏ | 3250/4000 [16:34<03:27,  3.61it/s]

train:  92%|█████████▏| 3250/3515 [16:31<01:13,  3.61it/s]
overall:  81%|████████▏ | 3251/4000 [16:34<03:24,  3.66it/s]

train:  92%|█████████▏| 3251/3515 [16:32<01:12,  3.66it/s]
overall:  81%|████████▏ | 3252/4000 [16:34<03:19,  3.74it/s]

train:  93%|█████████▎| 3252/3515 [16:32<01:10,  3.74it/s]
overall:  81%|████████▏ | 3253/4000 [16:35<03:22,  3.69it/s]

train:  93%|█████████▎| 3253/3515 [16:32<01:10,  3.69it/s]
overall:  81%|████████▏ | 3254/4000 [16:35<03:24,  3.64it/s]

train:  93%|█████████▎| 3254/3515 [16:32<01:11,  3.64it/s]
overall:  81%|████████▏ | 3255/4000 [16:35<03:24,  3.64it/s]

train:  93%|█████████▎| 3255/3515 [16:33<01:11,  3.64it/s]
overall:  81%|████████▏ | 3256/4000 [16:36<03:25,  3.62it/s]

train:  93%|█████████▎| 3256/3515 [16:33<01:11,  3.62it/s]
overall:  81%|████████▏ | 3257/4000 [16:36<03:18,  3.74it/s]

train:  93%|█████████▎| 3257/3515 [16:33<01:09,  3.74it/s]
overall:  81%|████████▏ | 3258/4000 [16:36<03:04,  4.02it/s]

train:  93%|█████████▎| 3258/3515 [16:33<01:03,  4.02it/s]
overall:  81%|████████▏ | 3259/4000 [16:36<03:08,  3.93it/s]

train:  93%|█████████▎| 3259/3515 [16:34<01:05,  3.93it/s]train loss: 0.603870689868927
train per: 0.16567462965434404

overall:  82%|████████▏ | 3260/4000 [16:37<03:12,  3.84it/s]

train:  93%|█████████▎| 3260/3515 [16:34<01:06,  3.84it/s]
overall:  82%|████████▏ | 3261/4000 [16:37<03:17,  3.74it/s]

train:  93%|█████████▎| 3261/3515 [16:34<01:07,  3.74it/s]
overall:  82%|████████▏ | 3262/4000 [16:37<03:12,  3.84it/s]

train:  93%|█████████▎| 3262/3515 [16:34<01:05,  3.84it/s]
overall:  82%|████████▏ | 3263/4000 [16:37<03:13,  3.81it/s]

train:  93%|█████████▎| 3263/3515 [16:35<01:06,  3.81it/s]
overall:  82%|████████▏ | 3264/4000 [16:38<03:13,  3.81it/s]

train:  93%|█████████▎| 3264/3515 [16:35<01:05,  3.81it/s]
overall:  82%|████████▏ | 3265/4000 [16:38<03:15,  3.76it/s]

train:  93%|█████████▎| 3265/3515 [16:35<01:06,  3.76it/s]
overall:  82%|████████▏ | 3266/4000 [16:38<03:05,  3.96it/s]

train:  93%|█████████▎| 3266/3515 [16:36<01:02,  3.96it/s]
overall:  82%|████████▏ | 3267/4000 [16:38<03:10,  3.85it/s]

train:  93%|█████████▎| 3267/3515 [16:36<01:04,  3.85it/s]
overall:  82%|████████▏ | 3268/4000 [16:39<03:10,  3.85it/s]

train:  93%|█████████▎| 3268/3515 [16:36<01:04,  3.85it/s]
overall:  82%|████████▏ | 3269/4000 [16:39<03:04,  3.97it/s]

train:  93%|█████████▎| 3269/3515 [16:36<01:01,  3.97it/s]
overall:  82%|████████▏ | 3270/4000 [16:39<03:03,  3.98it/s]

train:  93%|█████████▎| 3270/3515 [16:37<01:01,  3.98it/s]
overall:  82%|████████▏ | 3271/4000 [16:39<03:07,  3.88it/s]

train:  93%|█████████▎| 3271/3515 [16:37<01:02,  3.88it/s]
overall:  82%|████████▏ | 3272/4000 [16:40<02:33,  4.73it/s]

train:  93%|█████████▎| 3272/3515 [16:37<00:51,  4.73it/s]
overall:  82%|████████▏ | 3273/4000 [16:40<02:28,  4.88it/s]

train:  93%|█████████▎| 3273/3515 [16:37<00:49,  4.88it/s]
overall:  82%|████████▏ | 3274/4000 [16:40<02:42,  4.46it/s]

train:  93%|█████████▎| 3274/3515 [16:37<00:54,  4.46it/s]
overall:  82%|████████▏ | 3275/4000 [16:40<02:55,  4.14it/s]

train:  93%|█████████▎| 3275/3515 [16:38<00:57,  4.14it/s]
overall:  82%|████████▏ | 3276/4000 [16:40<02:31,  4.76it/s]

train:  93%|█████████▎| 3276/3515 [16:38<00:50,  4.76it/s]
overall:  82%|████████▏ | 3277/4000 [16:41<02:55,  4.11it/s]

train:  93%|█████████▎| 3277/3515 [16:38<00:57,  4.11it/s]
overall:  82%|████████▏ | 3278/4000 [16:41<02:27,  4.90it/s]

train:  93%|█████████▎| 3278/3515 [16:38<00:48,  4.90it/s]
overall:  82%|████████▏ | 3279/4000 [16:41<02:42,  4.44it/s]

train:  93%|█████████▎| 3279/3515 [16:38<00:53,  4.44it/s]train loss: 0.559743344783783
train per: 0.16503753307771238

overall:  82%|████████▏ | 3280/4000 [16:41<02:59,  4.00it/s]

train:  93%|█████████▎| 3280/3515 [16:39<00:58,  4.00it/s]
overall:  82%|████████▏ | 3281/4000 [16:42<03:04,  3.90it/s]

train:  93%|█████████▎| 3281/3515 [16:39<01:00,  3.90it/s]
overall:  82%|████████▏ | 3282/4000 [16:42<03:24,  3.51it/s]

train:  93%|█████████▎| 3282/3515 [16:39<01:06,  3.51it/s]
overall:  82%|████████▏ | 3283/4000 [16:42<03:15,  3.66it/s]

train:  93%|█████████▎| 3283/3515 [16:40<01:03,  3.66it/s]
overall:  82%|████████▏ | 3284/4000 [16:43<03:14,  3.68it/s]

train:  93%|█████████▎| 3284/3515 [16:40<01:02,  3.68it/s]
overall:  82%|████████▏ | 3285/4000 [16:43<03:15,  3.66it/s]

train:  93%|█████████▎| 3285/3515 [16:40<01:02,  3.66it/s]
overall:  82%|████████▏ | 3286/4000 [16:43<03:14,  3.67it/s]

train:  93%|█████████▎| 3286/3515 [16:40<01:02,  3.67it/s]
overall:  82%|████████▏ | 3287/4000 [16:43<03:20,  3.56it/s]

train:  94%|█████████▎| 3287/3515 [16:41<01:04,  3.56it/s]
overall:  82%|████████▏ | 3288/4000 [16:44<03:17,  3.60it/s]

train:  94%|█████████▎| 3288/3515 [16:41<01:03,  3.60it/s]
overall:  82%|████████▏ | 3289/4000 [16:44<03:17,  3.59it/s]

train:  94%|█████████▎| 3289/3515 [16:41<01:02,  3.59it/s]
overall:  82%|████████▏ | 3290/4000 [16:44<03:11,  3.72it/s]

train:  94%|█████████▎| 3290/3515 [16:42<01:00,  3.72it/s]
overall:  82%|████████▏ | 3291/4000 [16:44<03:12,  3.68it/s]

train:  94%|█████████▎| 3291/3515 [16:42<01:00,  3.68it/s]
overall:  82%|████████▏ | 3292/4000 [16:45<03:09,  3.74it/s]

train:  94%|█████████▎| 3292/3515 [16:42<00:59,  3.74it/s]
overall:  82%|████████▏ | 3293/4000 [16:45<03:16,  3.60it/s]

train:  94%|█████████▎| 3293/3515 [16:42<01:01,  3.60it/s]
overall:  82%|████████▏ | 3294/4000 [16:45<03:13,  3.64it/s]

train:  94%|█████████▎| 3294/3515 [16:43<01:00,  3.64it/s]
overall:  82%|████████▏ | 3295/4000 [16:46<03:11,  3.69it/s]

train:  94%|█████████▎| 3295/3515 [16:43<00:59,  3.69it/s]
overall:  82%|████████▏ | 3296/4000 [16:46<03:14,  3.62it/s]

train:  94%|█████████▍| 3296/3515 [16:43<01:00,  3.62it/s]
overall:  82%|████████▏ | 3297/4000 [16:46<03:21,  3.49it/s]

train:  94%|█████████▍| 3297/3515 [16:44<01:02,  3.49it/s]
overall:  82%|████████▏ | 3298/4000 [16:46<03:16,  3.58it/s]

train:  94%|█████████▍| 3298/3515 [16:44<01:00,  3.58it/s]
overall:  82%|████████▏ | 3299/4000 [16:47<03:12,  3.65it/s]

train:  94%|█████████▍| 3299/3515 [16:44<00:59,  3.65it/s]train loss: 0.5707898139953613
train per: 0.16618500347245557

overall:  82%|████████▎ | 3300/4000 [16:47<03:17,  3.54it/s]

train:  94%|█████████▍| 3300/3515 [16:44<01:00,  3.54it/s]
overall:  83%|████████▎ | 3301/4000 [16:47<03:24,  3.41it/s]

train:  94%|█████████▍| 3301/3515 [16:45<01:02,  3.41it/s]
overall:  83%|████████▎ | 3303/4000 [16:48<02:40,  4.35it/s]

train:  94%|█████████▍| 3303/3515 [16:45<00:48,  4.35it/s]
overall:  83%|████████▎ | 3304/4000 [16:48<02:51,  4.06it/s]

train:  94%|█████████▍| 3304/3515 [16:45<00:51,  4.06it/s]
overall:  83%|████████▎ | 3305/4000 [16:48<02:56,  3.93it/s]

train:  94%|█████████▍| 3305/3515 [16:46<00:53,  3.93it/s]
overall:  83%|████████▎ | 3306/4000 [16:49<03:13,  3.58it/s]

train:  94%|█████████▍| 3306/3515 [16:46<00:58,  3.58it/s]
overall:  83%|████████▎ | 3307/4000 [16:49<03:13,  3.58it/s]

train:  94%|█████████▍| 3307/3515 [16:46<00:58,  3.58it/s]
overall:  83%|████████▎ | 3308/4000 [16:49<03:05,  3.73it/s]

train:  94%|█████████▍| 3308/3515 [16:46<00:55,  3.73it/s]
overall:  83%|████████▎ | 3309/4000 [16:49<03:06,  3.71it/s]

train:  94%|█████████▍| 3309/3515 [16:47<00:55,  3.71it/s]
overall:  83%|████████▎ | 3310/4000 [16:50<02:43,  4.21it/s]

train:  94%|█████████▍| 3310/3515 [16:47<00:48,  4.21it/s]
overall:  83%|████████▎ | 3311/4000 [16:50<03:48,  3.01it/s]

train:  94%|█████████▍| 3311/3515 [16:47<01:07,  3.01it/s]
overall:  83%|████████▎ | 3312/4000 [16:50<03:26,  3.33it/s]

train:  94%|█████████▍| 3312/3515 [16:48<01:00,  3.33it/s]
overall:  83%|████████▎ | 3313/4000 [16:51<03:19,  3.45it/s]

train:  94%|█████████▍| 3313/3515 [16:48<00:58,  3.45it/s]
overall:  83%|████████▎ | 3314/4000 [16:51<03:17,  3.47it/s]

train:  94%|█████████▍| 3314/3515 [16:48<00:58,  3.47it/s]
overall:  83%|████████▎ | 3315/4000 [16:51<02:54,  3.92it/s]

train:  94%|█████████▍| 3315/3515 [16:48<00:51,  3.92it/s]
overall:  83%|████████▎ | 3316/4000 [16:51<02:56,  3.88it/s]

train:  94%|█████████▍| 3316/3515 [16:49<00:51,  3.88it/s]
overall:  83%|████████▎ | 3317/4000 [16:52<02:56,  3.87it/s]

train:  94%|█████████▍| 3317/3515 [16:49<00:51,  3.87it/s]
overall:  83%|████████▎ | 3318/4000 [16:52<02:57,  3.85it/s]

train:  94%|█████████▍| 3318/3515 [16:49<00:51,  3.85it/s]
overall:  83%|████████▎ | 3319/4000 [16:52<02:59,  3.80it/s]

train:  94%|█████████▍| 3319/3515 [16:49<00:51,  3.80it/s]train loss: 0.5576000213623047
train per: 0.1615813930617363

overall:  83%|████████▎ | 3320/4000 [16:52<03:00,  3.76it/s]

train:  94%|█████████▍| 3320/3515 [16:50<00:51,  3.75it/s]
overall:  83%|████████▎ | 3321/4000 [16:53<02:59,  3.78it/s]

train:  94%|█████████▍| 3321/3515 [16:50<00:51,  3.78it/s]
overall:  83%|████████▎ | 3322/4000 [16:53<02:55,  3.87it/s]

train:  95%|█████████▍| 3322/3515 [16:50<00:49,  3.87it/s]
overall:  83%|████████▎ | 3323/4000 [16:53<02:55,  3.85it/s]

train:  95%|█████████▍| 3323/3515 [16:50<00:49,  3.85it/s]
overall:  83%|████████▎ | 3324/4000 [16:53<02:57,  3.81it/s]

train:  95%|█████████▍| 3324/3515 [16:51<00:50,  3.82it/s]
overall:  83%|████████▎ | 3325/4000 [16:54<02:58,  3.79it/s]

train:  95%|█████████▍| 3325/3515 [16:51<00:50,  3.79it/s]
overall:  83%|████████▎ | 3326/4000 [16:54<02:58,  3.77it/s]

train:  95%|█████████▍| 3326/3515 [16:51<00:50,  3.77it/s]
overall:  83%|████████▎ | 3327/4000 [16:54<02:57,  3.79it/s]

train:  95%|█████████▍| 3327/3515 [16:52<00:49,  3.79it/s]
overall:  83%|████████▎ | 3328/4000 [16:54<02:59,  3.75it/s]

train:  95%|█████████▍| 3328/3515 [16:52<00:49,  3.75it/s]
overall:  83%|████████▎ | 3329/4000 [16:55<02:57,  3.79it/s]

train:  95%|█████████▍| 3329/3515 [16:52<00:49,  3.79it/s]
overall:  83%|████████▎ | 3330/4000 [16:55<02:59,  3.73it/s]

train:  95%|█████████▍| 3330/3515 [16:52<00:49,  3.73it/s]
overall:  83%|████████▎ | 3331/4000 [16:55<02:59,  3.72it/s]

train:  95%|█████████▍| 3331/3515 [16:53<00:49,  3.72it/s]
overall:  83%|████████▎ | 3332/4000 [16:56<02:59,  3.72it/s]

train:  95%|█████████▍| 3332/3515 [16:53<00:49,  3.72it/s]
overall:  83%|████████▎ | 3333/4000 [16:56<03:00,  3.69it/s]

train:  95%|█████████▍| 3333/3515 [16:53<00:49,  3.69it/s]
overall:  83%|████████▎ | 3334/4000 [16:56<02:58,  3.72it/s]

train:  95%|█████████▍| 3334/3515 [16:53<00:48,  3.72it/s]
overall:  83%|████████▎ | 3335/4000 [16:56<03:29,  3.17it/s]

train:  95%|█████████▍| 3335/3515 [16:54<00:56,  3.17it/s]
overall:  83%|████████▎ | 3336/4000 [16:57<03:18,  3.35it/s]

train:  95%|█████████▍| 3336/3515 [16:54<00:53,  3.35it/s]
overall:  83%|████████▎ | 3337/4000 [16:57<03:01,  3.66it/s]

train:  95%|█████████▍| 3337/3515 [16:54<00:48,  3.66it/s]
overall:  83%|████████▎ | 3338/4000 [16:57<02:43,  4.04it/s]

train:  95%|█████████▍| 3338/3515 [16:55<00:43,  4.04it/s]
overall:  83%|████████▎ | 3339/4000 [16:57<02:47,  3.96it/s]

train:  95%|█████████▍| 3339/3515 [16:55<00:44,  3.96it/s]train loss: 0.5570621490478516
train per: 0.16098779395824497

overall:  84%|████████▎ | 3340/4000 [16:58<02:49,  3.90it/s]

train:  95%|█████████▌| 3340/3515 [16:55<00:44,  3.90it/s]
overall:  84%|████████▎ | 3341/4000 [16:58<02:37,  4.17it/s]

train:  95%|█████████▌| 3341/3515 [16:55<00:41,  4.18it/s]
overall:  84%|████████▎ | 3342/4000 [16:58<02:44,  4.01it/s]

train:  95%|█████████▌| 3342/3515 [16:56<00:43,  4.01it/s]
overall:  84%|████████▎ | 3343/4000 [16:58<02:45,  3.97it/s]

train:  95%|█████████▌| 3343/3515 [16:56<00:43,  3.97it/s]
overall:  84%|████████▎ | 3344/4000 [16:59<02:48,  3.89it/s]

train:  95%|█████████▌| 3344/3515 [16:56<00:43,  3.89it/s]
overall:  84%|████████▎ | 3345/4000 [16:59<02:38,  4.13it/s]

train:  95%|█████████▌| 3345/3515 [16:56<00:41,  4.13it/s]
overall:  84%|████████▎ | 3346/4000 [16:59<02:41,  4.06it/s]

train:  95%|█████████▌| 3346/3515 [16:57<00:41,  4.06it/s]
overall:  84%|████████▎ | 3348/4000 [16:59<02:08,  5.06it/s]

train:  95%|█████████▌| 3348/3515 [16:57<00:32,  5.06it/s]
overall:  84%|████████▎ | 3349/4000 [17:00<02:12,  4.92it/s]

train:  95%|█████████▌| 3349/3515 [16:57<00:33,  4.92it/s]
overall:  84%|████████▍ | 3350/4000 [17:00<02:21,  4.60it/s]

train:  95%|█████████▌| 3350/3515 [16:57<00:35,  4.60it/s]
overall:  84%|████████▍ | 3351/4000 [17:00<02:31,  4.28it/s]

train:  95%|█████████▌| 3351/3515 [16:58<00:38,  4.28it/s]
overall:  84%|████████▍ | 3352/4000 [17:00<02:29,  4.32it/s]

train:  95%|█████████▌| 3352/3515 [16:58<00:37,  4.32it/s]
overall:  84%|████████▍ | 3354/4000 [17:01<02:13,  4.83it/s]

train:  95%|█████████▌| 3354/3515 [16:58<00:33,  4.83it/s]
overall:  84%|████████▍ | 3355/4000 [17:01<02:22,  4.53it/s]

train:  95%|█████████▌| 3355/3515 [16:58<00:35,  4.53it/s]
overall:  84%|████████▍ | 3356/4000 [17:01<02:30,  4.27it/s]

train:  95%|█████████▌| 3356/3515 [16:59<00:37,  4.27it/s]
overall:  84%|████████▍ | 3357/4000 [17:02<02:32,  4.22it/s]

train:  96%|█████████▌| 3357/3515 [16:59<00:37,  4.22it/s]
overall:  84%|████████▍ | 3358/4000 [17:02<02:33,  4.20it/s]

train:  96%|█████████▌| 3358/3515 [16:59<00:37,  4.20it/s]
overall:  84%|████████▍ | 3359/4000 [17:02<03:36,  2.97it/s]

train:  96%|█████████▌| 3359/3515 [17:00<00:52,  2.97it/s]train loss: 0.5441845655441284
train per: 0.1629421423204297

overall:  84%|████████▍ | 3360/4000 [17:03<03:21,  3.17it/s]

train:  96%|█████████▌| 3360/3515 [17:00<00:48,  3.17it/s]
overall:  84%|████████▍ | 3361/4000 [17:03<03:08,  3.40it/s]

train:  96%|█████████▌| 3361/3515 [17:00<00:45,  3.40it/s]
overall:  84%|████████▍ | 3362/4000 [17:03<03:01,  3.51it/s]

train:  96%|█████████▌| 3362/3515 [17:01<00:43,  3.51it/s]
overall:  84%|████████▍ | 3363/4000 [17:03<02:55,  3.64it/s]

train:  96%|█████████▌| 3363/3515 [17:01<00:41,  3.64it/s]
overall:  84%|████████▍ | 3364/4000 [17:04<02:50,  3.72it/s]

train:  96%|█████████▌| 3364/3515 [17:01<00:40,  3.72it/s]
overall:  84%|████████▍ | 3365/4000 [17:04<02:50,  3.73it/s]

train:  96%|█████████▌| 3365/3515 [17:01<00:40,  3.73it/s]
overall:  84%|████████▍ | 3366/4000 [17:04<02:50,  3.71it/s]

train:  96%|█████████▌| 3366/3515 [17:02<00:40,  3.71it/s]
overall:  84%|████████▍ | 3367/4000 [17:04<02:51,  3.70it/s]

train:  96%|█████████▌| 3367/3515 [17:02<00:40,  3.70it/s]
overall:  84%|████████▍ | 3368/4000 [17:05<02:50,  3.71it/s]

