[Runner] - Start a new experiment
2023-08-10 22:33:01 | INFO | s3prl.upstream.hubert.hubconf | Converting a fairseq checkpoint: /home/victor/checkpoints/hubert_d2v2_800k.pt
2023-08-10 22:33:01 | INFO | s3prl.upstream.hubert.hubconf | To: /home/victor/checkpoints/hubert_d2v2_800k.converted.pt
2023-08-10 22:33:01 | 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]

train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:   0%|          | 1/4000 [00:02<2:16:52,  2.05s/it]

train:   3%|▎         | 1/37 [00:02<01:13,  2.05s/it]
overall:   0%|          | 2/4000 [00:02<1:26:13,  1.29s/it]

train:   5%|▌         | 2/37 [00:02<00:45,  1.29s/it]
overall:   0%|          | 3/4000 [00:03<1:09:54,  1.05s/it]

train:   8%|▊         | 3/37 [00:03<00:35,  1.05s/it]
overall:   0%|          | 4/4000 [00:04<1:02:16,  1.07it/s]

train:  11%|█         | 4/37 [00:04<00:30,  1.07it/s]
overall:   0%|          | 5/4000 [00:05<58:13,  1.14it/s]  

train:  14%|█▎        | 5/37 [00:05<00:27,  1.14it/s]
overall:   0%|          | 6/4000 [00:05<56:01,  1.19it/s]

train:  16%|█▌        | 6/37 [00:05<00:26,  1.19it/s]
overall:   0%|          | 7/4000 [00:06<54:30,  1.22it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.22it/s]
overall:   0%|          | 8/4000 [00:07<53:14,  1.25it/s]

train:  22%|██▏       | 8/37 [00:07<00:23,  1.25it/s]
overall:   0%|          | 9/4000 [00:08<52:33,  1.27it/s]

train:  24%|██▍       | 9/37 [00:08<00:22,  1.27it/s]train loss: 2.9848227500915527
train acc: 0.07999999821186066

overall:   0%|          | 10/4000 [00:08<52:12,  1.27it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.27it/s]
overall:   0%|          | 11/4000 [00:09<52:00,  1.28it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.28it/s]
overall:   0%|          | 12/4000 [00:10<51:36,  1.29it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.29it/s]
overall:   0%|          | 13/4000 [00:11<51:39,  1.29it/s]

train:  35%|███▌      | 13/37 [00:11<00:18,  1.29it/s]
overall:   0%|          | 14/4000 [00:12<51:40,  1.29it/s]

train:  38%|███▊      | 14/37 [00:12<00:17,  1.29it/s]
overall:   0%|          | 15/4000 [00:12<51:22,  1.29it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.29it/s]
overall:   0%|          | 16/4000 [00:13<51:15,  1.30it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.30it/s]
overall:   0%|          | 17/4000 [00:14<51:07,  1.30it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.30it/s]
overall:   0%|          | 18/4000 [00:15<51:21,  1.29it/s]

train:  49%|████▊     | 18/37 [00:15<00:14,  1.29it/s]
overall:   0%|          | 19/4000 [00:15<51:23,  1.29it/s]

train:  51%|█████▏    | 19/37 [00:15<00:13,  1.29it/s]train loss: 2.9377896785736084
train acc: 0.1080000028014183

overall:   0%|          | 20/4000 [00:16<51:12,  1.30it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.30it/s]
overall:   1%|          | 21/4000 [00:17<51:12,  1.29it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.29it/s]
overall:   1%|          | 22/4000 [00:18<51:02,  1.30it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.30it/s]
overall:   1%|          | 23/4000 [00:18<51:00,  1.30it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.30it/s]
overall:   1%|          | 24/4000 [00:19<51:00,  1.30it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.30it/s]
overall:   1%|          | 25/4000 [00:20<51:13,  1.29it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.29it/s]
overall:   1%|          | 26/4000 [00:21<51:21,  1.29it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.29it/s]
overall:   1%|          | 27/4000 [00:22<51:08,  1.29it/s]

train:  73%|███████▎  | 27/37 [00:22<00:07,  1.29it/s]
overall:   1%|          | 28/4000 [00:22<51:17,  1.29it/s]

train:  76%|███████▌  | 28/37 [00:22<00:06,  1.29it/s]
overall:   1%|          | 29/4000 [00:23<51:09,  1.29it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.29it/s]train loss: 2.8786613941192627
train acc: 0.13300000131130219

overall:   1%|          | 30/4000 [00:24<51:24,  1.29it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.29it/s]
overall:   1%|          | 31/4000 [00:25<51:24,  1.29it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.29it/s]
overall:   1%|          | 32/4000 [00:25<51:13,  1.29it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.29it/s]
overall:   1%|          | 33/4000 [00:26<51:15,  1.29it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.29it/s]
overall:   1%|          | 34/4000 [00:27<51:13,  1.29it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.29it/s]
overall:   1%|          | 35/4000 [00:28<51:07,  1.29it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.29it/s]
overall:   1%|          | 36/4000 [00:29<51:05,  1.29it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.29it/s]
overall:   1%|          | 37/4000 [00:29<49:25,  1.34it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.34it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:   1%|          | 38/4000 [00:31<1:04:51,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:   1%|          | 39/4000 [00:32<1:01:12,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]train loss: 2.821606397628784
train acc: 0.13765181601047516

overall:   1%|          | 40/4000 [00:32<58:22,  1.13it/s]  

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:   1%|          | 41/4000 [00:33<56:08,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:   1%|          | 42/4000 [00:34<54:40,  1.21it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.20it/s]
overall:   1%|          | 43/4000 [00:35<53:34,  1.23it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.23it/s]
overall:   1%|          | 44/4000 [00:35<52:42,  1.25it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.25it/s]
overall:   1%|          | 45/4000 [00:36<52:18,  1.26it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.26it/s]
overall:   1%|          | 46/4000 [00:37<51:56,  1.27it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.27it/s]
overall:   1%|          | 47/4000 [00:38<51:32,  1.28it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.28it/s]
overall:   1%|          | 48/4000 [00:39<51:25,  1.28it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.28it/s]
overall:   1%|          | 49/4000 [00:39<51:17,  1.28it/s]

train:  32%|███▏      | 12/37 [00:09<00:19,  1.28it/s]train loss: 2.7619502544403076
train acc: 0.14499999582767487

overall:   1%|▏         | 50/4000 [00:40<51:14,  1.28it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.28it/s]
overall:   1%|▏         | 51/4000 [00:41<51:33,  1.28it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.28it/s]
overall:   1%|▏         | 52/4000 [00:42<51:24,  1.28it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.28it/s]
overall:   1%|▏         | 53/4000 [00:42<51:11,  1.29it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.29it/s]
overall:   1%|▏         | 54/4000 [00:43<51:04,  1.29it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.29it/s]
overall:   1%|▏         | 55/4000 [00:44<50:56,  1.29it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.29it/s]
overall:   1%|▏         | 56/4000 [00:45<51:01,  1.29it/s]

train:  51%|█████▏    | 19/37 [00:15<00:13,  1.29it/s]
overall:   1%|▏         | 57/4000 [00:46<51:01,  1.29it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.29it/s]
overall:   1%|▏         | 58/4000 [00:46<50:56,  1.29it/s]

train:  57%|█████▋    | 21/37 [00:16<00:12,  1.29it/s]
overall:   1%|▏         | 59/4000 [00:47<50:48,  1.29it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.29it/s]train loss: 2.729332447052002
train acc: 0.1459999978542328

overall:   2%|▏         | 60/4000 [00:48<50:56,  1.29it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.29it/s]
overall:   2%|▏         | 61/4000 [00:49<50:56,  1.29it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.29it/s]
overall:   2%|▏         | 62/4000 [00:49<50:57,  1.29it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.29it/s]
overall:   2%|▏         | 63/4000 [00:50<53:51,  1.22it/s]

train:  70%|███████   | 26/37 [00:21<00:09,  1.22it/s]
overall:   2%|▏         | 64/4000 [00:51<53:01,  1.24it/s]

train:  73%|███████▎  | 27/37 [00:21<00:08,  1.24it/s]
overall:   2%|▏         | 65/4000 [00:52<52:18,  1.25it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.25it/s]
overall:   2%|▏         | 66/4000 [00:53<51:53,  1.26it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.26it/s]
overall:   2%|▏         | 67/4000 [00:53<51:36,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:   2%|▏         | 68/4000 [00:54<51:26,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.27it/s]
overall:   2%|▏         | 69/4000 [00:55<51:12,  1.28it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.28it/s]train loss: 2.680602788925171
train acc: 0.19499999284744263

overall:   2%|▏         | 70/4000 [00:56<51:17,  1.28it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.28it/s]
overall:   2%|▏         | 71/4000 [00:57<51:01,  1.28it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.28it/s]
overall:   2%|▏         | 72/4000 [00:57<51:02,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:27<00:01,  1.28it/s]
overall:   2%|▏         | 73/4000 [00:58<50:57,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:   2%|▏         | 74/4000 [00:59<49:08,  1.33it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.33it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:   2%|▏         | 75/4000 [01:00<1:06:06,  1.01s/it]

train:   3%|▎         | 1/37 [00:01<00:54,  1.52s/it]
overall:   2%|▏         | 76/4000 [01:01<1:01:50,  1.06it/s]

train:   5%|▌         | 2/37 [00:02<00:38,  1.09s/it]
overall:   2%|▏         | 77/4000 [01:02<58:33,  1.12it/s]  