train:  96%|█████████▌| 3368/3515 [17:02<00:39,  3.71it/s]
overall:  84%|████████▍ | 3369/4000 [17:05<02:49,  3.72it/s]

train:  96%|█████████▌| 3369/3515 [17:02<00:39,  3.72it/s]
overall:  84%|████████▍ | 3370/4000 [17:05<02:48,  3.73it/s]

train:  96%|█████████▌| 3370/3515 [17:03<00:38,  3.73it/s]
overall:  84%|████████▍ | 3371/4000 [17:06<02:44,  3.83it/s]

train:  96%|█████████▌| 3371/3515 [17:03<00:37,  3.83it/s]
overall:  84%|████████▍ | 3372/4000 [17:06<02:45,  3.80it/s]

train:  96%|█████████▌| 3372/3515 [17:03<00:37,  3.80it/s]
overall:  84%|████████▍ | 3373/4000 [17:06<02:43,  3.84it/s]

train:  96%|█████████▌| 3373/3515 [17:03<00:36,  3.84it/s]
overall:  84%|████████▍ | 3374/4000 [17:06<02:42,  3.86it/s]

train:  96%|█████████▌| 3374/3515 [17:04<00:36,  3.86it/s]
overall:  84%|████████▍ | 3375/4000 [17:07<02:42,  3.84it/s]

train:  96%|█████████▌| 3375/3515 [17:04<00:36,  3.84it/s]
overall:  84%|████████▍ | 3377/4000 [17:07<02:19,  4.48it/s]

train:  96%|█████████▌| 3377/3515 [17:04<00:30,  4.48it/s]
overall:  84%|████████▍ | 3378/4000 [17:07<02:17,  4.52it/s]

train:  96%|█████████▌| 3378/3515 [17:04<00:30,  4.52it/s]
overall:  84%|████████▍ | 3379/4000 [17:07<02:13,  4.64it/s]

train:  96%|█████████▌| 3379/3515 [17:05<00:29,  4.64it/s]train loss: 0.5493574142456055
train per: 0.16063520347999266

overall:  84%|████████▍ | 3380/4000 [17:08<02:59,  3.45it/s]

train:  96%|█████████▌| 3380/3515 [17:05<00:39,  3.45it/s]
overall:  85%|████████▍ | 3381/4000 [17:08<02:55,  3.54it/s]

train:  96%|█████████▌| 3381/3515 [17:05<00:37,  3.54it/s]
overall:  85%|████████▍ | 3382/4000 [17:08<02:47,  3.69it/s]

train:  96%|█████████▌| 3382/3515 [17:06<00:36,  3.69it/s]
overall:  85%|████████▍ | 3383/4000 [17:09<02:54,  3.53it/s]

train:  96%|█████████▌| 3383/3515 [17:06<00:37,  3.53it/s]
overall:  85%|████████▍ | 3384/4000 [17:09<02:45,  3.71it/s]

train:  96%|█████████▋| 3384/3515 [17:06<00:35,  3.71it/s]
overall:  85%|████████▍ | 3385/4000 [17:09<02:42,  3.78it/s]

train:  96%|█████████▋| 3385/3515 [17:06<00:34,  3.78it/s]
overall:  85%|████████▍ | 3386/4000 [17:09<02:32,  4.02it/s]

train:  96%|█████████▋| 3386/3515 [17:07<00:32,  4.02it/s]
overall:  85%|████████▍ | 3387/4000 [17:10<02:23,  4.27it/s]

train:  96%|█████████▋| 3387/3515 [17:07<00:30,  4.27it/s]
overall:  85%|████████▍ | 3388/4000 [17:10<02:18,  4.41it/s]

train:  96%|█████████▋| 3388/3515 [17:07<00:28,  4.41it/s]
overall:  85%|████████▍ | 3389/4000 [17:10<01:55,  5.29it/s]

train:  96%|█████████▋| 3389/3515 [17:07<00:23,  5.29it/s]
overall:  85%|████████▍ | 3390/4000 [17:10<02:10,  4.69it/s]

train:  96%|█████████▋| 3390/3515 [17:07<00:26,  4.69it/s]
overall:  85%|████████▍ | 3391/4000 [17:10<02:17,  4.42it/s]

train:  96%|█████████▋| 3391/3515 [17:08<00:28,  4.42it/s]
overall:  85%|████████▍ | 3392/4000 [17:11<02:13,  4.54it/s]

train:  97%|█████████▋| 3392/3515 [17:08<00:27,  4.54it/s]
overall:  85%|████████▍ | 3393/4000 [17:11<02:21,  4.28it/s]

train:  97%|█████████▋| 3393/3515 [17:08<00:28,  4.28it/s]
overall:  85%|████████▍ | 3394/4000 [17:11<02:27,  4.11it/s]

train:  97%|█████████▋| 3394/3515 [17:08<00:29,  4.11it/s]
overall:  85%|████████▍ | 3396/4000 [17:11<01:41,  5.93it/s]

train:  97%|█████████▋| 3396/3515 [17:09<00:20,  5.93it/s]
overall:  85%|████████▍ | 3397/4000 [17:12<01:55,  5.24it/s]

train:  97%|█████████▋| 3397/3515 [17:09<00:22,  5.24it/s]
overall:  85%|████████▍ | 3398/4000 [17:12<02:08,  4.69it/s]

train:  97%|█████████▋| 3398/3515 [17:09<00:24,  4.69it/s]
overall:  85%|████████▍ | 3399/4000 [17:12<02:11,  4.56it/s]

train:  97%|█████████▋| 3399/3515 [17:09<00:25,  4.56it/s]train loss: 0.552820086479187
train per: 0.16244688900529655

overall:  85%|████████▌ | 3400/4000 [17:12<02:19,  4.30it/s]

train:  97%|█████████▋| 3400/3515 [17:10<00:26,  4.29it/s]
overall:  85%|████████▌ | 3401/4000 [17:13<02:26,  4.10it/s]

train:  97%|█████████▋| 3401/3515 [17:10<00:27,  4.10it/s]
overall:  85%|████████▌ | 3402/4000 [17:13<02:29,  3.99it/s]

train:  97%|█████████▋| 3402/3515 [17:10<00:28,  3.99it/s]
overall:  85%|████████▌ | 3403/4000 [17:13<02:29,  3.98it/s]

train:  97%|█████████▋| 3403/3515 [17:10<00:28,  3.98it/s]
overall:  85%|████████▌ | 3404/4000 [17:14<03:12,  3.09it/s]

train:  97%|█████████▋| 3404/3515 [17:11<00:35,  3.09it/s]
overall:  85%|████████▌ | 3405/4000 [17:14<03:04,  3.23it/s]

train:  97%|█████████▋| 3405/3515 [17:11<00:34,  3.23it/s]
overall:  85%|████████▌ | 3406/4000 [17:14<02:51,  3.46it/s]

train:  97%|█████████▋| 3406/3515 [17:11<00:31,  3.46it/s]
overall:  85%|████████▌ | 3407/4000 [17:14<03:02,  3.25it/s]

train:  97%|█████████▋| 3407/3515 [17:12<00:33,  3.25it/s]
overall:  85%|████████▌ | 3408/4000 [17:15<02:50,  3.47it/s]

train:  97%|█████████▋| 3408/3515 [17:12<00:30,  3.47it/s]
overall:  85%|████████▌ | 3409/4000 [17:15<02:42,  3.64it/s]

train:  97%|█████████▋| 3409/3515 [17:12<00:29,  3.64it/s]
overall:  85%|████████▌ | 3410/4000 [17:15<02:41,  3.65it/s]

train:  97%|█████████▋| 3410/3515 [17:13<00:28,  3.65it/s]
overall:  85%|████████▌ | 3412/4000 [17:16<02:09,  4.54it/s]

train:  97%|█████████▋| 3412/3515 [17:13<00:22,  4.54it/s]
overall:  85%|████████▌ | 3413/4000 [17:16<02:15,  4.33it/s]

train:  97%|█████████▋| 3413/3515 [17:13<00:23,  4.33it/s]
overall:  85%|████████▌ | 3414/4000 [17:16<02:17,  4.26it/s]

train:  97%|█████████▋| 3414/3515 [17:13<00:23,  4.26it/s]
overall:  85%|████████▌ | 3415/4000 [17:16<02:24,  4.05it/s]

train:  97%|█████████▋| 3415/3515 [17:14<00:24,  4.05it/s]
overall:  85%|████████▌ | 3416/4000 [17:17<02:28,  3.94it/s]

train:  97%|█████████▋| 3416/3515 [17:14<00:25,  3.94it/s]
overall:  85%|████████▌ | 3417/4000 [17:17<02:29,  3.89it/s]

train:  97%|█████████▋| 3417/3515 [17:14<00:25,  3.89it/s]
overall:  85%|████████▌ | 3418/4000 [17:17<02:30,  3.86it/s]

train:  97%|█████████▋| 3418/3515 [17:14<00:25,  3.86it/s]
overall:  85%|████████▌ | 3419/4000 [17:17<02:12,  4.38it/s]

train:  97%|█████████▋| 3419/3515 [17:15<00:21,  4.38it/s]train loss: 0.5815741419792175
train per: 0.16973496865086304

overall:  86%|████████▌ | 3420/4000 [17:18<02:18,  4.20it/s]

train:  97%|█████████▋| 3420/3515 [17:15<00:22,  4.20it/s]
overall:  86%|████████▌ | 3421/4000 [17:18<02:20,  4.12it/s]

train:  97%|█████████▋| 3421/3515 [17:15<00:22,  4.12it/s]
overall:  86%|████████▌ | 3422/4000 [17:18<02:22,  4.07it/s]

train:  97%|█████████▋| 3422/3515 [17:15<00:22,  4.07it/s]
overall:  86%|████████▌ | 3423/4000 [17:18<02:25,  3.97it/s]

train:  97%|█████████▋| 3423/3515 [17:16<00:23,  3.97it/s]
overall:  86%|████████▌ | 3424/4000 [17:19<02:27,  3.90it/s]

train:  97%|█████████▋| 3424/3515 [17:16<00:23,  3.90it/s]
overall:  86%|████████▌ | 3425/4000 [17:19<02:30,  3.83it/s]

train:  97%|█████████▋| 3425/3515 [17:16<00:23,  3.83it/s]
overall:  86%|████████▌ | 3426/4000 [17:19<02:29,  3.84it/s]

train:  97%|█████████▋| 3426/3515 [17:16<00:23,  3.84it/s]
overall:  86%|████████▌ | 3427/4000 [17:19<02:28,  3.86it/s]

train:  97%|█████████▋| 3427/3515 [17:17<00:22,  3.86it/s]
overall:  86%|████████▌ | 3428/4000 [17:20<03:07,  3.05it/s]

train:  98%|█████████▊| 3428/3515 [17:17<00:28,  3.05it/s]
overall:  86%|████████▌ | 3429/4000 [17:20<02:52,  3.31it/s]

train:  98%|█████████▊| 3429/3515 [17:17<00:26,  3.31it/s]
overall:  86%|████████▌ | 3430/4000 [17:20<02:43,  3.48it/s]

train:  98%|█████████▊| 3430/3515 [17:18<00:24,  3.48it/s]
overall:  86%|████████▌ | 3431/4000 [17:21<02:30,  3.79it/s]

train:  98%|█████████▊| 3431/3515 [17:18<00:22,  3.79it/s]
overall:  86%|████████▌ | 3432/4000 [17:21<02:31,  3.76it/s]

train:  98%|█████████▊| 3432/3515 [17:18<00:22,  3.76it/s]
overall:  86%|████████▌ | 3433/4000 [17:21<02:29,  3.79it/s]

train:  98%|█████████▊| 3433/3515 [17:18<00:21,  3.79it/s]
overall:  86%|████████▌ | 3434/4000 [17:21<02:20,  4.03it/s]

train:  98%|█████████▊| 3434/3515 [17:19<00:20,  4.03it/s]
overall:  86%|████████▌ | 3435/4000 [17:22<02:23,  3.94it/s]

train:  98%|█████████▊| 3435/3515 [17:19<00:20,  3.94it/s]
overall:  86%|████████▌ | 3436/4000 [17:22<02:21,  3.98it/s]

train:  98%|█████████▊| 3436/3515 [17:19<00:19,  3.98it/s]
overall:  86%|████████▌ | 3437/4000 [17:22<02:23,  3.93it/s]

train:  98%|█████████▊| 3437/3515 [17:19<00:19,  3.93it/s]
overall:  86%|████████▌ | 3438/4000 [17:22<02:23,  3.92it/s]

train:  98%|█████████▊| 3438/3515 [17:20<00:19,  3.92it/s]
overall:  86%|████████▌ | 3439/4000 [17:23<02:21,  3.95it/s]

train:  98%|█████████▊| 3439/3515 [17:20<00:19,  3.95it/s]train loss: 0.5691909790039062
train per: 0.16515947226847702

overall:  86%|████████▌ | 3441/4000 [17:23<01:50,  5.04it/s]

train:  98%|█████████▊| 3441/3515 [17:20<00:14,  5.04it/s]
overall:  86%|████████▌ | 3442/4000 [17:23<01:58,  4.70it/s]

train:  98%|█████████▊| 3442/3515 [17:20<00:15,  4.70it/s]
overall:  86%|████████▌ | 3443/4000 [17:23<02:06,  4.39it/s]

train:  98%|█████████▊| 3443/3515 [17:21<00:16,  4.39it/s]
overall:  86%|████████▌ | 3444/4000 [17:24<02:13,  4.16it/s]

train:  98%|█████████▊| 3444/3515 [17:21<00:17,  4.16it/s]
overall:  86%|████████▌ | 3445/4000 [17:24<02:16,  4.07it/s]

train:  98%|█████████▊| 3445/3515 [17:21<00:17,  4.07it/s]
overall:  86%|████████▌ | 3446/4000 [17:24<02:20,  3.94it/s]

train:  98%|█████████▊| 3446/3515 [17:22<00:17,  3.94it/s]
overall:  86%|████████▌ | 3447/4000 [17:24<02:23,  3.86it/s]

train:  98%|█████████▊| 3447/3515 [17:22<00:17,  3.86it/s]
overall:  86%|████████▌ | 3448/4000 [17:25<02:22,  3.86it/s]

train:  98%|█████████▊| 3448/3515 [17:22<00:17,  3.86it/s]
overall:  86%|████████▌ | 3449/4000 [17:25<02:24,  3.83it/s]

train:  98%|█████████▊| 3449/3515 [17:22<00:17,  3.82it/s]
overall:  86%|████████▋ | 3450/4000 [17:25<02:23,  3.84it/s]

train:  98%|█████████▊| 3450/3515 [17:23<00:16,  3.84it/s]
overall:  86%|████████▋ | 3451/4000 [17:26<02:24,  3.81it/s]

train:  98%|█████████▊| 3451/3515 [17:23<00:16,  3.81it/s]
overall:  86%|████████▋ | 3452/4000 [17:26<03:02,  3.00it/s]

train:  98%|█████████▊| 3452/3515 [17:23<00:21,  3.00it/s]
overall:  86%|████████▋ | 3453/4000 [17:26<02:45,  3.31it/s]

train:  98%|█████████▊| 3453/3515 [17:24<00:18,  3.31it/s]
overall:  86%|████████▋ | 3454/4000 [17:26<02:38,  3.44it/s]

train:  98%|█████████▊| 3454/3515 [17:24<00:17,  3.44it/s]
overall:  86%|████████▋ | 3455/4000 [17:27<02:47,  3.26it/s]

train:  98%|█████████▊| 3455/3515 [17:24<00:18,  3.26it/s]
overall:  86%|████████▋ | 3456/4000 [17:27<02:38,  3.43it/s]

train:  98%|█████████▊| 3456/3515 [17:24<00:17,  3.43it/s]
overall:  86%|████████▋ | 3457/4000 [17:27<02:33,  3.53it/s]

train:  98%|█████████▊| 3457/3515 [17:25<00:16,  3.53it/s]
overall:  86%|████████▋ | 3458/4000 [17:28<02:31,  3.57it/s]

train:  98%|█████████▊| 3458/3515 [17:25<00:15,  3.57it/s]
overall:  86%|████████▋ | 3459/4000 [17:28<02:27,  3.67it/s]

train:  98%|█████████▊| 3459/3515 [17:25<00:15,  3.67it/s]train loss: 0.5588217973709106
train per: 0.16110490180424716

overall:  86%|████████▋ | 3460/4000 [17:28<02:28,  3.63it/s]

train:  98%|█████████▊| 3460/3515 [17:26<00:15,  3.63it/s]
overall:  87%|████████▋ | 3461/4000 [17:28<02:27,  3.65it/s]

train:  98%|█████████▊| 3461/3515 [17:26<00:14,  3.65it/s]
overall:  87%|████████▋ | 3462/4000 [17:29<02:27,  3.66it/s]

train:  98%|█████████▊| 3462/3515 [17:26<00:14,  3.66it/s]
overall:  87%|████████▋ | 3463/4000 [17:29<02:22,  3.77it/s]

train:  99%|█████████▊| 3463/3515 [17:26<00:13,  3.77it/s]
overall:  87%|████████▋ | 3464/4000 [17:29<02:21,  3.79it/s]

train:  99%|█████████▊| 3464/3515 [17:27<00:13,  3.79it/s]
overall:  87%|████████▋ | 3465/4000 [17:29<02:22,  3.76it/s]

train:  99%|█████████▊| 3465/3515 [17:27<00:13,  3.76it/s]
overall:  87%|████████▋ | 3466/4000 [17:30<02:23,  3.72it/s]

train:  99%|█████████▊| 3466/3515 [17:27<00:13,  3.72it/s]
overall:  87%|████████▋ | 3467/4000 [17:30<02:23,  3.71it/s]

train:  99%|█████████▊| 3467/3515 [17:27<00:12,  3.72it/s]
overall:  87%|████████▋ | 3468/4000 [17:30<02:21,  3.75it/s]

train:  99%|█████████▊| 3468/3515 [17:28<00:12,  3.75it/s]
overall:  87%|████████▋ | 3469/4000 [17:31<02:17,  3.87it/s]

train:  99%|█████████▊| 3469/3515 [17:28<00:11,  3.87it/s]
overall:  87%|████████▋ | 3470/4000 [17:31<02:20,  3.78it/s]

train:  99%|█████████▊| 3470/3515 [17:28<00:11,  3.78it/s]
overall:  87%|████████▋ | 3471/4000 [17:31<02:03,  4.30it/s]

train:  99%|█████████▊| 3471/3515 [17:28<00:10,  4.30it/s]
overall:  87%|████████▋ | 3472/4000 [17:31<02:08,  4.11it/s]

train:  99%|█████████▉| 3472/3515 [17:29<00:10,  4.11it/s]
overall:  87%|████████▋ | 3473/4000 [17:31<01:51,  4.72it/s]

train:  99%|█████████▉| 3473/3515 [17:29<00:08,  4.72it/s]
overall:  87%|████████▋ | 3474/4000 [17:32<01:58,  4.44it/s]

train:  99%|█████████▉| 3474/3515 [17:29<00:09,  4.44it/s]
overall:  87%|████████▋ | 3475/4000 [17:32<02:03,  4.24it/s]

train:  99%|█████████▉| 3475/3515 [17:29<00:09,  4.24it/s]
overall:  87%|████████▋ | 3476/4000 [17:32<02:44,  3.18it/s]

train:  99%|█████████▉| 3476/3515 [17:30<00:12,  3.18it/s]
overall:  87%|████████▋ | 3477/4000 [17:33<02:32,  3.43it/s]

train:  99%|█████████▉| 3477/3515 [17:30<00:11,  3.43it/s]
overall:  87%|████████▋ | 3478/4000 [17:33<02:28,  3.52it/s]

train:  99%|█████████▉| 3478/3515 [17:30<00:10,  3.52it/s]
overall:  87%|████████▋ | 3479/4000 [17:33<02:36,  3.34it/s]

train:  99%|█████████▉| 3479/3515 [17:31<00:10,  3.34it/s]train loss: 0.5628894567489624
train per: 0.16206759814326702

overall:  87%|████████▋ | 3480/4000 [17:34<02:32,  3.40it/s]

train:  99%|█████████▉| 3480/3515 [17:31<00:10,  3.40it/s]
overall:  87%|████████▋ | 3481/4000 [17:34<02:25,  3.56it/s]

train:  99%|█████████▉| 3481/3515 [17:31<00:09,  3.56it/s]
overall:  87%|████████▋ | 3482/4000 [17:34<02:17,  3.76it/s]

train:  99%|█████████▉| 3482/3515 [17:31<00:08,  3.76it/s]
overall:  87%|████████▋ | 3483/4000 [17:34<02:17,  3.77it/s]

train:  99%|█████████▉| 3483/3515 [17:32<00:08,  3.77it/s]
overall:  87%|████████▋ | 3484/4000 [17:35<02:15,  3.80it/s]

train:  99%|█████████▉| 3484/3515 [17:32<00:08,  3.80it/s]
overall:  87%|████████▋ | 3485/4000 [17:35<02:12,  3.89it/s]

train:  99%|█████████▉| 3485/3515 [17:32<00:07,  3.89it/s]
overall:  87%|████████▋ | 3486/4000 [17:35<01:50,  4.64it/s]

train:  99%|█████████▉| 3486/3515 [17:32<00:06,  4.64it/s]
overall:  87%|████████▋ | 3487/4000 [17:35<01:57,  4.37it/s]

train:  99%|█████████▉| 3487/3515 [17:33<00:06,  4.37it/s]
overall:  87%|████████▋ | 3488/4000 [17:35<01:59,  4.28it/s]

train:  99%|█████████▉| 3488/3515 [17:33<00:06,  4.28it/s]
overall:  87%|████████▋ | 3489/4000 [17:36<02:05,  4.08it/s]

train:  99%|█████████▉| 3489/3515 [17:33<00:06,  4.08it/s]
overall:  87%|████████▋ | 3490/4000 [17:36<02:09,  3.93it/s]

train:  99%|█████████▉| 3490/3515 [17:33<00:06,  3.93it/s]
overall:  87%|████████▋ | 3491/4000 [17:36<02:11,  3.88it/s]

train:  99%|█████████▉| 3491/3515 [17:34<00:06,  3.88it/s]
overall:  87%|████████▋ | 3492/4000 [17:36<02:11,  3.86it/s]

train:  99%|█████████▉| 3492/3515 [17:34<00:05,  3.86it/s]
overall:  87%|████████▋ | 3493/4000 [17:37<02:11,  3.85it/s]

train:  99%|█████████▉| 3493/3515 [17:34<00:05,  3.85it/s]
overall:  87%|████████▋ | 3494/4000 [17:37<01:54,  4.42it/s]

train:  99%|█████████▉| 3494/3515 [17:34<00:04,  4.42it/s]
overall:  87%|████████▋ | 3495/4000 [17:37<01:56,  4.34it/s]

train:  99%|█████████▉| 3495/3515 [17:34<00:04,  4.34it/s]
overall:  87%|████████▋ | 3496/4000 [17:37<02:01,  4.16it/s]

train:  99%|█████████▉| 3496/3515 [17:35<00:04,  4.16it/s]
overall:  87%|████████▋ | 3497/4000 [17:38<02:05,  4.00it/s]

train:  99%|█████████▉| 3497/3515 [17:35<00:04,  4.00it/s]
overall:  87%|████████▋ | 3498/4000 [17:38<02:07,  3.94it/s]

train: 100%|█████████▉| 3498/3515 [17:35<00:04,  3.94it/s]
overall:  87%|████████▋ | 3499/4000 [17:38<02:09,  3.86it/s]

train: 100%|█████████▉| 3499/3515 [17:36<00:04,  3.86it/s]train loss: 0.5614519715309143
train per: 0.1667933782955242

overall:  88%|████████▊ | 3500/4000 [17:39<02:35,  3.22it/s]

train: 100%|█████████▉| 3500/3515 [17:36<00:04,  3.22it/s]
overall:  88%|████████▊ | 3501/4000 [17:39<02:18,  3.60it/s]

train: 100%|█████████▉| 3501/3515 [17:36<00:03,  3.60it/s]
overall:  88%|████████▊ | 3502/4000 [17:39<02:07,  3.90it/s]

train: 100%|█████████▉| 3502/3515 [17:36<00:03,  3.90it/s]
overall:  88%|████████▊ | 3503/4000 [17:39<02:10,  3.81it/s]

train: 100%|█████████▉| 3503/3515 [17:37<00:03,  3.81it/s]
overall:  88%|████████▊ | 3504/4000 [17:40<02:10,  3.81it/s]

train: 100%|█████████▉| 3504/3515 [17:37<00:02,  3.81it/s]
overall:  88%|████████▊ | 3505/4000 [17:40<01:53,  4.37it/s]

train: 100%|█████████▉| 3505/3515 [17:37<00:02,  4.37it/s]
overall:  88%|████████▊ | 3506/4000 [17:40<01:57,  4.21it/s]

train: 100%|█████████▉| 3506/3515 [17:37<00:02,  4.21it/s]
overall:  88%|████████▊ | 3507/4000 [17:40<02:03,  3.98it/s]

train: 100%|█████████▉| 3507/3515 [17:38<00:02,  3.98it/s]
overall:  88%|████████▊ | 3508/4000 [17:41<02:06,  3.88it/s]

train: 100%|█████████▉| 3508/3515 [17:38<00:01,  3.88it/s]
overall:  88%|████████▊ | 3509/4000 [17:41<02:08,  3.81it/s]

train: 100%|█████████▉| 3509/3515 [17:38<00:01,  3.81it/s]
overall:  88%|████████▊ | 3510/4000 [17:41<02:08,  3.82it/s]

train: 100%|█████████▉| 3510/3515 [17:38<00:01,  3.82it/s]
overall:  88%|████████▊ | 3511/4000 [17:41<02:07,  3.84it/s]

train: 100%|█████████▉| 3511/3515 [17:39<00:01,  3.84it/s]
overall:  88%|████████▊ | 3512/4000 [17:42<02:04,  3.92it/s]

train: 100%|█████████▉| 3512/3515 [17:39<00:00,  3.92it/s]
overall:  88%|████████▊ | 3513/4000 [17:42<02:07,  3.82it/s]

train: 100%|█████████▉| 3513/3515 [17:39<00:00,  3.82it/s]
overall:  88%|████████▊ | 3514/4000 [17:42<02:07,  3.82it/s]

train: 100%|█████████▉| 3514/3515 [17:39<00:00,  3.82it/s]
overall:  88%|████████▊ | 3515/4000 [17:42<01:58,  4.10it/s]

train: 100%|██████████| 3515/3515 [17:40<00:00,  4.10it/s]
train: 100%|██████████| 3515/3515 [17:40<00:00,  3.32it/s]
[LibriPhone] - Taking the first phoneme sequences for a deterministic behavior.


word -> phonemes:   0%|          | 0/3515 [00:00<?, ?it/s]

word -> phonemes:   8%|▊         | 277/3515 [00:00<00:01, 2768.26it/s]

word -> phonemes:  16%|█▌        | 569/3515 [00:00<00:01, 2855.45it/s]

word -> phonemes:  25%|██▍       | 875/3515 [00:00<00:00, 2946.19it/s]

word -> phonemes:  33%|███▎      | 1170/3515 [00:00<00:00, 2917.59it/s]

word -> phonemes:  42%|████▏     | 1477/3515 [00:00<00:00, 2968.79it/s]

word -> phonemes:  50%|█████     | 1774/3515 [00:00<00:00, 2968.14it/s]

word -> phonemes:  59%|█████▉    | 2071/3515 [00:00<00:00, 2883.01it/s]

word -> phonemes:  67%|██████▋   | 2360/3515 [00:00<00:00, 2424.38it/s]

word -> phonemes:  74%|███████▍  | 2614/3515 [00:01<00:00, 2272.60it/s]

word -> phonemes:  81%|████████  | 2850/3515 [00:01<00:00, 2151.22it/s]

word -> phonemes:  87%|████████▋ | 3072/3515 [00:01<00:00, 2055.89it/s]

word -> phonemes:  93%|█████████▎| 3282/3515 [00:01<00:00, 2006.32it/s]

word -> phonemes:  99%|█████████▉| 3486/3515 [00:01<00:00, 1923.66it/s]
word -> phonemes: 100%|██████████| 3515/3515 [00:01<00:00, 2343.62it/s]


train:   0%|          | 0/3515 [00:00<?, ?it/s]
overall:  88%|████████▊ | 3516/4000 [17:46<10:28,  1.30s/it]

train:   0%|          | 1/3515 [00:01<1:04:59,  1.11s/it]
overall:  88%|████████▊ | 3517/4000 [17:46<08:13,  1.02s/it]

train:   0%|          | 2/3515 [00:01<39:43,  1.47it/s]  
overall:  88%|████████▊ | 3518/4000 [17:47<06:24,  1.25it/s]

train:   0%|          | 3/3515 [00:01<28:55,  2.02it/s]
overall:  88%|████████▊ | 3519/4000 [17:47<05:08,  1.56it/s]

train:   0%|          | 4/3515 [00:02<23:56,  2.44it/s]train loss: 0.5641670227050781
train per: 0.1653785099018919

overall:  88%|████████▊ | 3520/4000 [17:47<03:52,  2.07it/s]

train:   0%|          | 5/3515 [00:02<17:41,  3.31it/s]
overall:  88%|████████▊ | 3521/4000 [17:47<03:19,  2.41it/s]

train:   0%|          | 6/3515 [00:02<16:46,  3.49it/s]
overall:  88%|████████▊ | 3522/4000 [17:48<03:00,  2.65it/s]

train:   0%|          | 7/3515 [00:02<16:50,  3.47it/s]
overall:  88%|████████▊ | 3523/4000 [17:48<02:47,  2.84it/s]

train:   0%|          | 8/3515 [00:02<16:52,  3.46it/s]
overall:  88%|████████▊ | 3524/4000 [17:48<02:29,  3.18it/s]

train:   0%|          | 9/3515 [00:03<15:45,  3.71it/s]
overall:  88%|████████▊ | 3525/4000 [17:48<02:24,  3.28it/s]

train:   0%|          | 10/3515 [00:03<15:59,  3.65it/s]
overall:  88%|████████▊ | 3526/4000 [17:49<02:17,  3.44it/s]

train:   0%|          | 11/3515 [00:03<15:42,  3.72it/s]
overall:  88%|████████▊ | 3527/4000 [17:49<02:16,  3.46it/s]

train:   0%|          | 12/3515 [00:04<16:00,  3.65it/s]
overall:  88%|████████▊ | 3528/4000 [17:49<02:15,  3.49it/s]

train:   0%|          | 13/3515 [00:04<16:06,  3.62it/s]
overall:  88%|████████▊ | 3529/4000 [17:49<01:56,  4.05it/s]

train:   0%|          | 14/3515 [00:04<13:57,  4.18it/s]
overall:  88%|████████▊ | 3530/4000 [17:50<01:40,  4.65it/s]

train:   0%|          | 15/3515 [00:04<12:12,  4.78it/s]
overall:  88%|████████▊ | 3531/4000 [17:50<01:48,  4.31it/s]

train:   0%|          | 16/3515 [00:04<13:17,  4.39it/s]
overall:  88%|████████▊ | 3532/4000 [17:50<01:53,  4.13it/s]

train:   0%|          | 17/3515 [00:05<13:57,  4.18it/s]
overall:  88%|████████▊ | 3533/4000 [17:50<01:56,  4.00it/s]

train:   1%|          | 18/3515 [00:05<14:28,  4.03it/s]
overall:  88%|████████▊ | 3534/4000 [17:51<01:57,  3.98it/s]

train:   1%|          | 19/3515 [00:05<14:33,  4.00it/s]
overall:  88%|████████▊ | 3535/4000 [17:51<02:02,  3.79it/s]

train:   1%|          | 20/3515 [00:05<15:19,  3.80it/s]
overall:  88%|████████▊ | 3536/4000 [17:51<02:04,  3.74it/s]

train:   1%|          | 21/3515 [00:06<15:33,  3.74it/s]
overall:  88%|████████▊ | 3537/4000 [17:51<02:05,  3.70it/s]

train:   1%|          | 22/3515 [00:06<15:41,  3.71it/s]
overall:  88%|████████▊ | 3538/4000 [17:52<02:03,  3.73it/s]

train:   1%|          | 23/3515 [00:06<15:34,  3.74it/s]
overall:  88%|████████▊ | 3539/4000 [17:52<02:05,  3.69it/s]

train:   1%|          | 24/3515 [00:07<15:45,  3.69it/s]train loss: 0.564103364944458
train per: 0.16407518575013533

overall:  88%|████████▊ | 3540/4000 [17:52<02:08,  3.57it/s]

train:   1%|          | 25/3515 [00:07<16:16,  3.57it/s]
overall:  89%|████████▊ | 3541/4000 [17:53<02:12,  3.46it/s]

train:   1%|          | 26/3515 [00:07<16:49,  3.46it/s]
overall:  89%|████████▊ | 3542/4000 [17:53<02:07,  3.58it/s]

train:   1%|          | 27/3515 [00:07<16:13,  3.58it/s]
overall:  89%|████████▊ | 3543/4000 [17:53<02:08,  3.56it/s]

train:   1%|          | 28/3515 [00:08<16:20,  3.56it/s]
overall:  89%|████████▊ | 3544/4000 [17:53<02:03,  3.69it/s]

train:   1%|          | 29/3515 [00:08<15:43,  3.69it/s]
overall:  89%|████████▊ | 3545/4000 [17:54<02:01,  3.76it/s]

train:   1%|          | 30/3515 [00:08<15:27,  3.76it/s]
overall:  89%|████████▊ | 3546/4000 [17:54<02:01,  3.72it/s]

train:   1%|          | 31/3515 [00:08<15:35,  3.72it/s]
overall:  89%|████████▊ | 3547/4000 [17:54<02:05,  3.61it/s]

train:   1%|          | 32/3515 [00:09<16:04,  3.61it/s]
overall:  89%|████████▊ | 3548/4000 [17:54<01:46,  4.26it/s]

train:   1%|          | 33/3515 [00:09<13:36,  4.26it/s]
overall:  89%|████████▊ | 3549/4000 [17:55<01:51,  4.06it/s]

train:   1%|          | 34/3515 [00:09<14:17,  4.06it/s]
overall:  89%|████████▉ | 3550/4000 [17:55<01:53,  3.97it/s]

train:   1%|          | 35/3515 [00:09<14:36,  3.97it/s]
overall:  89%|████████▉ | 3551/4000 [17:55<01:56,  3.86it/s]

train:   1%|          | 36/3515 [00:10<15:00,  3.86it/s]
overall:  89%|████████▉ | 3552/4000 [17:55<01:55,  3.89it/s]

train:   1%|          | 37/3515 [00:10<14:54,  3.89it/s]
overall:  89%|████████▉ | 3553/4000 [17:56<01:59,  3.73it/s]

train:   1%|          | 38/3515 [00:10<15:31,  3.73it/s]
overall:  89%|████████▉ | 3554/4000 [17:56<02:00,  3.71it/s]

train:   1%|          | 39/3515 [00:11<15:35,  3.71it/s]
overall:  89%|████████▉ | 3555/4000 [17:56<01:55,  3.86it/s]

train:   1%|          | 40/3515 [00:11<14:59,  3.86it/s]
overall:  89%|████████▉ | 3556/4000 [17:57<01:56,  3.81it/s]

train:   1%|          | 41/3515 [00:11<15:11,  3.81it/s]
overall:  89%|████████▉ | 3557/4000 [17:57<01:55,  3.82it/s]

train:   1%|          | 42/3515 [00:11<15:09,  3.82it/s]
overall:  89%|████████▉ | 3558/4000 [17:57<01:57,  3.78it/s]

train:   1%|          | 43/3515 [00:12<15:19,  3.78it/s]
overall:  89%|████████▉ | 3559/4000 [17:57<02:00,  3.65it/s]

train:   1%|▏         | 44/3515 [00:12<15:49,  3.65it/s]train loss: 0.5375961065292358
train per: 0.1560834213564568

overall:  89%|████████▉ | 3560/4000 [17:58<02:00,  3.66it/s]

train:   1%|▏         | 45/3515 [00:12<15:48,  3.66it/s]
overall:  89%|████████▉ | 3561/4000 [17:58<01:59,  3.67it/s]

train:   1%|▏         | 46/3515 [00:12<15:45,  3.67it/s]
overall:  89%|████████▉ | 3562/4000 [17:58<01:57,  3.71it/s]

train:   1%|▏         | 47/3515 [00:13<15:34,  3.71it/s]
overall:  89%|████████▉ | 3563/4000 [17:58<01:37,  4.48it/s]

train:   1%|▏         | 48/3515 [00:13<12:54,  4.48it/s]
overall:  89%|████████▉ | 3564/4000 [17:59<01:43,  4.22it/s]

train:   1%|▏         | 49/3515 [00:13<13:41,  4.22it/s]
overall:  89%|████████▉ | 3565/4000 [17:59<01:52,  3.85it/s]

train:   1%|▏         | 50/3515 [00:13<14:59,  3.85it/s]
overall:  89%|████████▉ | 3566/4000 [17:59<01:53,  3.84it/s]

train:   1%|▏         | 51/3515 [00:14<15:02,  3.84it/s]
overall:  89%|████████▉ | 3567/4000 [17:59<01:43,  4.17it/s]

train:   1%|▏         | 52/3515 [00:14<13:50,  4.17it/s]
overall:  89%|████████▉ | 3568/4000 [18:00<01:47,  4.01it/s]

train:   2%|▏         | 53/3515 [00:14<14:24,  4.01it/s]
overall:  89%|████████▉ | 3569/4000 [18:00<01:48,  3.97it/s]

train:   2%|▏         | 54/3515 [00:14<14:31,  3.97it/s]
overall:  89%|████████▉ | 3570/4000 [18:00<01:52,  3.82it/s]

train:   2%|▏         | 55/3515 [00:15<15:05,  3.82it/s]
overall:  89%|████████▉ | 3571/4000 [18:00<01:56,  3.69it/s]

train:   2%|▏         | 56/3515 [00:15<15:38,  3.69it/s]
overall:  89%|████████▉ | 3572/4000 [18:01<01:55,  3.69it/s]

train:   2%|▏         | 57/3515 [00:15<15:36,  3.69it/s]
overall:  89%|████████▉ | 3573/4000 [18:01<01:56,  3.65it/s]

train:   2%|▏         | 58/3515 [00:16<15:47,  3.65it/s]
overall:  89%|████████▉ | 3574/4000 [18:01<01:37,  4.37it/s]

train:   2%|▏         | 59/3515 [00:16<13:10,  4.37it/s]
overall:  89%|████████▉ | 3575/4000 [18:01<01:43,  4.11it/s]

train:   2%|▏         | 60/3515 [00:16<14:01,  4.11it/s]
overall:  89%|████████▉ | 3576/4000 [18:02<01:39,  4.27it/s]

train:   2%|▏         | 61/3515 [00:16<13:29,  4.27it/s]
overall:  89%|████████▉ | 3577/4000 [18:02<01:23,  5.07it/s]

train:   2%|▏         | 62/3515 [00:16<11:20,  5.07it/s]
overall:  89%|████████▉ | 3578/4000 [18:02<01:33,  4.50it/s]

train:   2%|▏         | 63/3515 [00:17<12:46,  4.50it/s]
overall:  89%|████████▉ | 3579/4000 [18:02<01:41,  4.13it/s]

train:   2%|▏         | 64/3515 [00:17<13:54,  4.13it/s]train loss: 0.5584509968757629
train per: 0.16468735477874735

overall:  90%|████████▉ | 3580/4000 [18:03<01:45,  3.98it/s]

train:   2%|▏         | 65/3515 [00:17<14:28,  3.97it/s]
overall:  90%|████████▉ | 3581/4000 [18:03<01:47,  3.91it/s]

train:   2%|▏         | 66/3515 [00:17<14:41,  3.91it/s]
overall:  90%|████████▉ | 3582/4000 [18:03<01:50,  3.79it/s]

train:   2%|▏         | 67/3515 [00:18<15:08,  3.79it/s]
overall:  90%|████████▉ | 3583/4000 [18:03<01:51,  3.73it/s]

train:   2%|▏         | 68/3515 [00:18<15:24,  3.73it/s]
overall:  90%|████████▉ | 3584/4000 [18:04<01:50,  3.77it/s]

train:   2%|▏         | 69/3515 [00:18<15:14,  3.77it/s]
overall:  90%|████████▉ | 3585/4000 [18:04<01:48,  3.82it/s]

train:   2%|▏         | 70/3515 [00:18<15:02,  3.82it/s]
overall:  90%|████████▉ | 3586/4000 [18:04<01:49,  3.78it/s]

train:   2%|▏         | 71/3515 [00:19<15:10,  3.78it/s]
overall:  90%|████████▉ | 3587/4000 [18:04<01:48,  3.82it/s]

train:   2%|▏         | 72/3515 [00:19<15:01,  3.82it/s]
overall:  90%|████████▉ | 3588/4000 [18:05<01:50,  3.72it/s]

train:   2%|▏         | 73/3515 [00:19<15:26,  3.72it/s]
overall:  90%|████████▉ | 3589/4000 [18:05<01:56,  3.52it/s]

train:   2%|▏         | 74/3515 [00:20<16:16,  3.52it/s]
overall:  90%|████████▉ | 3590/4000 [18:05<01:52,  3.63it/s]

train:   2%|▏         | 75/3515 [00:20<15:47,  3.63it/s]
overall:  90%|████████▉ | 3591/4000 [18:06<01:52,  3.63it/s]

train:   2%|▏         | 76/3515 [00:20<15:47,  3.63it/s]
overall:  90%|████████▉ | 3592/4000 [18:06<01:52,  3.61it/s]

train:   2%|▏         | 77/3515 [00:20<15:51,  3.61it/s]
overall:  90%|████████▉ | 3593/4000 [18:06<01:39,  4.08it/s]

train:   2%|▏         | 78/3515 [00:21<14:02,  4.08it/s]
overall:  90%|████████▉ | 3594/4000 [18:06<01:44,  3.87it/s]

train:   2%|▏         | 79/3515 [00:21<14:48,  3.87it/s]
overall:  90%|████████▉ | 3595/4000 [18:07<01:47,  3.76it/s]

train:   2%|▏         | 80/3515 [00:21<15:12,  3.76it/s]
overall:  90%|████████▉ | 3596/4000 [18:07<01:39,  4.04it/s]

train:   2%|▏         | 81/3515 [00:21<14:09,  4.04it/s]
overall:  90%|████████▉ | 3597/4000 [18:07<01:43,  3.90it/s]