train:   8%|▊         | 3/37 [00:03<00:32,  1.05it/s]
overall:   2%|▏         | 78/4000 [01:03<56:07,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:29,  1.14it/s]
overall:   2%|▏         | 79/4000 [01:04<54:31,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]train loss: 2.616130828857422
train acc: 0.26619434356689453

overall:   2%|▏         | 80/4000 [01:04<53:27,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:   2%|▏         | 81/4000 [01:05<52:40,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.24it/s]
overall:   2%|▏         | 82/4000 [01:06<52:08,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:   2%|▏         | 83/4000 [01:07<51:44,  1.26it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.26it/s]
overall:   2%|▏         | 84/4000 [01:07<51:29,  1.27it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.27it/s]
overall:   2%|▏         | 85/4000 [01:08<51:15,  1.27it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.27it/s]
overall:   2%|▏         | 86/4000 [01:09<51:07,  1.28it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.28it/s]
overall:   2%|▏         | 87/4000 [01:10<51:09,  1.27it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.28it/s]
overall:   2%|▏         | 88/4000 [01:11<51:19,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:   2%|▏         | 89/4000 [01:11<51:14,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]train loss: 2.585419178009033
train acc: 0.23800000548362732

overall:   2%|▏         | 90/4000 [01:12<51:07,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:   2%|▏         | 91/4000 [01:13<50:56,  1.28it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.28it/s]
overall:   2%|▏         | 92/4000 [01:14<50:57,  1.28it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.28it/s]
overall:   2%|▏         | 93/4000 [01:14<50:46,  1.28it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.28it/s]
overall:   2%|▏         | 94/4000 [01:15<50:46,  1.28it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.28it/s]
overall:   2%|▏         | 95/4000 [01:16<50:45,  1.28it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.28it/s]
overall:   2%|▏         | 96/4000 [01:17<50:46,  1.28it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.28it/s]
overall:   2%|▏         | 97/4000 [01:18<50:46,  1.28it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.28it/s]
overall:   2%|▏         | 98/4000 [01:18<50:38,  1.28it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.28it/s]
overall:   2%|▏         | 99/4000 [01:19<50:46,  1.28it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.28it/s]train loss: 2.541546106338501
train acc: 0.2680000066757202

overall:   2%|▎         | 100/4000 [01:20<51:09,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:   3%|▎         | 101/4000 [01:21<51:05,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:   3%|▎         | 102/4000 [01:22<50:48,  1.28it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.28it/s]
overall:   3%|▎         | 103/4000 [01:22<50:48,  1.28it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.28it/s]
overall:   3%|▎         | 104/4000 [01:23<50:43,  1.28it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.28it/s]
overall:   3%|▎         | 105/4000 [01:24<50:39,  1.28it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.28it/s]
overall:   3%|▎         | 106/4000 [01:25<50:37,  1.28it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.28it/s]
overall:   3%|▎         | 107/4000 [01:25<50:36,  1.28it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.28it/s]
overall:   3%|▎         | 108/4000 [01:26<50:34,  1.28it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.28it/s]
overall:   3%|▎         | 109/4000 [01:27<50:31,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]train loss: 2.525893449783325
train acc: 0.28700000047683716

overall:   3%|▎         | 110/4000 [01:28<50:33,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:   3%|▎         | 111/4000 [01:28<48:46,  1.33it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.33it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:   3%|▎         | 112/4000 [01:30<1:03:51,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:   3%|▎         | 113/4000 [01:31<1:00:07,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:   3%|▎         | 114/4000 [01:32<57:13,  1.13it/s]  

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:   3%|▎         | 115/4000 [01:32<55:15,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:   3%|▎         | 116/4000 [01:33<53:52,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:   3%|▎         | 117/4000 [01:34<52:59,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:   3%|▎         | 118/4000 [01:35<52:20,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:   3%|▎         | 119/4000 [01:35<51:54,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]train loss: 2.4630134105682373
train acc: 0.2753036320209503

overall:   3%|▎         | 120/4000 [01:36<51:31,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:   3%|▎         | 121/4000 [01:37<51:07,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:   3%|▎         | 122/4000 [01:38<50:57,  1.27it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.27it/s]
overall:   3%|▎         | 123/4000 [01:39<50:44,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:   3%|▎         | 124/4000 [01:39<51:05,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.26it/s]
overall:   3%|▎         | 125/4000 [01:40<50:55,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:   3%|▎         | 126/4000 [01:41<50:43,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:   3%|▎         | 127/4000 [01:42<50:34,  1.28it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.28it/s]
overall:   3%|▎         | 128/4000 [01:43<50:28,  1.28it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.28it/s]
overall:   3%|▎         | 129/4000 [01:43<50:21,  1.28it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.28it/s]train loss: 2.4296085834503174
train acc: 0.3019999861717224