train:   2%|▏         | 82/3515 [00:22<14:39,  3.90it/s]
overall:  90%|████████▉ | 3598/4000 [18:07<01:43,  3.88it/s]

train:   2%|▏         | 83/3515 [00:22<14:43,  3.88it/s]
overall:  90%|████████▉ | 3599/4000 [18:08<01:44,  3.83it/s]

train:   2%|▏         | 84/3515 [00:22<14:56,  3.83it/s]train loss: 0.5849035382270813
train per: 0.17080432495354927

overall:  90%|█████████ | 3600/4000 [18:08<01:47,  3.72it/s]

train:   2%|▏         | 85/3515 [00:22<15:22,  3.72it/s]
overall:  90%|█████████ | 3601/4000 [18:08<01:53,  3.50it/s]

train:   2%|▏         | 86/3515 [00:23<16:19,  3.50it/s]
overall:  90%|█████████ | 3602/4000 [18:08<01:51,  3.56it/s]

train:   2%|▏         | 87/3515 [00:23<16:02,  3.56it/s]
overall:  90%|█████████ | 3603/4000 [18:09<01:54,  3.46it/s]

train:   3%|▎         | 88/3515 [00:23<16:29,  3.46it/s]
overall:  90%|█████████ | 3604/4000 [18:09<01:50,  3.58it/s]

train:   3%|▎         | 89/3515 [00:24<15:57,  3.58it/s]
overall:  90%|█████████ | 3605/4000 [18:09<01:47,  3.68it/s]

train:   3%|▎         | 90/3515 [00:24<15:30,  3.68it/s]
overall:  90%|█████████ | 3606/4000 [18:09<01:27,  4.49it/s]

train:   3%|▎         | 91/3515 [00:24<12:42,  4.49it/s]
overall:  90%|█████████ | 3607/4000 [18:10<01:36,  4.08it/s]

train:   3%|▎         | 92/3515 [00:24<13:59,  4.08it/s]
overall:  90%|█████████ | 3608/4000 [18:10<01:39,  3.95it/s]

train:   3%|▎         | 93/3515 [00:24<14:27,  3.95it/s]
overall:  90%|█████████ | 3609/4000 [18:10<01:41,  3.87it/s]

train:   3%|▎         | 94/3515 [00:25<14:44,  3.87it/s]
overall:  90%|█████████ | 3610/4000 [18:10<01:42,  3.82it/s]

train:   3%|▎         | 95/3515 [00:25<14:55,  3.82it/s]
overall:  90%|█████████ | 3611/4000 [18:11<01:44,  3.74it/s]

train:   3%|▎         | 96/3515 [00:25<15:15,  3.74it/s]
overall:  90%|█████████ | 3612/4000 [18:11<01:42,  3.80it/s]

train:   3%|▎         | 97/3515 [00:26<14:58,  3.80it/s]
overall:  90%|█████████ | 3613/4000 [18:11<01:49,  3.55it/s]

train:   3%|▎         | 98/3515 [00:26<16:02,  3.55it/s]
overall:  90%|█████████ | 3614/4000 [18:12<01:40,  3.83it/s]

train:   3%|▎         | 99/3515 [00:26<14:52,  3.83it/s]
overall:  90%|█████████ | 3615/4000 [18:12<01:45,  3.65it/s]

train:   3%|▎         | 100/3515 [00:26<15:36,  3.65it/s]
overall:  90%|█████████ | 3616/4000 [18:12<01:43,  3.70it/s]

train:   3%|▎         | 101/3515 [00:27<15:22,  3.70it/s]
overall:  90%|█████████ | 3617/4000 [18:12<01:42,  3.72it/s]

train:   3%|▎         | 102/3515 [00:27<15:17,  3.72it/s]
overall:  90%|█████████ | 3618/4000 [18:13<01:38,  3.86it/s]

train:   3%|▎         | 103/3515 [00:27<14:43,  3.86it/s]
overall:  90%|█████████ | 3619/4000 [18:13<01:42,  3.70it/s]

train:   3%|▎         | 104/3515 [00:27<15:20,  3.70it/s]train loss: 0.5562740564346313
train per: 0.1616896809740834

overall:  90%|█████████ | 3620/4000 [18:13<01:42,  3.71it/s]

train:   3%|▎         | 105/3515 [00:28<15:20,  3.71it/s]
overall:  91%|█████████ | 3621/4000 [18:13<01:42,  3.69it/s]

train:   3%|▎         | 106/3515 [00:28<15:22,  3.70it/s]
overall:  91%|█████████ | 3622/4000 [18:14<01:42,  3.68it/s]

train:   3%|▎         | 107/3515 [00:28<15:25,  3.68it/s]
overall:  91%|█████████ | 3623/4000 [18:14<01:44,  3.62it/s]

train:   3%|▎         | 108/3515 [00:29<15:40,  3.62it/s]
overall:  91%|█████████ | 3624/4000 [18:14<01:42,  3.65it/s]

train:   3%|▎         | 109/3515 [00:29<15:32,  3.65it/s]
overall:  91%|█████████ | 3625/4000 [18:15<01:47,  3.48it/s]

train:   3%|▎         | 110/3515 [00:29<16:18,  3.48it/s]
overall:  91%|█████████ | 3626/4000 [18:15<01:45,  3.53it/s]

train:   3%|▎         | 111/3515 [00:29<16:04,  3.53it/s]
overall:  91%|█████████ | 3627/4000 [18:15<01:46,  3.50it/s]

train:   3%|▎         | 112/3515 [00:30<16:11,  3.50it/s]
overall:  91%|█████████ | 3628/4000 [18:15<01:44,  3.57it/s]

train:   3%|▎         | 113/3515 [00:30<15:54,  3.57it/s]
overall:  91%|█████████ | 3629/4000 [18:16<01:39,  3.72it/s]

train:   3%|▎         | 114/3515 [00:30<15:14,  3.72it/s]
overall:  91%|█████████ | 3630/4000 [18:16<01:40,  3.68it/s]

train:   3%|▎         | 115/3515 [00:31<15:23,  3.68it/s]
overall:  91%|█████████ | 3631/4000 [18:16<01:42,  3.58it/s]

train:   3%|▎         | 116/3515 [00:31<15:48,  3.58it/s]
overall:  91%|█████████ | 3632/4000 [18:17<01:39,  3.68it/s]

train:   3%|▎         | 117/3515 [00:31<15:22,  3.68it/s]
overall:  91%|█████████ | 3633/4000 [18:17<01:32,  3.95it/s]

train:   3%|▎         | 118/3515 [00:31<14:20,  3.95it/s]
overall:  91%|█████████ | 3634/4000 [18:17<01:28,  4.12it/s]

train:   3%|▎         | 119/3515 [00:31<13:44,  4.12it/s]
overall:  91%|█████████ | 3635/4000 [18:17<01:21,  4.47it/s]

train:   3%|▎         | 120/3515 [00:32<12:40,  4.47it/s]
overall:  91%|█████████ | 3636/4000 [18:17<01:26,  4.20it/s]

train:   3%|▎         | 121/3515 [00:32<13:28,  4.20it/s]
overall:  91%|█████████ | 3637/4000 [18:18<01:36,  3.78it/s]

train:   3%|▎         | 122/3515 [00:32<14:57,  3.78it/s]
overall:  91%|█████████ | 3638/4000 [18:18<01:36,  3.75it/s]

train:   3%|▎         | 123/3515 [00:33<15:04,  3.75it/s]
overall:  91%|█████████ | 3639/4000 [18:18<01:38,  3.67it/s]

train:   4%|▎         | 124/3515 [00:33<15:22,  3.67it/s]train loss: 0.5532155632972717
train per: 0.16117531922825987

overall:  91%|█████████ | 3640/4000 [18:19<01:38,  3.65it/s]

train:   4%|▎         | 125/3515 [00:33<15:28,  3.65it/s]
overall:  91%|█████████ | 3641/4000 [18:19<01:36,  3.70it/s]

train:   4%|▎         | 126/3515 [00:33<15:14,  3.71it/s]
overall:  91%|█████████ | 3642/4000 [18:19<01:38,  3.64it/s]

train:   4%|▎         | 127/3515 [00:34<15:29,  3.65it/s]
overall:  91%|█████████ | 3643/4000 [18:19<01:39,  3.59it/s]

train:   4%|▎         | 128/3515 [00:34<15:44,  3.59it/s]
overall:  91%|█████████ | 3644/4000 [18:20<01:38,  3.62it/s]

train:   4%|▎         | 129/3515 [00:34<15:34,  3.63it/s]
overall:  91%|█████████ | 3645/4000 [18:20<01:37,  3.63it/s]

train:   4%|▎         | 130/3515 [00:34<15:33,  3.63it/s]
overall:  91%|█████████ | 3646/4000 [18:20<01:37,  3.65it/s]

train:   4%|▎         | 131/3515 [00:35<15:27,  3.65it/s]
overall:  91%|█████████ | 3647/4000 [18:20<01:35,  3.69it/s]

train:   4%|▍         | 132/3515 [00:35<15:17,  3.69it/s]
overall:  91%|█████████ | 3648/4000 [18:21<01:37,  3.62it/s]

train:   4%|▍         | 133/3515 [00:35<15:34,  3.62it/s]
overall:  91%|█████████ | 3649/4000 [18:21<01:39,  3.52it/s]

train:   4%|▍         | 134/3515 [00:36<16:01,  3.52it/s]
overall:  91%|█████████▏| 3651/4000 [18:21<01:23,  4.16it/s]

train:   4%|▍         | 136/3515 [00:36<13:32,  4.16it/s]
overall:  91%|█████████▏| 3652/4000 [18:22<01:23,  4.15it/s]

train:   4%|▍         | 137/3515 [00:36<13:34,  4.15it/s]
overall:  91%|█████████▏| 3653/4000 [18:22<01:23,  4.15it/s]

train:   4%|▍         | 138/3515 [00:36<13:33,  4.15it/s]
overall:  91%|█████████▏| 3654/4000 [18:22<01:24,  4.11it/s]

train:   4%|▍         | 139/3515 [00:37<13:42,  4.11it/s]
overall:  91%|█████████▏| 3655/4000 [18:22<01:28,  3.88it/s]

train:   4%|▍         | 140/3515 [00:37<14:29,  3.88it/s]
overall:  91%|█████████▏| 3656/4000 [18:23<01:28,  3.88it/s]

train:   4%|▍         | 141/3515 [00:37<14:30,  3.88it/s]
overall:  91%|█████████▏| 3657/4000 [18:23<01:49,  3.13it/s]

train:   4%|▍         | 142/3515 [00:38<17:58,  3.13it/s]
overall:  91%|█████████▏| 3658/4000 [18:23<01:41,  3.37it/s]

train:   4%|▍         | 143/3515 [00:38<16:40,  3.37it/s]
overall:  91%|█████████▏| 3659/4000 [18:24<01:37,  3.49it/s]

train:   4%|▍         | 144/3515 [00:38<16:07,  3.49it/s]train loss: 0.58301842212677
train per: 0.15970051610089409

overall:  92%|█████████▏| 3660/4000 [18:24<01:38,  3.45it/s]

train:   4%|▍         | 145/3515 [00:39<16:17,  3.45it/s]
overall:  92%|█████████▏| 3661/4000 [18:24<01:36,  3.53it/s]

train:   4%|▍         | 146/3515 [00:39<15:54,  3.53it/s]
overall:  92%|█████████▏| 3662/4000 [18:25<01:33,  3.62it/s]

train:   4%|▍         | 147/3515 [00:39<15:31,  3.62it/s]
overall:  92%|█████████▏| 3663/4000 [18:25<01:32,  3.63it/s]

train:   4%|▍         | 148/3515 [00:39<15:26,  3.63it/s]
overall:  92%|█████████▏| 3664/4000 [18:25<01:31,  3.67it/s]

train:   4%|▍         | 149/3515 [00:40<15:16,  3.67it/s]
overall:  92%|█████████▏| 3665/4000 [18:25<01:31,  3.66it/s]

train:   4%|▍         | 150/3515 [00:40<15:18,  3.66it/s]
overall:  92%|█████████▏| 3666/4000 [18:26<01:31,  3.65it/s]

train:   4%|▍         | 151/3515 [00:40<15:20,  3.65it/s]
overall:  92%|█████████▏| 3668/4000 [18:26<01:12,  4.61it/s]

train:   4%|▍         | 153/3515 [00:40<12:09,  4.61it/s]
overall:  92%|█████████▏| 3669/4000 [18:26<01:24,  3.90it/s]

train:   4%|▍         | 154/3515 [00:41<14:20,  3.90it/s]
overall:  92%|█████████▏| 3670/4000 [18:27<01:25,  3.87it/s]

train:   4%|▍         | 155/3515 [00:41<14:28,  3.87it/s]
overall:  92%|█████████▏| 3671/4000 [18:27<01:23,  3.94it/s]

train:   4%|▍         | 156/3515 [00:41<14:13,  3.94it/s]
overall:  92%|█████████▏| 3672/4000 [18:27<01:19,  4.11it/s]

train:   4%|▍         | 157/3515 [00:42<13:37,  4.11it/s]
overall:  92%|█████████▏| 3673/4000 [18:27<01:16,  4.27it/s]

train:   4%|▍         | 158/3515 [00:42<13:06,  4.27it/s]
overall:  92%|█████████▏| 3674/4000 [18:27<01:15,  4.30it/s]

train:   5%|▍         | 159/3515 [00:42<13:00,  4.30it/s]
overall:  92%|█████████▏| 3675/4000 [18:28<01:18,  4.12it/s]

train:   5%|▍         | 160/3515 [00:42<13:33,  4.12it/s]
overall:  92%|█████████▏| 3676/4000 [18:28<01:22,  3.93it/s]

train:   5%|▍         | 161/3515 [00:43<14:12,  3.93it/s]
overall:  92%|█████████▏| 3677/4000 [18:28<01:23,  3.86it/s]

train:   5%|▍         | 162/3515 [00:43<14:29,  3.86it/s]
overall:  92%|█████████▏| 3678/4000 [18:29<01:24,  3.81it/s]

train:   5%|▍         | 163/3515 [00:43<14:40,  3.81it/s]
overall:  92%|█████████▏| 3679/4000 [18:29<01:22,  3.89it/s]

train:   5%|▍         | 164/3515 [00:43<14:20,  3.89it/s]train loss: 0.5378901958465576
train per: 0.16071168149336446

overall:  92%|█████████▏| 3680/4000 [18:29<01:24,  3.81it/s]

train:   5%|▍         | 165/3515 [00:44<14:41,  3.80it/s]
overall:  92%|█████████▏| 3681/4000 [18:30<01:42,  3.12it/s]

train:   5%|▍         | 166/3515 [00:44<17:52,  3.12it/s]
overall:  92%|█████████▏| 3682/4000 [18:30<01:33,  3.40it/s]

train:   5%|▍         | 167/3515 [00:44<16:23,  3.40it/s]
overall:  92%|█████████▏| 3683/4000 [18:30<01:30,  3.50it/s]

train:   5%|▍         | 168/3515 [00:45<15:56,  3.50it/s]
overall:  92%|█████████▏| 3684/4000 [18:30<01:30,  3.50it/s]

train:   5%|▍         | 169/3515 [00:45<15:56,  3.50it/s]
overall:  92%|█████████▏| 3685/4000 [18:31<01:28,  3.56it/s]

train:   5%|▍         | 170/3515 [00:45<15:40,  3.56it/s]
overall:  92%|█████████▏| 3686/4000 [18:31<01:25,  3.65it/s]

train:   5%|▍         | 171/3515 [00:45<15:15,  3.65it/s]
overall:  92%|█████████▏| 3687/4000 [18:31<01:23,  3.77it/s]

train:   5%|▍         | 172/3515 [00:46<14:47,  3.77it/s]
overall:  92%|█████████▏| 3688/4000 [18:31<01:23,  3.72it/s]

train:   5%|▍         | 173/3515 [00:46<14:59,  3.72it/s]
overall:  92%|█████████▏| 3689/4000 [18:32<01:21,  3.82it/s]

train:   5%|▍         | 174/3515 [00:46<14:33,  3.82it/s]
overall:  92%|█████████▏| 3690/4000 [18:32<01:20,  3.86it/s]

train:   5%|▍         | 175/3515 [00:46<14:26,  3.86it/s]
overall:  92%|█████████▏| 3691/4000 [18:32<01:20,  3.82it/s]

train:   5%|▌         | 176/3515 [00:47<14:34,  3.82it/s]
overall:  92%|█████████▏| 3692/4000 [18:32<01:21,  3.76it/s]

train:   5%|▌         | 177/3515 [00:47<14:46,  3.76it/s]
overall:  92%|█████████▏| 3693/4000 [18:33<01:37,  3.13it/s]

train:   5%|▌         | 178/3515 [00:47<17:44,  3.13it/s]
overall:  92%|█████████▏| 3694/4000 [18:33<01:32,  3.31it/s]

train:   5%|▌         | 179/3515 [00:48<16:49,  3.31it/s]
overall:  92%|█████████▏| 3696/4000 [18:33<01:08,  4.44it/s]

train:   5%|▌         | 181/3515 [00:48<12:30,  4.44it/s]
overall:  92%|█████████▏| 3697/4000 [18:34<01:11,  4.22it/s]

train:   5%|▌         | 182/3515 [00:48<13:10,  4.22it/s]
overall:  92%|█████████▏| 3698/4000 [18:34<01:14,  4.08it/s]

train:   5%|▌         | 183/3515 [00:48<13:36,  4.08it/s]
overall:  92%|█████████▏| 3699/4000 [18:34<01:15,  3.98it/s]

train:   5%|▌         | 184/3515 [00:49<13:57,  3.98it/s]train loss: 0.5212406516075134
train per: 0.15520985789150654

overall:  92%|█████████▎| 3700/4000 [18:34<01:18,  3.84it/s]

train:   5%|▌         | 185/3515 [00:49<14:27,  3.84it/s]
overall:  93%|█████████▎| 3701/4000 [18:35<01:14,  4.03it/s]

train:   5%|▌         | 186/3515 [00:49<13:46,  4.03it/s]
overall:  93%|█████████▎| 3702/4000 [18:35<01:15,  3.95it/s]

train:   5%|▌         | 187/3515 [00:49<14:02,  3.95it/s]
overall:  93%|█████████▎| 3703/4000 [18:35<01:16,  3.86it/s]

train:   5%|▌         | 188/3515 [00:50<14:21,  3.86it/s]
overall:  93%|█████████▎| 3704/4000 [18:35<01:17,  3.84it/s]

train:   5%|▌         | 189/3515 [00:50<14:27,  3.84it/s]
overall:  93%|█████████▎| 3705/4000 [18:36<01:31,  3.22it/s]

train:   5%|▌         | 190/3515 [00:50<17:12,  3.22it/s]
overall:  93%|█████████▎| 3706/4000 [18:36<01:27,  3.36it/s]

train:   5%|▌         | 191/3515 [00:51<16:29,  3.36it/s]
overall:  93%|█████████▎| 3707/4000 [18:36<01:24,  3.47it/s]

train:   5%|▌         | 192/3515 [00:51<15:57,  3.47it/s]
overall:  93%|█████████▎| 3708/4000 [18:37<01:23,  3.51it/s]

train:   5%|▌         | 193/3515 [00:51<15:47,  3.51it/s]
overall:  93%|█████████▎| 3709/4000 [18:37<01:21,  3.59it/s]

train:   6%|▌         | 194/3515 [00:52<15:24,  3.59it/s]
overall:  93%|█████████▎| 3710/4000 [18:37<01:20,  3.62it/s]

train:   6%|▌         | 195/3515 [00:52<15:17,  3.62it/s]
overall:  93%|█████████▎| 3711/4000 [18:38<01:19,  3.65it/s]

train:   6%|▌         | 196/3515 [00:52<15:10,  3.65it/s]
overall:  93%|█████████▎| 3712/4000 [18:38<01:14,  3.84it/s]

train:   6%|▌         | 197/3515 [00:52<14:23,  3.84it/s]
overall:  93%|█████████▎| 3713/4000 [18:38<01:14,  3.86it/s]

train:   6%|▌         | 198/3515 [00:53<14:19,  3.86it/s]
overall:  93%|█████████▎| 3714/4000 [18:38<01:14,  3.84it/s]

train:   6%|▌         | 199/3515 [00:53<14:24,  3.84it/s]
overall:  93%|█████████▎| 3715/4000 [18:38<01:09,  4.07it/s]

train:   6%|▌         | 200/3515 [00:53<13:33,  4.07it/s]
overall:  93%|█████████▎| 3716/4000 [18:39<01:12,  3.92it/s]

train:   6%|▌         | 201/3515 [00:53<14:05,  3.92it/s]
overall:  93%|█████████▎| 3717/4000 [18:39<01:26,  3.26it/s]

train:   6%|▌         | 202/3515 [00:54<16:55,  3.26it/s]
overall:  93%|█████████▎| 3718/4000 [18:39<01:23,  3.39it/s]

train:   6%|▌         | 203/3515 [00:54<16:18,  3.39it/s]train loss: 0.5429618954658508
train per: 0.15752610998366579

overall:  93%|█████████▎| 3720/4000 [18:40<00:58,  4.79it/s]

train:   6%|▌         | 205/3515 [00:54<11:31,  4.79it/s]
overall:  93%|█████████▎| 3721/4000 [18:40<01:03,  4.40it/s]

train:   6%|▌         | 206/3515 [00:55<12:31,  4.40it/s]
overall:  93%|█████████▎| 3722/4000 [18:40<01:05,  4.23it/s]

train:   6%|▌         | 207/3515 [00:55<13:01,  4.23it/s]
overall:  93%|█████████▎| 3723/4000 [18:40<01:08,  4.06it/s]

train:   6%|▌         | 208/3515 [00:55<13:33,  4.06it/s]
overall:  93%|█████████▎| 3724/4000 [18:41<01:04,  4.27it/s]

train:   6%|▌         | 209/3515 [00:55<12:53,  4.27it/s]
overall:  93%|█████████▎| 3725/4000 [18:41<01:07,  4.08it/s]

train:   6%|▌         | 210/3515 [00:56<13:30,  4.08it/s]
overall:  93%|█████████▎| 3726/4000 [18:41<01:10,  3.90it/s]

train:   6%|▌         | 211/3515 [00:56<14:06,  3.90it/s]
overall:  93%|█████████▎| 3727/4000 [18:42<01:09,  3.92it/s]

train:   6%|▌         | 212/3515 [00:56<14:03,  3.92it/s]
overall:  93%|█████████▎| 3728/4000 [18:42<01:09,  3.92it/s]

train:   6%|▌         | 213/3515 [00:56<14:02,  3.92it/s]
overall:  93%|█████████▎| 3729/4000 [18:42<01:18,  3.44it/s]

train:   6%|▌         | 214/3515 [00:57<16:00,  3.44it/s]
overall:  93%|█████████▎| 3730/4000 [18:42<01:03,  4.26it/s]

train:   6%|▌         | 215/3515 [00:57<12:54,  4.26it/s]
overall:  93%|█████████▎| 3731/4000 [18:42<01:00,  4.42it/s]

train:   6%|▌         | 216/3515 [00:57<12:27,  4.42it/s]
overall:  93%|█████████▎| 3732/4000 [18:43<01:04,  4.17it/s]

train:   6%|▌         | 217/3515 [00:57<13:10,  4.17it/s]
overall:  93%|█████████▎| 3733/4000 [18:43<01:04,  4.14it/s]

train:   6%|▌         | 218/3515 [00:58<13:16,  4.14it/s]
overall:  93%|█████████▎| 3734/4000 [18:43<01:05,  4.06it/s]

train:   6%|▌         | 219/3515 [00:58<13:31,  4.06it/s]
overall:  93%|█████████▎| 3735/4000 [18:43<01:06,  3.96it/s]

train:   6%|▋         | 220/3515 [00:58<13:51,  3.96it/s]
overall:  93%|█████████▎| 3736/4000 [18:44<01:06,  4.00it/s]

train:   6%|▋         | 221/3515 [00:58<13:43,  4.00it/s]
overall:  93%|█████████▎| 3737/4000 [18:44<01:05,  4.02it/s]

train:   6%|▋         | 222/3515 [00:59<13:39,  4.02it/s]
overall:  93%|█████████▎| 3738/4000 [18:44<01:05,  3.98it/s]

train:   6%|▋         | 223/3515 [00:59<13:48,  3.98it/s]train loss: 0.5614560842514038
train per: 0.16436871423890304

overall:  94%|█████████▎| 3740/4000 [18:45<00:57,  4.56it/s]

train:   6%|▋         | 225/3515 [00:59<12:01,  4.56it/s]
overall:  94%|█████████▎| 3741/4000 [18:45<01:10,  3.67it/s]

train:   6%|▋         | 226/3515 [01:00<14:55,  3.67it/s]
overall:  94%|█████████▎| 3742/4000 [18:45<01:09,  3.70it/s]

train:   6%|▋         | 227/3515 [01:00<14:47,  3.70it/s]
overall:  94%|█████████▎| 3743/4000 [18:46<01:08,  3.75it/s]

train:   6%|▋         | 228/3515 [01:00<14:35,  3.75it/s]
overall:  94%|█████████▎| 3744/4000 [18:46<01:06,  3.85it/s]

train:   7%|▋         | 229/3515 [01:00<14:13,  3.85it/s]
overall:  94%|█████████▎| 3745/4000 [18:46<01:05,  3.87it/s]

train:   7%|▋         | 230/3515 [01:01<14:09,  3.87it/s]
overall:  94%|█████████▎| 3746/4000 [18:46<01:06,  3.85it/s]

train:   7%|▋         | 231/3515 [01:01<14:13,  3.85it/s]
overall:  94%|█████████▎| 3747/4000 [18:47<01:05,  3.88it/s]

train:   7%|▋         | 232/3515 [01:01<14:05,  3.88it/s]
overall:  94%|█████████▎| 3748/4000 [18:47<01:06,  3.81it/s]

train:   7%|▋         | 233/3515 [01:01<14:22,  3.81it/s]
overall:  94%|█████████▎| 3749/4000 [18:47<01:04,  3.87it/s]

train:   7%|▋         | 234/3515 [01:02<14:08,  3.87it/s]
overall:  94%|█████████▍| 3750/4000 [18:47<01:04,  3.88it/s]

train:   7%|▋         | 235/3515 [01:02<14:06,  3.88it/s]
overall:  94%|█████████▍| 3751/4000 [18:48<01:04,  3.84it/s]

train:   7%|▋         | 236/3515 [01:02<14:14,  3.84it/s]
overall:  94%|█████████▍| 3752/4000 [18:48<01:04,  3.82it/s]

train:   7%|▋         | 237/3515 [01:02<14:18,  3.82it/s]
overall:  94%|█████████▍| 3753/4000 [18:48<01:18,  3.13it/s]

train:   7%|▋         | 238/3515 [01:03<17:27,  3.13it/s]
overall:  94%|█████████▍| 3754/4000 [18:49<01:15,  3.27it/s]

train:   7%|▋         | 239/3515 [01:03<16:42,  3.27it/s]
overall:  94%|█████████▍| 3755/4000 [18:49<01:11,  3.45it/s]

train:   7%|▋         | 240/3515 [01:03<15:49,  3.45it/s]
overall:  94%|█████████▍| 3756/4000 [18:49<01:09,  3.51it/s]

train:   7%|▋         | 241/3515 [01:04<15:33,  3.51it/s]
overall:  94%|█████████▍| 3757/4000 [18:49<01:08,  3.56it/s]

train:   7%|▋         | 242/3515 [01:04<15:20,  3.56it/s]
overall:  94%|█████████▍| 3758/4000 [18:50<01:07,  3.61it/s]

train:   7%|▋         | 243/3515 [01:04<15:06,  3.61it/s]
overall:  94%|█████████▍| 3759/4000 [18:50<01:05,  3.66it/s]

train:   7%|▋         | 244/3515 [01:04<14:54,  3.66it/s]train loss: 0.5452902913093567
train per: 0.1596179137537205

overall:  94%|█████████▍| 3760/4000 [18:50<01:03,  3.76it/s]

train:   7%|▋         | 245/3515 [01:05<14:29,  3.76it/s]
overall:  94%|█████████▍| 3761/4000 [18:50<01:03,  3.79it/s]

train:   7%|▋         | 246/3515 [01:05<14:22,  3.79it/s]
overall:  94%|█████████▍| 3762/4000 [18:51<01:02,  3.78it/s]

train:   7%|▋         | 247/3515 [01:05<14:24,  3.78it/s]
overall:  94%|█████████▍| 3763/4000 [18:51<01:03,  3.73it/s]

train:   7%|▋         | 248/3515 [01:06<14:35,  3.73it/s]
overall:  94%|█████████▍| 3764/4000 [18:51<01:03,  3.70it/s]

train:   7%|▋         | 249/3515 [01:06<14:42,  3.70it/s]
overall:  94%|█████████▍| 3765/4000 [18:52<01:13,  3.22it/s]

train:   7%|▋         | 250/3515 [01:06<16:54,  3.22it/s]
overall:  94%|█████████▍| 3766/4000 [18:52<01:07,  3.48it/s]

train:   7%|▋         | 251/3515 [01:06<15:38,  3.48it/s]
overall:  94%|█████████▍| 3767/4000 [18:52<00:55,  4.18it/s]

train:   7%|▋         | 252/3515 [01:07<13:01,  4.18it/s]
overall:  94%|█████████▍| 3768/4000 [18:52<00:56,  4.09it/s]

train:   7%|▋         | 253/3515 [01:07<13:17,  4.09it/s]
overall:  94%|█████████▍| 3769/4000 [18:53<00:57,  3.99it/s]

train:   7%|▋         | 254/3515 [01:07<13:36,  3.99it/s]
overall:  94%|█████████▍| 3770/4000 [18:53<00:59,  3.90it/s]

train:   7%|▋         | 255/3515 [01:07<13:56,  3.90it/s]
overall:  94%|█████████▍| 3771/4000 [18:53<00:58,  3.91it/s]

train:   7%|▋         | 256/3515 [01:08<13:53,  3.91it/s]
overall:  94%|█████████▍| 3773/4000 [18:53<00:41,  5.46it/s]

train:   7%|▋         | 258/3515 [01:08<09:56,  5.46it/s]
overall:  94%|█████████▍| 3774/4000 [18:53<00:36,  6.15it/s]

train:   7%|▋         | 259/3515 [01:08<08:49,  6.15it/s]
overall:  94%|█████████▍| 3775/4000 [18:54<00:43,  5.22it/s]

train:   7%|▋         | 260/3515 [01:08<10:23,  5.22it/s]
overall:  94%|█████████▍| 3776/4000 [18:54<00:47,  4.75it/s]

train:   7%|▋         | 261/3515 [01:08<11:24,  4.75it/s]
overall:  94%|█████████▍| 3777/4000 [18:54<00:59,  3.74it/s]

train:   7%|▋         | 262/3515 [01:09<14:29,  3.74it/s]
overall:  94%|█████████▍| 3778/4000 [18:55<00:58,  3.82it/s]

train:   7%|▋         | 263/3515 [01:09<14:10,  3.82it/s]
overall:  94%|█████████▍| 3779/4000 [18:55<00:58,  3.81it/s]

train:   8%|▊         | 264/3515 [01:09<14:13,  3.81it/s]train loss: 0.5566369295120239
train per: 0.16211831045855943

overall:  94%|█████████▍| 3780/4000 [18:55<00:59,  3.71it/s]

train:   8%|▊         | 265/3515 [01:10<14:37,  3.70it/s]
overall:  95%|█████████▍| 3781/4000 [18:55<00:57,  3.84it/s]

train:   8%|▊         | 266/3515 [01:10<14:06,  3.84it/s]
overall:  95%|█████████▍| 3782/4000 [18:56<00:57,  3.81it/s]

train:   8%|▊         | 267/3515 [01:10<14:12,  3.81it/s]
overall:  95%|█████████▍| 3783/4000 [18:56<00:56,  3.84it/s]

train:   8%|▊         | 268/3515 [01:10<14:04,  3.85it/s]
overall:  95%|█████████▍| 3785/4000 [18:56<00:46,  4.65it/s]

train:   8%|▊         | 270/3515 [01:11<11:38,  4.65it/s]
overall:  95%|█████████▍| 3787/4000 [18:57<00:42,  5.00it/s]

train:   8%|▊         | 272/3515 [01:11<10:48,  5.00it/s]
overall:  95%|█████████▍| 3788/4000 [18:57<00:45,  4.70it/s]

train:   8%|▊         | 273/3515 [01:11<11:29,  4.70it/s]
overall:  95%|█████████▍| 3789/4000 [18:57<00:54,  3.89it/s]

train:   8%|▊         | 274/3515 [01:12<13:53,  3.89it/s]
overall:  95%|█████████▍| 3790/4000 [18:57<00:54,  3.88it/s]

train:   8%|▊         | 275/3515 [01:12<13:54,  3.88it/s]
overall:  95%|█████████▍| 3791/4000 [18:58<00:54,  3.85it/s]

train:   8%|▊         | 276/3515 [01:12<14:02,  3.85it/s]
overall:  95%|█████████▍| 3792/4000 [18:58<00:54,  3.78it/s]

train:   8%|▊         | 277/3515 [01:13<14:15,  3.78it/s]
overall:  95%|█████████▍| 3793/4000 [18:58<00:55,  3.74it/s]

train:   8%|▊         | 278/3515 [01:13<14:26,  3.74it/s]
overall:  95%|█████████▍| 3794/4000 [18:58<00:51,  4.01it/s]

train:   8%|▊         | 279/3515 [01:13<13:26,  4.01it/s]
overall:  95%|█████████▍| 3795/4000 [18:59<00:49,  4.12it/s]

train:   8%|▊         | 280/3515 [01:13<13:05,  4.12it/s]
overall:  95%|█████████▍| 3796/4000 [18:59<00:51,  3.96it/s]

train:   8%|▊         | 281/3515 [01:14<13:36,  3.96it/s]
overall:  95%|█████████▍| 3797/4000 [18:59<00:52,  3.88it/s]

train:   8%|▊         | 282/3515 [01:14<13:53,  3.88it/s]
overall:  95%|█████████▍| 3798/4000 [19:00<00:52,  3.83it/s]

train:   8%|▊         | 283/3515 [01:14<14:03,  3.83it/s]
overall:  95%|█████████▍| 3799/4000 [19:00<00:52,  3.80it/s]

train:   8%|▊         | 284/3515 [01:14<14:11,  3.80it/s]train loss: 0.5871920585632324
train per: 0.17269395391752312

overall:  95%|█████████▌| 3800/4000 [19:00<00:47,  4.24it/s]

train:   8%|▊         | 285/3515 [01:15<12:41,  4.24it/s]
overall:  95%|█████████▌| 3801/4000 [19:00<00:58,  3.39it/s]

train:   8%|▊         | 286/3515 [01:15<15:53,  3.39it/s]
overall:  95%|█████████▌| 3802/4000 [19:01<00:56,  3.51it/s]

train:   8%|▊         | 287/3515 [01:15<15:20,  3.51it/s]
overall:  95%|█████████▌| 3803/4000 [19:01<00:51,  3.85it/s]

train:   8%|▊         | 288/3515 [01:15<13:58,  3.85it/s]
overall:  95%|█████████▌| 3804/4000 [19:01<00:51,  3.81it/s]

train:   8%|▊         | 289/3515 [01:16<14:05,  3.81it/s]
overall:  95%|█████████▌| 3805/4000 [19:01<00:51,  3.77it/s]

train:   8%|▊         | 290/3515 [01:16<14:14,  3.77it/s]
overall:  95%|█████████▌| 3806/4000 [19:02<00:50,  3.85it/s]

train:   8%|▊         | 291/3515 [01:16<13:58,  3.85it/s]
overall:  95%|█████████▌| 3807/4000 [19:02<00:41,  4.64it/s]

train:   8%|▊         | 292/3515 [01:16<11:35,  4.64it/s]
overall:  95%|█████████▌| 3808/4000 [19:02<00:44,  4.27it/s]

train:   8%|▊         | 293/3515 [01:17<12:34,  4.27it/s]
overall:  95%|█████████▌| 3809/4000 [19:02<00:44,  4.25it/s]

train:   8%|▊         | 294/3515 [01:17<12:37,  4.25it/s]
overall:  95%|█████████▌| 3810/4000 [19:02<00:43,  4.38it/s]

train:   8%|▊         | 295/3515 [01:17<12:15,  4.38it/s]
overall:  95%|█████████▌| 3811/4000 [19:03<00:45,  4.18it/s]

train:   8%|▊         | 296/3515 [01:17<12:50,  4.18it/s]
overall:  95%|█████████▌| 3812/4000 [19:03<00:46,  4.03it/s]

train:   8%|▊         | 297/3515 [01:18<13:17,  4.03it/s]
overall:  95%|█████████▌| 3813/4000 [19:03<00:54,  3.44it/s]

train:   8%|▊         | 298/3515 [01:18<15:35,  3.44it/s]
overall:  95%|█████████▌| 3814/4000 [19:04<00:52,  3.52it/s]

train:   9%|▊         | 299/3515 [01:18<15:12,  3.52it/s]
overall:  95%|█████████▌| 3815/4000 [19:04<00:51,  3.58it/s]

train:   9%|▊         | 300/3515 [01:19<14:57,  3.58it/s]
overall:  95%|█████████▌| 3816/4000 [19:04<00:41,  4.39it/s]

train:   9%|▊         | 301/3515 [01:19<12:12,  4.39it/s]
overall:  95%|█████████▌| 3817/4000 [19:04<00:44,  4.12it/s]

train:   9%|▊         | 302/3515 [01:19<13:00,  4.12it/s]
overall:  95%|█████████▌| 3819/4000 [19:05<00:36,  4.90it/s]

train:   9%|▊         | 304/3515 [01:19<10:55,  4.90it/s]train loss: 0.5977693796157837
train per: 0.16926070038910507

overall:  96%|█████████▌| 3820/4000 [19:05<00:40,  4.50it/s]

train:   9%|▊         | 305/3515 [01:19<11:54,  4.49it/s]
overall:  96%|█████████▌| 3821/4000 [19:05<00:40,  4.42it/s]

train:   9%|▊         | 306/3515 [01:20<12:06,  4.42it/s]
overall:  96%|█████████▌| 3822/4000 [19:05<00:35,  4.95it/s]

train:   9%|▊         | 307/3515 [01:20<10:48,  4.95it/s]
overall:  96%|█████████▌| 3823/4000 [19:06<00:37,  4.69it/s]

train:   9%|▉         | 308/3515 [01:20<11:24,  4.69it/s]
overall:  96%|█████████▌| 3824/4000 [19:06<00:40,  4.38it/s]

train:   9%|▉         | 309/3515 [01:20<12:12,  4.38it/s]
overall:  96%|█████████▌| 3825/4000 [19:06<00:50,  3.44it/s]

train:   9%|▉         | 310/3515 [01:21<15:30,  3.44it/s]
overall:  96%|█████████▌| 3826/4000 [19:07<00:49,  3.55it/s]

train:   9%|▉         | 311/3515 [01:21<15:02,  3.55it/s]
overall:  96%|█████████▌| 3827/4000 [19:07<00:47,  3.61it/s]

train:   9%|▉         | 312/3515 [01:21<14:47,  3.61it/s]
overall:  96%|█████████▌| 3828/4000 [19:07<00:48,  3.58it/s]

train:   9%|▉         | 313/3515 [01:22<14:54,  3.58it/s]
overall:  96%|█████████▌| 3829/4000 [19:07<00:46,  3.67it/s]

train:   9%|▉         | 314/3515 [01:22<14:31,  3.67it/s]
overall:  96%|█████████▌| 3830/4000 [19:08<00:45,  3.73it/s]

train:   9%|▉         | 315/3515 [01:22<14:17,  3.73it/s]
overall:  96%|█████████▌| 3831/4000 [19:08<00:43,  3.87it/s]

train:   9%|▉         | 316/3515 [01:22<13:46,  3.87it/s]
overall:  96%|█████████▌| 3832/4000 [19:08<00:44,  3.77it/s]

train:   9%|▉         | 317/3515 [01:23<14:08,  3.77it/s]
overall:  96%|█████████▌| 3833/4000 [19:08<00:43,  3.86it/s]

train:   9%|▉         | 318/3515 [01:23<13:49,  3.86it/s]
overall:  96%|█████████▌| 3834/4000 [19:09<00:43,  3.81it/s]

train:   9%|▉         | 319/3515 [01:23<13:58,  3.81it/s]
overall:  96%|█████████▌| 3835/4000 [19:09<00:41,  3.94it/s]

train:   9%|▉         | 320/3515 [01:23<13:30,  3.94it/s]
overall:  96%|█████████▌| 3836/4000 [19:09<00:42,  3.90it/s]

train:   9%|▉         | 321/3515 [01:24<13:39,  3.90it/s]
overall:  96%|█████████▌| 3837/4000 [19:10<00:49,  3.29it/s]

train:   9%|▉         | 322/3515 [01:24<16:10,  3.29it/s]
overall:  96%|█████████▌| 3838/4000 [19:10<00:46,  3.48it/s]

train:   9%|▉         | 323/3515 [01:24<15:15,  3.48it/s]
overall:  96%|█████████▌| 3839/4000 [19:10<00:45,  3.52it/s]

train:   9%|▉         | 324/3515 [01:25<15:07,  3.52it/s]train loss: 0.5591119527816772
train per: 0.16512936947164278

overall:  96%|█████████▌| 3840/4000 [19:10<00:44,  3.63it/s]

train:   9%|▉         | 325/3515 [01:25<14:40,  3.62it/s]
overall:  96%|█████████▌| 3841/4000 [19:11<00:40,  3.91it/s]

train:   9%|▉         | 326/3515 [01:25<13:34,  3.91it/s]
overall:  96%|█████████▌| 3842/4000 [19:11<00:40,  3.90it/s]

train:   9%|▉         | 327/3515 [01:25<13:38,  3.90it/s]
overall:  96%|█████████▌| 3843/4000 [19:11<00:41,  3.82it/s]

train:   9%|▉         | 328/3515 [01:26<13:54,  3.82it/s]
overall:  96%|█████████▌| 3844/4000 [19:11<00:41,  3.74it/s]

train:   9%|▉         | 329/3515 [01:26<14:12,  3.74it/s]
overall:  96%|█████████▌| 3845/4000 [19:12<00:41,  3.71it/s]

train:   9%|▉         | 330/3515 [01:26<14:17,  3.71it/s]
overall:  96%|█████████▌| 3846/4000 [19:12<00:38,  3.96it/s]

train:   9%|▉         | 331/3515 [01:26<13:24,  3.96it/s]
overall:  96%|█████████▌| 3847/4000 [19:12<00:38,  3.94it/s]

train:   9%|▉         | 332/3515 [01:27<13:27,  3.94it/s]
overall:  96%|█████████▌| 3848/4000 [19:12<00:38,  3.90it/s]

train:   9%|▉         | 333/3515 [01:27<13:36,  3.90it/s]
overall:  96%|█████████▌| 3849/4000 [19:13<00:46,  3.27it/s]

train:  10%|▉         | 334/3515 [01:27<16:13,  3.27it/s]
overall:  96%|█████████▋| 3850/4000 [19:13<00:43,  3.45it/s]

train:  10%|▉         | 335/3515 [01:28<15:20,  3.45it/s]
overall:  96%|█████████▋| 3851/4000 [19:13<00:42,  3.53it/s]

train:  10%|▉         | 336/3515 [01:28<15:01,  3.53it/s]
overall:  96%|█████████▋| 3852/4000 [19:14<00:41,  3.53it/s]

train:  10%|▉         | 337/3515 [01:28<15:00,  3.53it/s]
overall:  96%|█████████▋| 3854/4000 [19:14<00:33,  4.32it/s]

train:  10%|▉         | 339/3515 [01:28<12:14,  4.32it/s]
overall:  96%|█████████▋| 3855/4000 [19:14<00:32,  4.48it/s]

train:  10%|▉         | 340/3515 [01:29<11:48,  4.48it/s]
overall:  96%|█████████▋| 3856/4000 [19:14<00:33,  4.25it/s]

train:  10%|▉         | 341/3515 [01:29<12:25,  4.25it/s]
overall:  96%|█████████▋| 3857/4000 [19:15<00:34,  4.12it/s]

train:  10%|▉         | 342/3515 [01:29<12:50,  4.12it/s]
overall:  96%|█████████▋| 3858/4000 [19:15<00:35,  4.00it/s]

train:  10%|▉         | 343/3515 [01:29<13:13,  4.00it/s]
overall:  96%|█████████▋| 3859/4000 [19:15<00:33,  4.15it/s]

train:  10%|▉         | 344/3515 [01:30<12:43,  4.15it/s]train loss: 0.5406535267829895
train per: 0.15666064800882737

overall:  96%|█████████▋| 3860/4000 [19:15<00:34,  4.00it/s]

train:  10%|▉         | 345/3515 [01:30<13:12,  4.00it/s]
overall:  97%|█████████▋| 3861/4000 [19:16<00:41,  3.35it/s]

train:  10%|▉         | 346/3515 [01:30<15:45,  3.35it/s]
overall:  97%|█████████▋| 3863/4000 [19:16<00:32,  4.16it/s]

train:  10%|▉         | 348/3515 [01:31<12:41,  4.16it/s]
overall:  97%|█████████▋| 3864/4000 [19:16<00:32,  4.20it/s]

train:  10%|▉         | 349/3515 [01:31<12:33,  4.20it/s]
overall:  97%|█████████▋| 3865/4000 [19:17<00:33,  4.04it/s]

train:  10%|▉         | 350/3515 [01:31<13:02,  4.04it/s]
overall:  97%|█████████▋| 3866/4000 [19:17<00:33,  3.97it/s]

train:  10%|▉         | 351/3515 [01:31<13:16,  3.97it/s]
overall:  97%|█████████▋| 3867/4000 [19:17<00:33,  4.00it/s]

train:  10%|█         | 352/3515 [01:32<13:10,  4.00it/s]
overall:  97%|█████████▋| 3868/4000 [19:17<00:33,  3.92it/s]

train:  10%|█         | 353/3515 [01:32<13:26,  3.92it/s]
overall:  97%|█████████▋| 3869/4000 [19:18<00:34,  3.80it/s]

train:  10%|█         | 354/3515 [01:32<13:52,  3.80it/s]
overall:  97%|█████████▋| 3870/4000 [19:18<00:34,  3.74it/s]

train:  10%|█         | 355/3515 [01:33<14:04,  3.74it/s]
overall:  97%|█████████▋| 3871/4000 [19:18<00:34,  3.73it/s]

train:  10%|█         | 356/3515 [01:33<14:06,  3.73it/s]
overall:  97%|█████████▋| 3872/4000 [19:19<00:34,  3.73it/s]

train:  10%|█         | 357/3515 [01:33<14:07,  3.73it/s]
overall:  97%|█████████▋| 3873/4000 [19:19<00:34,  3.65it/s]

train:  10%|█         | 358/3515 [01:33<14:24,  3.65it/s]
overall:  97%|█████████▋| 3874/4000 [19:19<00:34,  3.66it/s]

train:  10%|█         | 359/3515 [01:34<14:22,  3.66it/s]
overall:  97%|█████████▋| 3875/4000 [19:19<00:33,  3.75it/s]

train:  10%|█         | 360/3515 [01:34<14:02,  3.75it/s]
overall:  97%|█████████▋| 3876/4000 [19:20<00:33,  3.70it/s]

train:  10%|█         | 361/3515 [01:34<14:11,  3.70it/s]
overall:  97%|█████████▋| 3877/4000 [19:20<00:32,  3.83it/s]

train:  10%|█         | 362/3515 [01:34<13:43,  3.83it/s]
overall:  97%|█████████▋| 3878/4000 [19:20<00:32,  3.77it/s]

train:  10%|█         | 363/3515 [01:35<13:56,  3.77it/s]
overall:  97%|█████████▋| 3879/4000 [19:20<00:28,  4.17it/s]

train:  10%|█         | 364/3515 [01:35<12:34,  4.17it/s]train loss: 0.568781852722168
train per: 0.16455999233936608

overall:  97%|█████████▋| 3880/4000 [19:21<00:30,  3.96it/s]

train:  10%|█         | 365/3515 [01:35<13:16,  3.96it/s]
overall:  97%|█████████▋| 3881/4000 [19:21<00:30,  3.93it/s]

train:  10%|█         | 366/3515 [01:35<13:20,  3.94it/s]
overall:  97%|█████████▋| 3882/4000 [19:21<00:30,  3.83it/s]

train:  10%|█         | 367/3515 [01:36<13:42,  3.83it/s]
overall:  97%|█████████▋| 3883/4000 [19:21<00:30,  3.84it/s]

train:  10%|█         | 368/3515 [01:36<13:39,  3.84it/s]
overall:  97%|█████████▋| 3884/4000 [19:22<00:30,  3.84it/s]

train:  10%|█         | 369/3515 [01:36<13:38,  3.84it/s]
overall:  97%|█████████▋| 3885/4000 [19:22<00:34,  3.33it/s]

train:  11%|█         | 370/3515 [01:37<15:43,  3.33it/s]
overall:  97%|█████████▋| 3886/4000 [19:22<00:33,  3.43it/s]

train:  11%|█         | 371/3515 [01:37<15:15,  3.44it/s]
overall:  97%|█████████▋| 3887/4000 [19:23<00:31,  3.58it/s]

train:  11%|█         | 372/3515 [01:37<14:36,  3.58it/s]
overall:  97%|█████████▋| 3888/4000 [19:23<00:31,  3.57it/s]

train:  11%|█         | 373/3515 [01:37<14:40,  3.57it/s]
overall:  97%|█████████▋| 3889/4000 [19:23<00:30,  3.67it/s]

train:  11%|█         | 374/3515 [01:38<14:14,  3.67it/s]
overall:  97%|█████████▋| 3890/4000 [19:23<00:29,  3.75it/s]

train:  11%|█         | 375/3515 [01:38<13:58,  3.75it/s]
overall:  97%|█████████▋| 3891/4000 [19:24<00:28,  3.78it/s]

train:  11%|█         | 376/3515 [01:38<13:51,  3.78it/s]
overall:  97%|█████████▋| 3892/4000 [19:24<00:28,  3.76it/s]

train:  11%|█         | 377/3515 [01:38<13:55,  3.76it/s]
overall:  97%|█████████▋| 3893/4000 [19:24<00:24,  4.30it/s]

train:  11%|█         | 378/3515 [01:39<12:09,  4.30it/s]
overall:  97%|█████████▋| 3894/4000 [19:24<00:26,  4.03it/s]

train:  11%|█         | 379/3515 [01:39<12:57,  4.03it/s]
overall:  97%|█████████▋| 3895/4000 [19:25<00:26,  3.94it/s]

train:  11%|█         | 380/3515 [01:39<13:16,  3.94it/s]
overall:  97%|█████████▋| 3896/4000 [19:25<00:26,  3.90it/s]

train:  11%|█         | 381/3515 [01:39<13:22,  3.90it/s]
overall:  97%|█████████▋| 3897/4000 [19:25<00:30,  3.36it/s]

train:  11%|█         | 382/3515 [01:40<15:32,  3.36it/s]
overall:  97%|█████████▋| 3898/4000 [19:26<00:29,  3.45it/s]

train:  11%|█         | 383/3515 [01:40<15:07,  3.45it/s]
overall:  97%|█████████▋| 3899/4000 [19:26<00:28,  3.52it/s]

train:  11%|█         | 384/3515 [01:40<14:49,  3.52it/s]train loss: 0.5424734354019165
train per: 0.15873659547625485

overall:  98%|█████████▊| 3900/4000 [19:26<00:28,  3.55it/s]

train:  11%|█         | 385/3515 [01:41<14:41,  3.55it/s]
overall:  98%|█████████▊| 3901/4000 [19:26<00:27,  3.64it/s]

train:  11%|█         | 386/3515 [01:41<14:18,  3.64it/s]
overall:  98%|█████████▊| 3902/4000 [19:27<00:26,  3.68it/s]

train:  11%|█         | 387/3515 [01:41<14:10,  3.68it/s]
overall:  98%|█████████▊| 3903/4000 [19:27<00:21,  4.53it/s]

train:  11%|█         | 388/3515 [01:41<11:30,  4.53it/s]
overall:  98%|█████████▊| 3904/4000 [19:27<00:22,  4.28it/s]

train:  11%|█         | 389/3515 [01:42<12:10,  4.28it/s]
overall:  98%|█████████▊| 3905/4000 [19:27<00:23,  4.05it/s]

train:  11%|█         | 390/3515 [01:42<12:51,  4.05it/s]
overall:  98%|█████████▊| 3906/4000 [19:28<00:24,  3.86it/s]

train:  11%|█         | 391/3515 [01:42<13:28,  3.86it/s]
overall:  98%|█████████▊| 3907/4000 [19:28<00:24,  3.81it/s]

train:  11%|█         | 392/3515 [01:42<13:40,  3.81it/s]
overall:  98%|█████████▊| 3908/4000 [19:28<00:23,  3.85it/s]

train:  11%|█         | 393/3515 [01:43<13:31,  3.85it/s]
overall:  98%|█████████▊| 3909/4000 [19:28<00:27,  3.30it/s]

train:  11%|█         | 394/3515 [01:43<15:44,  3.30it/s]
overall:  98%|█████████▊| 3910/4000 [19:29<00:26,  3.42it/s]

train:  11%|█         | 395/3515 [01:43<15:13,  3.42it/s]
overall:  98%|█████████▊| 3911/4000 [19:29<00:25,  3.56it/s]

train:  11%|█▏        | 396/3515 [01:44<14:37,  3.56it/s]
overall:  98%|█████████▊| 3912/4000 [19:29<00:23,  3.69it/s]

train:  11%|█▏        | 397/3515 [01:44<14:05,  3.69it/s]
overall:  98%|█████████▊| 3913/4000 [19:29<00:23,  3.68it/s]

train:  11%|█▏        | 398/3515 [01:44<14:06,  3.68it/s]
overall:  98%|█████████▊| 3914/4000 [19:30<00:20,  4.13it/s]

train:  11%|█▏        | 399/3515 [01:44<12:34,  4.13it/s]
overall:  98%|█████████▊| 3915/4000 [19:30<00:20,  4.12it/s]

train:  11%|█▏        | 400/3515 [01:44<12:35,  4.12it/s]
overall:  98%|█████████▊| 3916/4000 [19:30<00:21,  3.99it/s]

train:  11%|█▏        | 401/3515 [01:45<13:00,  3.99it/s]
overall:  98%|█████████▊| 3917/4000 [19:30<00:17,  4.63it/s]

train:  11%|█▏        | 402/3515 [01:45<11:12,  4.63it/s]
overall:  98%|█████████▊| 3918/4000 [19:31<00:19,  4.21it/s]

train:  11%|█▏        | 403/3515 [01:45<12:18,  4.21it/s]
overall:  98%|█████████▊| 3919/4000 [19:31<00:19,  4.19it/s]

train:  11%|█▏        | 404/3515 [01:45<12:22,  4.19it/s]train loss: 0.5305823683738708
train per: 0.1607917105553242

overall:  98%|█████████▊| 3920/4000 [19:31<00:20,  4.00it/s]

train:  12%|█▏        | 405/3515 [01:46<12:58,  4.00it/s]
overall:  98%|█████████▊| 3921/4000 [19:32<00:22,  3.44it/s]

train:  12%|█▏        | 406/3515 [01:46<15:04,  3.44it/s]
overall:  98%|█████████▊| 3922/4000 [19:32<00:21,  3.58it/s]

train:  12%|█▏        | 407/3515 [01:46<14:28,  3.58it/s]
overall:  98%|█████████▊| 3923/4000 [19:32<00:18,  4.20it/s]

train:  12%|█▏        | 408/3515 [01:46<12:19,  4.20it/s]
overall:  98%|█████████▊| 3924/4000 [19:32<00:17,  4.23it/s]

train:  12%|█▏        | 409/3515 [01:47<12:14,  4.23it/s]
overall:  98%|█████████▊| 3925/4000 [19:32<00:17,  4.20it/s]

train:  12%|█▏        | 410/3515 [01:47<12:18,  4.20it/s]
overall:  98%|█████████▊| 3926/4000 [19:33<00:15,  4.83it/s]

train:  12%|█▏        | 411/3515 [01:47<10:42,  4.83it/s]
overall:  98%|█████████▊| 3927/4000 [19:33<00:15,  4.59it/s]

train:  12%|█▏        | 412/3515 [01:47<11:15,  4.59it/s]
overall:  98%|█████████▊| 3928/4000 [19:33<00:16,  4.29it/s]

train:  12%|█▏        | 413/3515 [01:48<12:03,  4.29it/s]
overall:  98%|█████████▊| 3929/4000 [19:33<00:17,  4.12it/s]

train:  12%|█▏        | 414/3515 [01:48<12:32,  4.12it/s]
overall:  98%|█████████▊| 3930/4000 [19:34<00:17,  3.98it/s]

train:  12%|█▏        | 415/3515 [01:48<12:57,  3.98it/s]
overall:  98%|█████████▊| 3931/4000 [19:34<00:16,  4.10it/s]

train:  12%|█▏        | 416/3515 [01:48<12:35,  4.10it/s]
overall:  98%|█████████▊| 3932/4000 [19:34<00:16,  4.08it/s]

train:  12%|█▏        | 417/3515 [01:49<12:39,  4.08it/s]
overall:  98%|█████████▊| 3933/4000 [19:34<00:19,  3.43it/s]

train:  12%|█▏        | 418/3515 [01:49<15:04,  3.43it/s]
overall:  98%|█████████▊| 3934/4000 [19:35<00:18,  3.52it/s]

train:  12%|█▏        | 419/3515 [01:49<14:38,  3.52it/s]
overall:  98%|█████████▊| 3935/4000 [19:35<00:18,  3.58it/s]

train:  12%|█▏        | 420/3515 [01:50<14:23,  3.58it/s]
overall:  98%|█████████▊| 3936/4000 [19:35<00:17,  3.63it/s]

train:  12%|█▏        | 421/3515 [01:50<14:11,  3.63it/s]
overall:  98%|█████████▊| 3937/4000 [19:35<00:17,  3.68it/s]

train:  12%|█▏        | 422/3515 [01:50<14:01,  3.68it/s]
overall:  98%|█████████▊| 3939/4000 [19:36<00:13,  4.43it/s]

train:  12%|█▏        | 424/3515 [01:50<11:38,  4.43it/s]train loss: 0.5280081033706665
train per: 0.15287337026467462

overall:  98%|█████████▊| 3940/4000 [19:36<00:14,  4.25it/s]

train:  12%|█▏        | 425/3515 [01:51<12:08,  4.24it/s]
overall:  99%|█████████▊| 3941/4000 [19:36<00:14,  4.10it/s]

train:  12%|█▏        | 426/3515 [01:51<12:33,  4.10it/s]
overall:  99%|█████████▊| 3942/4000 [19:37<00:14,  3.94it/s]

train:  12%|█▏        | 427/3515 [01:51<13:03,  3.94it/s]
overall:  99%|█████████▊| 3943/4000 [19:37<00:14,  3.99it/s]

train:  12%|█▏        | 428/3515 [01:51<12:53,  3.99it/s]
overall:  99%|█████████▊| 3944/4000 [19:37<00:14,  3.93it/s]

train:  12%|█▏        | 429/3515 [01:52<13:04,  3.93it/s]
overall:  99%|█████████▊| 3945/4000 [19:38<00:16,  3.33it/s]

train:  12%|█▏        | 430/3515 [01:52<15:25,  3.33it/s]
overall:  99%|█████████▊| 3946/4000 [19:38<00:15,  3.41it/s]

train:  12%|█▏        | 431/3515 [01:52<15:04,  3.41it/s]
overall:  99%|█████████▊| 3947/4000 [19:38<00:15,  3.48it/s]

train:  12%|█▏        | 432/3515 [01:53<14:45,  3.48it/s]
overall:  99%|█████████▊| 3948/4000 [19:38<00:13,  3.76it/s]

train:  12%|█▏        | 433/3515 [01:53<13:39,  3.76it/s]
overall:  99%|█████████▊| 3949/4000 [19:39<00:12,  4.00it/s]

train:  12%|█▏        | 434/3515 [01:53<12:49,  4.00it/s]
overall:  99%|█████████▉| 3950/4000 [19:39<00:12,  3.91it/s]

train:  12%|█▏        | 435/3515 [01:53<13:07,  3.91it/s]
overall:  99%|█████████▉| 3951/4000 [19:39<00:12,  3.85it/s]

train:  12%|█▏        | 436/3515 [01:54<13:19,  3.85it/s]
overall:  99%|█████████▉| 3952/4000 [19:39<00:11,  4.09it/s]

train:  12%|█▏        | 437/3515 [01:54<12:32,  4.09it/s]
overall:  99%|█████████▉| 3953/4000 [19:40<00:11,  3.96it/s]

train:  12%|█▏        | 438/3515 [01:54<12:56,  3.96it/s]
overall:  99%|█████████▉| 3954/4000 [19:40<00:11,  3.86it/s]

train:  12%|█▏        | 439/3515 [01:54<13:17,  3.86it/s]
overall:  99%|█████████▉| 3955/4000 [19:40<00:11,  3.80it/s]

train:  13%|█▎        | 440/3515 [01:55<13:29,  3.80it/s]
overall:  99%|█████████▉| 3956/4000 [19:40<00:10,  4.08it/s]

train:  13%|█▎        | 441/3515 [01:55<12:33,  4.08it/s]
overall:  99%|█████████▉| 3957/4000 [19:41<00:12,  3.56it/s]

train:  13%|█▎        | 442/3515 [01:55<14:23,  3.56it/s]
overall:  99%|█████████▉| 3959/4000 [19:41<00:09,  4.54it/s]

train:  13%|█▎        | 444/3515 [01:56<11:16,  4.54it/s]train loss: 0.5484418272972107
train per: 0.1679178190438562

overall:  99%|█████████▉| 3960/4000 [19:41<00:09,  4.29it/s]

train:  13%|█▎        | 445/3515 [01:56<11:55,  4.29it/s]
overall:  99%|█████████▉| 3961/4000 [19:42<00:09,  4.13it/s]

train:  13%|█▎        | 446/3515 [01:56<12:22,  4.13it/s]
overall:  99%|█████████▉| 3963/4000 [19:42<00:07,  4.80it/s]

train:  13%|█▎        | 448/3515 [01:56<10:39,  4.80it/s]
overall:  99%|█████████▉| 3964/4000 [19:42<00:07,  4.50it/s]

train:  13%|█▎        | 449/3515 [01:57<11:21,  4.50it/s]
overall:  99%|█████████▉| 3965/4000 [19:42<00:07,  4.85it/s]

train:  13%|█▎        | 450/3515 [01:57<10:32,  4.85it/s]
overall:  99%|█████████▉| 3966/4000 [19:43<00:07,  4.35it/s]

train:  13%|█▎        | 451/3515 [01:57<11:44,  4.35it/s]
overall:  99%|█████████▉| 3967/4000 [19:43<00:07,  4.21it/s]

train:  13%|█▎        | 452/3515 [01:57<12:07,  4.21it/s]
overall:  99%|█████████▉| 3968/4000 [19:43<00:07,  4.04it/s]

train:  13%|█▎        | 453/3515 [01:58<12:37,  4.04it/s]
overall:  99%|█████████▉| 3969/4000 [19:44<00:09,  3.32it/s]

train:  13%|█▎        | 454/3515 [01:58<15:22,  3.32it/s]
overall:  99%|█████████▉| 3970/4000 [19:44<00:08,  3.47it/s]

train:  13%|█▎        | 455/3515 [01:58<14:42,  3.47it/s]
overall:  99%|█████████▉| 3971/4000 [19:44<00:08,  3.57it/s]

train:  13%|█▎        | 456/3515 [01:59<14:16,  3.57it/s]
overall:  99%|█████████▉| 3972/4000 [19:44<00:07,  3.55it/s]

train:  13%|█▎        | 457/3515 [01:59<14:22,  3.55it/s]
overall:  99%|█████████▉| 3973/4000 [19:45<00:07,  3.62it/s]

train:  13%|█▎        | 458/3515 [01:59<14:04,  3.62it/s]
overall:  99%|█████████▉| 3974/4000 [19:45<00:07,  3.64it/s]

train:  13%|█▎        | 459/3515 [01:59<13:58,  3.64it/s]
overall:  99%|█████████▉| 3975/4000 [19:45<00:06,  3.87it/s]

train:  13%|█▎        | 460/3515 [02:00<13:09,  3.87it/s]
overall:  99%|█████████▉| 3976/4000 [19:45<00:06,  3.79it/s]

train:  13%|█▎        | 461/3515 [02:00<13:26,  3.79it/s]
overall:  99%|█████████▉| 3977/4000 [19:46<00:06,  3.75it/s]

train:  13%|█▎        | 462/3515 [02:00<13:34,  3.75it/s]
overall:  99%|█████████▉| 3978/4000 [19:46<00:05,  3.69it/s]

train:  13%|█▎        | 463/3515 [02:00<13:47,  3.69it/s]
overall:  99%|█████████▉| 3979/4000 [19:46<00:05,  3.75it/s]

train:  13%|█▎        | 464/3515 [02:01<13:34,  3.75it/s]train loss: 0.5496253371238708
train per: 0.16909728179612246

overall: 100%|█████████▉| 3981/4000 [19:47<00:04,  3.91it/s]

train:  13%|█▎        | 466/3515 [02:01<12:58,  3.91it/s]
overall: 100%|█████████▉| 3982/4000 [19:47<00:04,  3.93it/s]

train:  13%|█▎        | 467/3515 [02:01<12:56,  3.93it/s]
overall: 100%|█████████▉| 3983/4000 [19:47<00:04,  3.92it/s]

train:  13%|█▎        | 468/3515 [02:02<12:57,  3.92it/s]
overall: 100%|█████████▉| 3984/4000 [19:47<00:04,  3.95it/s]

train:  13%|█▎        | 469/3515 [02:02<12:51,  3.95it/s]
overall: 100%|█████████▉| 3985/4000 [19:48<00:03,  3.97it/s]

train:  13%|█▎        | 470/3515 [02:02<12:46,  3.97it/s]
overall: 100%|█████████▉| 3986/4000 [19:48<00:03,  4.01it/s]

train:  13%|█▎        | 471/3515 [02:02<12:38,  4.01it/s]
overall: 100%|█████████▉| 3987/4000 [19:48<00:02,  4.67it/s]

train:  13%|█▎        | 472/3515 [02:03<10:50,  4.68it/s]
overall: 100%|█████████▉| 3988/4000 [19:48<00:02,  4.49it/s]

train:  13%|█▎        | 473/3515 [02:03<11:17,  4.49it/s]
overall: 100%|█████████▉| 3989/4000 [19:49<00:02,  4.24it/s]

train:  13%|█▎        | 474/3515 [02:03<11:56,  4.24it/s]
overall: 100%|█████████▉| 3990/4000 [19:49<00:02,  4.06it/s]

train:  14%|█▎        | 475/3515 [02:03<12:28,  4.06it/s]
overall: 100%|█████████▉| 3991/4000 [19:49<00:02,  3.97it/s]

train:  14%|█▎        | 476/3515 [02:04<12:44,  3.97it/s]
overall: 100%|█████████▉| 3992/4000 [19:49<00:02,  3.92it/s]

train:  14%|█▎        | 477/3515 [02:04<12:54,  3.92it/s]
overall: 100%|█████████▉| 3993/4000 [19:50<00:02,  3.30it/s]

train:  14%|█▎        | 478/3515 [02:04<15:20,  3.30it/s]
overall: 100%|█████████▉| 3994/4000 [19:50<00:01,  3.40it/s]

train:  14%|█▎        | 479/3515 [02:05<14:52,  3.40it/s]
overall: 100%|█████████▉| 3995/4000 [19:50<00:01,  3.75it/s]

train:  14%|█▎        | 480/3515 [02:05<13:29,  3.75it/s]
overall: 100%|█████████▉| 3996/4000 [19:51<00:01,  3.70it/s]

train:  14%|█▎        | 481/3515 [02:05<13:40,  3.70it/s]
overall: 100%|█████████▉| 3997/4000 [19:51<00:00,  3.71it/s]

train:  14%|█▎        | 482/3515 [02:05<13:37,  3.71it/s]
overall: 100%|█████████▉| 3998/4000 [19:51<00:00,  3.77it/s]

train:  14%|█▎        | 483/3515 [02:06<13:25,  3.77it/s]
overall: 100%|█████████▉| 3999/4000 [19:51<00:00,  3.72it/s]

train:  14%|█▍        | 484/3515 [02:06<13:34,  3.72it/s]train loss: 0.5500874519348145
train per: 0.1608711908767994
[LibriPhone] - Taking the first phoneme sequences for a deterministic behavior.



word -> phonemes:   0%|          | 0/2703 [00:00<?, ?it/s]


word -> phonemes:  13%|█▎        | 345/2703 [00:00<00:00, 3440.24it/s]


word -> phonemes:  26%|██▌       | 690/2703 [00:00<00:00, 3380.02it/s]


word -> phonemes:  38%|███▊      | 1029/2703 [00:00<00:00, 3326.10it/s]


word -> phonemes:  50%|█████     | 1362/2703 [00:00<00:00, 3318.06it/s]


word -> phonemes:  63%|██████▎   | 1696/2703 [00:00<00:00, 3323.21it/s]


word -> phonemes:  75%|███████▌  | 2029/2703 [00:00<00:00, 3234.59it/s]


word -> phonemes:  87%|████████▋ | 2364/2703 [00:00<00:00, 3271.16it/s]


word -> phonemes: 100%|█████████▉| 2692/2703 [00:00<00:00, 3191.92it/s]
word -> phonemes: 100%|██████████| 2703/2703 [00:00<00:00, 3257.60it/s]



dev:   0%|          | 0/2703 [00:00<?, ?it/s]


dev:   0%|          | 1/2703 [00:00<43:04,  1.05it/s]


dev:   0%|          | 4/2703 [00:01<10:03,  4.48it/s]


dev:   0%|          | 7/2703 [00:01<05:46,  7.79it/s]


dev:   0%|          | 10/2703 [00:01<04:02, 11.11it/s]


dev:   0%|          | 13/2703 [00:01<03:10, 14.15it/s]


dev:   1%|          | 16/2703 [00:01<02:39, 16.86it/s]


dev:   1%|          | 19/2703 [00:01<02:18, 19.39it/s]


dev:   1%|          | 22/2703 [00:01<02:03, 21.69it/s]


dev:   1%|          | 25/2703 [00:01<01:54, 23.34it/s]


dev:   1%|          | 28/2703 [00:02<01:52, 23.84it/s]


dev:   1%|          | 32/2703 [00:02<01:41, 26.20it/s]


dev:   1%|▏         | 36/2703 [00:02<01:36, 27.70it/s]


dev:   1%|▏         | 40/2703 [00:02<01:32, 28.76it/s]


dev:   2%|▏         | 44/2703 [00:02<01:26, 30.63it/s]


dev:   2%|▏         | 48/2703 [00:02<01:23, 31.77it/s]


dev:   2%|▏         | 52/2703 [00:02<01:24, 31.55it/s]


dev:   2%|▏         | 56/2703 [00:02<01:20, 32.96it/s]


dev:   2%|▏         | 60/2703 [00:03<01:19, 33.19it/s]


dev:   2%|▏         | 64/2703 [00:03<01:17, 34.10it/s]


dev:   3%|▎         | 68/2703 [00:03<01:14, 35.18it/s]


dev:   3%|▎         | 72/2703 [00:03<01:14, 35.32it/s]


dev:   3%|▎         | 76/2703 [00:03<01:14, 35.47it/s]


dev:   3%|▎         | 80/2703 [00:03<01:12, 35.93it/s]


dev:   3%|▎         | 84/2703 [00:03<01:11, 36.64it/s]


dev:   3%|▎         | 88/2703 [00:03<01:14, 35.33it/s]


dev:   3%|▎         | 92/2703 [00:03<01:14, 35.19it/s]


dev:   4%|▎         | 96/2703 [00:04<01:12, 35.94it/s]


dev:   4%|▎         | 100/2703 [00:04<01:11, 36.55it/s]


dev:   4%|▍         | 104/2703 [00:04<01:11, 36.59it/s]


dev:   4%|▍         | 108/2703 [00:04<01:11, 36.39it/s]


dev:   4%|▍         | 112/2703 [00:04<01:10, 36.60it/s]


dev:   4%|▍         | 117/2703 [00:04<01:09, 37.29it/s]


dev:   4%|▍         | 121/2703 [00:04<01:09, 37.31it/s]


dev:   5%|▍         | 125/2703 [00:04<01:08, 37.79it/s]


dev:   5%|▍         | 130/2703 [00:04<01:04, 39.91it/s]


dev:   5%|▍         | 134/2703 [00:05<01:06, 38.42it/s]


dev:   5%|▌         | 139/2703 [00:05<01:05, 39.36it/s]


dev:   5%|▌         | 143/2703 [00:05<01:05, 39.20it/s]


dev:   5%|▌         | 147/2703 [00:05<01:06, 38.59it/s]


dev:   6%|▌         | 151/2703 [00:05<01:06, 38.41it/s]


dev:   6%|▌         | 156/2703 [00:05<01:06, 38.02it/s]


dev:   6%|▌         | 161/2703 [00:05<01:06, 38.23it/s]


dev:   6%|▌         | 166/2703 [00:05<01:03, 39.66it/s]


dev:   6%|▋         | 170/2703 [00:05<01:04, 39.49it/s]


dev:   6%|▋         | 175/2703 [00:06<01:02, 40.30it/s]


dev:   7%|▋         | 180/2703 [00:06<01:03, 40.01it/s]


dev:   7%|▋         | 185/2703 [00:06<01:00, 41.35it/s]


dev:   7%|▋         | 190/2703 [00:06<00:59, 41.97it/s]


dev:   7%|▋         | 195/2703 [00:06<01:00, 41.50it/s]


dev:   7%|▋         | 200/2703 [00:06<00:58, 43.08it/s]


dev:   8%|▊         | 205/2703 [00:06<00:58, 42.67it/s]


dev:   8%|▊         | 210/2703 [00:06<00:57, 43.39it/s]


dev:   8%|▊         | 215/2703 [00:06<00:56, 44.06it/s]


dev:   8%|▊         | 220/2703 [00:07<00:56, 43.83it/s]


dev:   8%|▊         | 225/2703 [00:07<00:56, 43.55it/s]


dev:   9%|▊         | 230/2703 [00:07<00:56, 43.55it/s]


dev:   9%|▊         | 235/2703 [00:07<00:56, 43.88it/s]


dev:   9%|▉         | 240/2703 [00:07<00:56, 43.68it/s]


dev:   9%|▉         | 245/2703 [00:07<00:55, 44.56it/s]


dev:   9%|▉         | 250/2703 [00:07<00:54, 44.68it/s]


dev:   9%|▉         | 255/2703 [00:07<00:55, 44.00it/s]


dev:  10%|▉         | 260/2703 [00:07<00:53, 45.32it/s]


dev:  10%|▉         | 265/2703 [00:08<00:56, 43.40it/s]


dev:  10%|█         | 271/2703 [00:08<00:53, 45.69it/s]


dev:  10%|█         | 276/2703 [00:08<00:55, 43.89it/s]


dev:  10%|█         | 281/2703 [00:08<00:53, 45.40it/s]


dev:  11%|█         | 286/2703 [00:08<00:51, 46.58it/s]


dev:  11%|█         | 291/2703 [00:08<00:52, 45.83it/s]


dev:  11%|█         | 297/2703 [00:08<00:50, 47.48it/s]


dev:  11%|█         | 302/2703 [00:08<00:51, 46.47it/s]


dev:  11%|█▏        | 307/2703 [00:09<00:51, 46.71it/s]


dev:  12%|█▏        | 312/2703 [00:09<00:51, 46.05it/s]


dev:  12%|█▏        | 317/2703 [00:09<00:50, 46.99it/s]


dev:  12%|█▏        | 322/2703 [00:09<00:50, 46.77it/s]


dev:  12%|█▏        | 327/2703 [00:09<00:50, 47.29it/s]


dev:  12%|█▏        | 332/2703 [00:09<00:49, 47.99it/s]


dev:  13%|█▎        | 338/2703 [00:09<00:47, 49.27it/s]


dev:  13%|█▎        | 344/2703 [00:09<00:47, 49.93it/s]


dev:  13%|█▎        | 349/2703 [00:09<00:47, 49.88it/s]


dev:  13%|█▎        | 355/2703 [00:09<00:46, 50.92it/s]


dev:  13%|█▎        | 361/2703 [00:10<00:48, 48.36it/s]


dev:  14%|█▎        | 367/2703 [00:10<00:47, 49.16it/s]


dev:  14%|█▍        | 372/2703 [00:10<00:47, 49.24it/s]


dev:  14%|█▍        | 378/2703 [00:10<00:46, 50.39it/s]


dev:  14%|█▍        | 384/2703 [00:10<00:46, 49.97it/s]


dev:  14%|█▍        | 390/2703 [00:10<00:45, 50.35it/s]


dev:  15%|█▍        | 396/2703 [00:10<00:46, 50.04it/s]


dev:  15%|█▍        | 402/2703 [00:10<00:45, 50.27it/s]


dev:  15%|█▌        | 408/2703 [00:11<00:45, 50.40it/s]


dev:  15%|█▌        | 414/2703 [00:11<00:45, 50.58it/s]


dev:  16%|█▌        | 420/2703 [00:11<00:45, 50.61it/s]


dev:  16%|█▌        | 426/2703 [00:11<00:43, 51.75it/s]


dev:  16%|█▌        | 432/2703 [00:11<00:43, 52.14it/s]


dev:  16%|█▌        | 438/2703 [00:11<00:43, 52.40it/s]


dev:  16%|█▋        | 444/2703 [00:11<00:44, 50.60it/s]


dev:  17%|█▋        | 450/2703 [00:11<00:43, 51.63it/s]


dev:  17%|█▋        | 456/2703 [00:11<00:43, 51.09it/s]


dev:  17%|█▋        | 462/2703 [00:12<00:43, 51.42it/s]


dev:  17%|█▋        | 468/2703 [00:12<00:43, 51.51it/s]


dev:  18%|█▊        | 474/2703 [00:12<00:42, 52.12it/s]


dev:  18%|█▊        | 480/2703 [00:12<00:43, 51.48it/s]


dev:  18%|█▊        | 486/2703 [00:12<00:42, 52.75it/s]


dev:  18%|█▊        | 492/2703 [00:12<00:41, 53.02it/s]


dev:  18%|█▊        | 498/2703 [00:12<00:40, 54.56it/s]


dev:  19%|█▊        | 504/2703 [00:12<00:40, 54.24it/s]


dev:  19%|█▉        | 510/2703 [00:12<00:39, 55.50it/s]


dev:  19%|█▉        | 516/2703 [00:13<00:39, 55.34it/s]


dev:  19%|█▉        | 522/2703 [00:13<00:39, 55.19it/s]


dev:  20%|█▉        | 528/2703 [00:13<00:39, 55.02it/s]


dev:  20%|█▉        | 534/2703 [00:13<00:38, 56.15it/s]


dev:  20%|█▉        | 540/2703 [00:13<00:39, 54.81it/s]


dev:  20%|██        | 546/2703 [00:13<00:38, 55.99it/s]


dev:  20%|██        | 552/2703 [00:13<00:38, 55.30it/s]


dev:  21%|██        | 558/2703 [00:13<00:38, 56.33it/s]


dev:  21%|██        | 564/2703 [00:13<00:38, 56.01it/s]


dev:  21%|██        | 570/2703 [00:14<00:38, 55.80it/s]


dev:  21%|██▏       | 576/2703 [00:14<00:38, 55.91it/s]


dev:  22%|██▏       | 582/2703 [00:14<00:38, 54.82it/s]


dev:  22%|██▏       | 588/2703 [00:14<00:37, 55.71it/s]


dev:  22%|██▏       | 594/2703 [00:14<00:38, 55.41it/s]


dev:  22%|██▏       | 600/2703 [00:14<00:38, 54.88it/s]


dev:  22%|██▏       | 606/2703 [00:14<00:37, 55.47it/s]


dev:  23%|██▎       | 612/2703 [00:14<00:37, 55.19it/s]


dev:  23%|██▎       | 618/2703 [00:14<00:37, 55.69it/s]


dev:  23%|██▎       | 624/2703 [00:15<00:36, 56.41it/s]


dev:  23%|██▎       | 630/2703 [00:15<00:36, 57.29it/s]


dev:  24%|██▎       | 636/2703 [00:15<00:36, 57.04it/s]


dev:  24%|██▍       | 643/2703 [00:15<00:35, 57.98it/s]


dev:  24%|██▍       | 649/2703 [00:15<00:35, 58.09it/s]


dev:  24%|██▍       | 655/2703 [00:15<00:34, 58.63it/s]


dev:  24%|██▍       | 661/2703 [00:15<00:35, 57.67it/s]


dev:  25%|██▍       | 667/2703 [00:15<00:35, 57.57it/s]


dev:  25%|██▍       | 673/2703 [00:15<00:35, 57.58it/s]


dev:  25%|██▌       | 679/2703 [00:15<00:35, 56.74it/s]


dev:  25%|██▌       | 685/2703 [00:16<00:35, 57.63it/s]


dev:  26%|██▌       | 691/2703 [00:16<00:35, 57.48it/s]


dev:  26%|██▌       | 697/2703 [00:16<00:34, 58.11it/s]


dev:  26%|██▌       | 703/2703 [00:16<00:34, 57.18it/s]


dev:  26%|██▌       | 709/2703 [00:16<00:34, 56.98it/s]


dev:  26%|██▋       | 716/2703 [00:16<00:33, 58.53it/s]


dev:  27%|██▋       | 723/2703 [00:16<00:33, 59.38it/s]


dev:  27%|██▋       | 729/2703 [00:16<00:33, 59.19it/s]


dev:  27%|██▋       | 735/2703 [00:16<00:33, 58.75it/s]


dev:  27%|██▋       | 742/2703 [00:17<00:32, 59.44it/s]


dev:  28%|██▊       | 749/2703 [00:17<00:32, 59.97it/s]


dev:  28%|██▊       | 756/2703 [00:17<00:32, 60.26it/s]


dev:  28%|██▊       | 763/2703 [00:17<00:32, 58.98it/s]


dev:  28%|██▊       | 770/2703 [00:17<00:32, 59.90it/s]


dev:  29%|██▊       | 776/2703 [00:17<00:32, 59.63it/s]


dev:  29%|██▉       | 782/2703 [00:17<00:32, 59.20it/s]


dev:  29%|██▉       | 789/2703 [00:17<00:31, 60.61it/s]


dev:  29%|██▉       | 796/2703 [00:17<00:31, 60.39it/s]


dev:  30%|██▉       | 803/2703 [00:18<00:30, 61.37it/s]


dev:  30%|██▉       | 810/2703 [00:18<00:30, 61.58it/s]


dev:  30%|███       | 817/2703 [00:18<00:31, 60.79it/s]


dev:  30%|███       | 824/2703 [00:18<00:30, 61.69it/s]


dev:  31%|███       | 831/2703 [00:18<00:29, 62.66it/s]


dev:  31%|███       | 838/2703 [00:18<00:29, 63.07it/s]


dev:  31%|███▏      | 845/2703 [00:18<00:29, 62.95it/s]


dev:  32%|███▏      | 852/2703 [00:18<00:30, 61.61it/s]


dev:  32%|███▏      | 859/2703 [00:18<00:29, 61.93it/s]


dev:  32%|███▏      | 866/2703 [00:19<00:30, 61.02it/s]


dev:  32%|███▏      | 873/2703 [00:19<00:29, 62.13it/s]


dev:  33%|███▎      | 880/2703 [00:19<00:28, 63.17it/s]


dev:  33%|███▎      | 887/2703 [00:19<00:29, 61.71it/s]


dev:  33%|███▎      | 894/2703 [00:19<00:29, 61.54it/s]


dev:  33%|███▎      | 901/2703 [00:19<00:29, 60.42it/s]


dev:  34%|███▎      | 908/2703 [00:19<00:29, 61.38it/s]


dev:  34%|███▍      | 915/2703 [00:19<00:29, 61.15it/s]


dev:  34%|███▍      | 922/2703 [00:19<00:29, 59.62it/s]


dev:  34%|███▍      | 929/2703 [00:20<00:28, 61.52it/s]


dev:  35%|███▍      | 936/2703 [00:20<00:28, 61.58it/s]


dev:  35%|███▍      | 943/2703 [00:20<00:28, 61.18it/s]


dev:  35%|███▌      | 950/2703 [00:20<00:27, 63.16it/s]


dev:  35%|███▌      | 957/2703 [00:20<00:28, 62.00it/s]


dev:  36%|███▌      | 964/2703 [00:20<00:27, 63.85it/s]


dev:  36%|███▌      | 971/2703 [00:20<00:26, 64.86it/s]


dev:  36%|███▌      | 978/2703 [00:20<00:26, 64.30it/s]


dev:  36%|███▋      | 985/2703 [00:20<00:26, 64.76it/s]


dev:  37%|███▋      | 992/2703 [00:21<00:27, 62.89it/s]


dev:  37%|███▋      | 1000/2703 [00:21<00:25, 65.66it/s]


dev:  37%|███▋      | 1007/2703 [00:21<00:26, 64.23it/s]


dev:  38%|███▊      | 1014/2703 [00:21<00:26, 63.30it/s]


dev:  38%|███▊      | 1021/2703 [00:21<00:26, 64.33it/s]


dev:  38%|███▊      | 1028/2703 [00:21<00:26, 64.22it/s]


dev:  38%|███▊      | 1035/2703 [00:21<00:25, 65.02it/s]


dev:  39%|███▊      | 1042/2703 [00:21<00:25, 63.91it/s]


dev:  39%|███▉      | 1049/2703 [00:21<00:26, 63.51it/s]


dev:  39%|███▉      | 1056/2703 [00:22<00:25, 64.89it/s]


dev:  39%|███▉      | 1063/2703 [00:22<00:25, 63.84it/s]


dev:  40%|███▉      | 1070/2703 [00:22<00:24, 65.48it/s]


dev:  40%|███▉      | 1077/2703 [00:22<00:24, 66.69it/s]


dev:  40%|████      | 1084/2703 [00:22<00:24, 66.55it/s]


dev:  40%|████      | 1091/2703 [00:22<00:24, 65.69it/s]


dev:  41%|████      | 1098/2703 [00:22<00:24, 65.45it/s]


dev:  41%|████      | 1105/2703 [00:22<00:24, 66.09it/s]


dev:  41%|████      | 1112/2703 [00:22<00:23, 66.61it/s]


dev:  41%|████▏     | 1119/2703 [00:23<00:23, 66.37it/s]


dev:  42%|████▏     | 1126/2703 [00:23<00:23, 66.45it/s]


dev:  42%|████▏     | 1134/2703 [00:23<00:23, 66.28it/s]


dev:  42%|████▏     | 1142/2703 [00:23<00:22, 68.13it/s]


dev:  43%|████▎     | 1149/2703 [00:23<00:22, 67.58it/s]


dev:  43%|████▎     | 1156/2703 [00:23<00:23, 66.95it/s]


dev:  43%|████▎     | 1164/2703 [00:23<00:22, 68.08it/s]


dev:  43%|████▎     | 1171/2703 [00:23<00:22, 67.94it/s]


dev:  44%|████▎     | 1178/2703 [00:23<00:22, 68.30it/s]


dev:  44%|████▍     | 1185/2703 [00:23<00:22, 68.25it/s]


dev:  44%|████▍     | 1192/2703 [00:24<00:22, 68.46it/s]


dev:  44%|████▍     | 1199/2703 [00:24<00:22, 68.12it/s]


dev:  45%|████▍     | 1207/2703 [00:24<00:21, 69.69it/s]


dev:  45%|████▍     | 1214/2703 [00:24<00:21, 69.47it/s]


dev:  45%|████▌     | 1221/2703 [00:24<00:21, 69.36it/s]


dev:  45%|████▌     | 1229/2703 [00:24<00:20, 70.50it/s]


dev:  46%|████▌     | 1237/2703 [00:24<00:21, 68.68it/s]


dev:  46%|████▌     | 1245/2703 [00:24<00:21, 69.17it/s]


dev:  46%|████▋     | 1253/2703 [00:24<00:20, 70.04it/s]


dev:  47%|████▋     | 1261/2703 [00:25<00:21, 67.95it/s]


dev:  47%|████▋     | 1269/2703 [00:25<00:20, 68.90it/s]


dev:  47%|████▋     | 1277/2703 [00:25<00:20, 70.42it/s]


dev:  48%|████▊     | 1285/2703 [00:25<00:20, 69.46it/s]


dev:  48%|████▊     | 1293/2703 [00:25<00:20, 69.95it/s]


dev:  48%|████▊     | 1301/2703 [00:25<00:20, 69.88it/s]


dev:  48%|████▊     | 1309/2703 [00:25<00:19, 70.41it/s]


dev:  49%|████▊     | 1317/2703 [00:25<00:19, 70.48it/s]


dev:  49%|████▉     | 1325/2703 [00:25<00:19, 70.61it/s]


dev:  49%|████▉     | 1333/2703 [00:26<00:19, 70.39it/s]


dev:  50%|████▉     | 1341/2703 [00:26<00:19, 70.46it/s]


dev:  50%|████▉     | 1349/2703 [00:26<00:19, 70.82it/s]


dev:  50%|█████     | 1357/2703 [00:26<00:19, 70.40it/s]


dev:  50%|█████     | 1365/2703 [00:26<00:18, 71.07it/s]


dev:  51%|█████     | 1373/2703 [00:26<00:18, 71.21it/s]


dev:  51%|█████     | 1381/2703 [00:26<00:18, 69.88it/s]


dev:  51%|█████▏    | 1389/2703 [00:26<00:18, 70.21it/s]


dev:  52%|█████▏    | 1397/2703 [00:27<00:18, 69.84it/s]


dev:  52%|█████▏    | 1405/2703 [00:27<00:18, 70.30it/s]


dev:  52%|█████▏    | 1413/2703 [00:27<00:18, 71.05it/s]


dev:  53%|█████▎    | 1421/2703 [00:27<00:18, 70.33it/s]


dev:  53%|█████▎    | 1429/2703 [00:27<00:18, 70.17it/s]


dev:  53%|█████▎    | 1437/2703 [00:27<00:18, 69.37it/s]


dev:  53%|█████▎    | 1444/2703 [00:27<00:18, 68.29it/s]


dev:  54%|█████▎    | 1452/2703 [00:27<00:18, 69.48it/s]


dev:  54%|█████▍    | 1460/2703 [00:27<00:17, 70.88it/s]


dev:  54%|█████▍    | 1468/2703 [00:28<00:17, 70.34it/s]


dev:  55%|█████▍    | 1476/2703 [00:28<00:17, 70.69it/s]


dev:  55%|█████▍    | 1484/2703 [00:28<00:17, 71.40it/s]


dev:  55%|█████▌    | 1492/2703 [00:28<00:17, 70.67it/s]


dev:  55%|█████▌    | 1500/2703 [00:28<00:17, 69.02it/s]


dev:  56%|█████▌    | 1508/2703 [00:28<00:17, 69.81it/s]


dev:  56%|█████▌    | 1516/2703 [00:28<00:16, 70.61it/s]


dev:  56%|█████▋    | 1524/2703 [00:28<00:16, 69.62it/s]


dev:  57%|█████▋    | 1532/2703 [00:28<00:16, 70.48it/s]


dev:  57%|█████▋    | 1540/2703 [00:29<00:16, 69.83it/s]


dev:  57%|█████▋    | 1548/2703 [00:29<00:16, 70.24it/s]


dev:  58%|█████▊    | 1556/2703 [00:29<00:16, 70.39it/s]


dev:  58%|█████▊    | 1564/2703 [00:29<00:16, 71.10it/s]


dev:  58%|█████▊    | 1572/2703 [00:29<00:15, 71.83it/s]


dev:  58%|█████▊    | 1580/2703 [00:29<00:15, 72.49it/s]


dev:  59%|█████▊    | 1588/2703 [00:29<00:15, 73.05it/s]


dev:  59%|█████▉    | 1596/2703 [00:29<00:15, 73.43it/s]


dev:  59%|█████▉    | 1604/2703 [00:29<00:14, 73.58it/s]


dev:  60%|█████▉    | 1612/2703 [00:30<00:14, 73.69it/s]


dev:  60%|█████▉    | 1620/2703 [00:30<00:14, 72.87it/s]


dev:  60%|██████    | 1628/2703 [00:30<00:14, 73.80it/s]


dev:  61%|██████    | 1636/2703 [00:30<00:14, 73.81it/s]


dev:  61%|██████    | 1644/2703 [00:30<00:14, 73.67it/s]


dev:  61%|██████    | 1652/2703 [00:30<00:14, 74.65it/s]


dev:  61%|██████▏   | 1660/2703 [00:30<00:14, 74.49it/s]


dev:  62%|██████▏   | 1668/2703 [00:30<00:13, 74.52it/s]


dev:  62%|██████▏   | 1676/2703 [00:30<00:13, 74.99it/s]


dev:  62%|██████▏   | 1684/2703 [00:31<00:13, 73.97it/s]


dev:  63%|██████▎   | 1692/2703 [00:31<00:13, 74.07it/s]


dev:  63%|██████▎   | 1700/2703 [00:31<00:13, 74.17it/s]


dev:  63%|██████▎   | 1708/2703 [00:31<00:13, 71.43it/s]


dev:  63%|██████▎   | 1716/2703 [00:31<00:13, 71.83it/s]


dev:  64%|██████▍   | 1724/2703 [00:31<00:13, 72.71it/s]


dev:  64%|██████▍   | 1732/2703 [00:31<00:13, 71.51it/s]


dev:  64%|██████▍   | 1740/2703 [00:31<00:13, 71.80it/s]


dev:  65%|██████▍   | 1748/2703 [00:31<00:13, 71.41it/s]


dev:  65%|██████▍   | 1756/2703 [00:32<00:13, 70.84it/s]


dev:  65%|██████▌   | 1764/2703 [00:32<00:13, 71.49it/s]


dev:  66%|██████▌   | 1772/2703 [00:32<00:12, 71.77it/s]


dev:  66%|██████▌   | 1780/2703 [00:32<00:12, 71.82it/s]


dev:  66%|██████▌   | 1788/2703 [00:32<00:12, 72.44it/s]


dev:  66%|██████▋   | 1796/2703 [00:32<00:12, 72.65it/s]


dev:  67%|██████▋   | 1804/2703 [00:32<00:12, 73.14it/s]


dev:  67%|██████▋   | 1812/2703 [00:32<00:12, 73.39it/s]


dev:  67%|██████▋   | 1820/2703 [00:32<00:12, 72.29it/s]


dev:  68%|██████▊   | 1828/2703 [00:33<00:11, 73.07it/s]


dev:  68%|██████▊   | 1836/2703 [00:33<00:12, 71.24it/s]


dev:  68%|██████▊   | 1844/2703 [00:33<00:11, 72.09it/s]


dev:  69%|██████▊   | 1852/2703 [00:33<00:11, 71.72it/s]


dev:  69%|██████▉   | 1860/2703 [00:33<00:11, 72.34it/s]


dev:  69%|██████▉   | 1868/2703 [00:33<00:11, 73.27it/s]


dev:  69%|██████▉   | 1876/2703 [00:33<00:11, 73.06it/s]


dev:  70%|██████▉   | 1884/2703 [00:33<00:11, 73.37it/s]


dev:  70%|██████▉   | 1892/2703 [00:33<00:11, 72.16it/s]


dev:  70%|███████   | 1900/2703 [00:34<00:11, 72.85it/s]


dev:  71%|███████   | 1908/2703 [00:34<00:10, 73.58it/s]


dev:  71%|███████   | 1916/2703 [00:34<00:10, 72.52it/s]


dev:  71%|███████   | 1924/2703 [00:34<00:10, 70.91it/s]


dev:  71%|███████▏  | 1932/2703 [00:34<00:10, 72.08it/s]


dev:  72%|███████▏  | 1940/2703 [00:34<00:10, 72.06it/s]


dev:  72%|███████▏  | 1948/2703 [00:34<00:10, 72.68it/s]


dev:  72%|███████▏  | 1956/2703 [00:34<00:10, 71.73it/s]


dev:  73%|███████▎  | 1964/2703 [00:34<00:10, 72.70it/s]


dev:  73%|███████▎  | 1972/2703 [00:34<00:09, 73.36it/s]


dev:  73%|███████▎  | 1980/2703 [00:35<00:09, 73.08it/s]


dev:  74%|███████▎  | 1988/2703 [00:35<00:09, 73.34it/s]


dev:  74%|███████▍  | 1996/2703 [00:35<00:09, 73.42it/s]


dev:  74%|███████▍  | 2004/2703 [00:35<00:09, 72.86it/s]


dev:  74%|███████▍  | 2012/2703 [00:35<00:09, 73.32it/s]


dev:  75%|███████▍  | 2020/2703 [00:35<00:09, 72.49it/s]


dev:  75%|███████▌  | 2028/2703 [00:35<00:09, 73.52it/s]


dev:  75%|███████▌  | 2036/2703 [00:35<00:09, 74.05it/s]


dev:  76%|███████▌  | 2044/2703 [00:35<00:09, 72.58it/s]


dev:  76%|███████▌  | 2052/2703 [00:36<00:09, 72.21it/s]


dev:  76%|███████▌  | 2060/2703 [00:36<00:08, 72.75it/s]


dev:  77%|███████▋  | 2068/2703 [00:36<00:08, 73.13it/s]


dev:  77%|███████▋  | 2076/2703 [00:36<00:08, 72.72it/s]


dev:  77%|███████▋  | 2084/2703 [00:36<00:08, 73.23it/s]


dev:  77%|███████▋  | 2092/2703 [00:36<00:08, 72.07it/s]


dev:  78%|███████▊  | 2100/2703 [00:36<00:08, 72.96it/s]


dev:  78%|███████▊  | 2108/2703 [00:36<00:08, 72.25it/s]


dev:  78%|███████▊  | 2116/2703 [00:36<00:08, 73.03it/s]


dev:  79%|███████▊  | 2124/2703 [00:37<00:07, 72.51it/s]


dev:  79%|███████▉  | 2132/2703 [00:37<00:07, 73.20it/s]


dev:  79%|███████▉  | 2140/2703 [00:37<00:07, 73.96it/s]


dev:  79%|███████▉  | 2148/2703 [00:37<00:07, 74.05it/s]


dev:  80%|███████▉  | 2156/2703 [00:37<00:07, 74.00it/s]


dev:  80%|████████  | 2164/2703 [00:37<00:07, 72.74it/s]


dev:  80%|████████  | 2172/2703 [00:37<00:07, 73.49it/s]


dev:  81%|████████  | 2180/2703 [00:37<00:07, 72.60it/s]


dev:  81%|████████  | 2188/2703 [00:37<00:07, 72.44it/s]


dev:  81%|████████  | 2196/2703 [00:38<00:06, 73.61it/s]


dev:  82%|████████▏ | 2204/2703 [00:38<00:06, 74.01it/s]


dev:  82%|████████▏ | 2212/2703 [00:38<00:06, 74.10it/s]


dev:  82%|████████▏ | 2220/2703 [00:38<00:06, 74.61it/s]


dev:  82%|████████▏ | 2228/2703 [00:38<00:06, 73.29it/s]


dev:  83%|████████▎ | 2236/2703 [00:38<00:06, 73.78it/s]


dev:  83%|████████▎ | 2244/2703 [00:38<00:06, 73.21it/s]


dev:  83%|████████▎ | 2252/2703 [00:38<00:06, 73.59it/s]


dev:  84%|████████▎ | 2260/2703 [00:38<00:06, 73.38it/s]


dev:  84%|████████▍ | 2268/2703 [00:39<00:06, 71.70it/s]


dev:  84%|████████▍ | 2276/2703 [00:39<00:05, 72.32it/s]


dev:  84%|████████▍ | 2284/2703 [00:39<00:05, 72.40it/s]


dev:  85%|████████▍ | 2292/2703 [00:39<00:05, 73.42it/s]


dev:  85%|████████▌ | 2300/2703 [00:39<00:05, 73.79it/s]


dev:  85%|████████▌ | 2308/2703 [00:39<00:05, 73.29it/s]


dev:  86%|████████▌ | 2316/2703 [00:39<00:05, 73.83it/s]


dev:  86%|████████▌ | 2324/2703 [00:39<00:05, 73.98it/s]


dev:  86%|████████▋ | 2332/2703 [00:39<00:04, 74.25it/s]


dev:  87%|████████▋ | 2340/2703 [00:40<00:04, 74.29it/s]


dev:  87%|████████▋ | 2348/2703 [00:40<00:04, 73.21it/s]


dev:  87%|████████▋ | 2356/2703 [00:40<00:04, 74.11it/s]


dev:  87%|████████▋ | 2364/2703 [00:40<00:04, 74.36it/s]


dev:  88%|████████▊ | 2372/2703 [00:40<00:04, 73.42it/s]


dev:  88%|████████▊ | 2380/2703 [00:40<00:04, 73.98it/s]


dev:  88%|████████▊ | 2388/2703 [00:40<00:04, 74.42it/s]


dev:  89%|████████▊ | 2396/2703 [00:40<00:04, 71.96it/s]


dev:  89%|████████▉ | 2404/2703 [00:40<00:04, 73.73it/s]


dev:  89%|████████▉ | 2413/2703 [00:41<00:03, 75.90it/s]


dev:  90%|████████▉ | 2421/2703 [00:41<00:03, 75.44it/s]


dev:  90%|████████▉ | 2429/2703 [00:41<00:03, 76.13it/s]


dev:  90%|█████████ | 2437/2703 [00:41<00:03, 75.86it/s]


dev:  90%|█████████ | 2445/2703 [00:41<00:03, 75.20it/s]


dev:  91%|█████████ | 2453/2703 [00:41<00:03, 75.78it/s]


dev:  91%|█████████ | 2461/2703 [00:41<00:03, 73.78it/s]


dev:  91%|█████████▏| 2469/2703 [00:41<00:03, 73.07it/s]


dev:  92%|█████████▏| 2477/2703 [00:41<00:03, 73.52it/s]


dev:  92%|█████████▏| 2485/2703 [00:41<00:02, 73.05it/s]


dev:  92%|█████████▏| 2493/2703 [00:42<00:02, 73.90it/s]


dev:  93%|█████████▎| 2501/2703 [00:42<00:02, 74.22it/s]


dev:  93%|█████████▎| 2509/2703 [00:42<00:02, 75.54it/s]


dev:  93%|█████████▎| 2517/2703 [00:42<00:02, 75.57it/s]


dev:  93%|█████████▎| 2525/2703 [00:42<00:02, 75.34it/s]


dev:  94%|█████████▎| 2533/2703 [00:42<00:02, 75.42it/s]


dev:  94%|█████████▍| 2541/2703 [00:42<00:02, 75.43it/s]


dev:  94%|█████████▍| 2549/2703 [00:42<00:02, 74.60it/s]


dev:  95%|█████████▍| 2558/2703 [00:42<00:01, 76.76it/s]


dev:  95%|█████████▍| 2566/2703 [00:43<00:01, 77.15it/s]


dev:  95%|█████████▌| 2574/2703 [00:43<00:01, 77.15it/s]


dev:  96%|█████████▌| 2582/2703 [00:43<00:01, 76.65it/s]


dev:  96%|█████████▌| 2591/2703 [00:43<00:01, 79.60it/s]


dev:  96%|█████████▌| 2599/2703 [00:43<00:01, 78.85it/s]


dev:  96%|█████████▋| 2607/2703 [00:43<00:01, 77.59it/s]


dev:  97%|█████████▋| 2615/2703 [00:43<00:01, 76.55it/s]


dev:  97%|█████████▋| 2623/2703 [00:43<00:01, 76.44it/s]


dev:  97%|█████████▋| 2632/2703 [00:43<00:00, 78.63it/s]


dev:  98%|█████████▊| 2640/2703 [00:43<00:00, 77.35it/s]


dev:  98%|█████████▊| 2648/2703 [00:44<00:00, 77.02it/s]


dev:  98%|█████████▊| 2656/2703 [00:44<00:00, 77.54it/s]


dev:  99%|█████████▊| 2665/2703 [00:44<00:00, 79.11it/s]


dev:  99%|█████████▉| 2674/2703 [00:44<00:00, 81.57it/s]


dev:  99%|█████████▉| 2683/2703 [00:44<00:00, 82.71it/s]


dev: 100%|█████████▉| 2693/2703 [00:44<00:00, 84.97it/s]
dev: 100%|██████████| 2703/2703 [00:44<00:00, 60.22it/s]
dev loss: 0.7637030482292175
dev per: 0.19178459937565037

                                                            

                                                         
[Runner] - Save the checkpoint to:

overall: 100%|█████████▉| 3999/4000 [20:39<00:00,  3.72it/s]

train:  14%|█▍        | 484/3515 [02:53<13:34,  3.72it/s]
                                                            

                                                         
1. result/downstream/grid/dev-best.ckpt

overall: 100%|█████████▉| 3999/4000 [20:39<00:00,  3.72it/s]

train:  14%|█▍        | 484/3515 [02:53<13:34,  3.72it/s]
                                                            

                                                         
2. result/downstream/grid/states-4000.ckpt

overall: 100%|█████████▉| 3999/4000 [20:39<00:00,  3.72it/s]

train:  14%|█▍        | 484/3515 [02:53<13:34,  3.72it/s]
overall: 100%|██████████| 4000/4000 [20:39<00:00, 14.41s/it]

train:  14%|█▍        | 485/3515 [02:53<12:07:46, 14.41s/it]
train:  14%|█▍        | 485/3515 [02:54<18:07,  2.79it/s]   

overall: 100%|██████████| 4000/4000 [20:39<00:00,  3.23it/s]