overall:   3%|▎         | 130/4000 [01:44<50:28,  1.28it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.28it/s]
overall:   3%|▎         | 131/4000 [01:45<50:21,  1.28it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.28it/s]
overall:   3%|▎         | 132/4000 [01:46<50:23,  1.28it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.28it/s]
overall:   3%|▎         | 133/4000 [01:46<50:20,  1.28it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.28it/s]
overall:   3%|▎         | 134/4000 [01:47<50:24,  1.28it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.28it/s]
overall:   3%|▎         | 135/4000 [01:48<50:26,  1.28it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.28it/s]
overall:   3%|▎         | 136/4000 [01:49<50:43,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:   3%|▎         | 137/4000 [01:50<50:37,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:   3%|▎         | 138/4000 [01:50<50:28,  1.28it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.28it/s]
overall:   3%|▎         | 139/4000 [01:51<50:24,  1.28it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.28it/s]train loss: 2.405127763748169
train acc: 0.30300000309944153

overall:   4%|▎         | 140/4000 [01:52<50:20,  1.28it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.28it/s]
overall:   4%|▎         | 141/4000 [01:53<50:12,  1.28it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.28it/s]
overall:   4%|▎         | 142/4000 [01:53<50:15,  1.28it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.28it/s]
overall:   4%|▎         | 143/4000 [01:54<50:09,  1.28it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.28it/s]
overall:   4%|▎         | 144/4000 [01:55<50:12,  1.28it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.28it/s]
overall:   4%|▎         | 145/4000 [01:56<50:06,  1.28it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.28it/s]
overall:   4%|▎         | 146/4000 [01:57<50:05,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:   4%|▎         | 147/4000 [01:57<50:04,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:   4%|▎         | 148/4000 [01:58<48:16,  1.33it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.33it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:   4%|▎         | 149/4000 [02:00<1:03:53,  1.00it/s]

train:   3%|▎         | 1/37 [00:01<00:52,  1.47s/it]train loss: 2.365628719329834
train acc: 0.3674089014530182

overall:   4%|▍         | 150/4000 [02:00<1:00:07,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.08s/it]
overall:   4%|▍         | 151/4000 [02:01<57:12,  1.12it/s]  

train:   8%|▊         | 3/37 [00:03<00:32,  1.06it/s]
overall:   4%|▍         | 152/4000 [02:02<55:17,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:29,  1.13it/s]
overall:   4%|▍         | 153/4000 [02:03<53:47,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:   4%|▍         | 154/4000 [02:04<52:39,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:   4%|▍         | 155/4000 [02:04<51:53,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:   4%|▍         | 156/4000 [02:05<51:18,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:   4%|▍         | 157/4000 [02:06<50:53,  1.26it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.26it/s]
overall:   4%|▍         | 158/4000 [02:07<50:42,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:   4%|▍         | 159/4000 [02:07<50:27,  1.27it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.27it/s]train loss: 2.329458475112915
train acc: 0.37700000405311584

overall:   4%|▍         | 160/4000 [02:08<50:25,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:   4%|▍         | 161/4000 [02:09<50:26,  1.27it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.27it/s]
overall:   4%|▍         | 162/4000 [02:10<50:33,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:   4%|▍         | 163/4000 [02:11<50:22,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:   4%|▍         | 164/4000 [02:11<50:29,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:   4%|▍         | 165/4000 [02:12<50:23,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:   4%|▍         | 166/4000 [02:13<50:13,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:   4%|▍         | 167/4000 [02:14<50:16,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:   4%|▍         | 168/4000 [02:15<50:13,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:   4%|▍         | 169/4000 [02:15<50:05,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]train loss: 2.293951988220215
train acc: 0.35899999737739563

overall:   4%|▍         | 170/4000 [02:16<50:02,  1.28it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.28it/s]
overall:   4%|▍         | 171/4000 [02:17<49:58,  1.28it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.28it/s]
overall:   4%|▍         | 172/4000 [02:18<49:57,  1.28it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.28it/s]
overall:   4%|▍         | 173/4000 [02:18<49:51,  1.28it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.28it/s]
overall:   4%|▍         | 174/4000 [02:19<50:09,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:   4%|▍         | 175/4000 [02:20<50:04,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:   4%|▍         | 176/4000 [02:21<50:11,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:   4%|▍         | 177/4000 [02:22<50:10,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:   4%|▍         | 178/4000 [02:22<49:59,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:   4%|▍         | 179/4000 [02:23<49:52,  1.28it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.28it/s]train loss: 2.27599835395813
train acc: 0.35199999809265137

overall:   4%|▍         | 180/4000 [02:24<49:47,  1.28it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.28it/s]
overall:   5%|▍         | 181/4000 [02:25<49:46,  1.28it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.28it/s]
overall:   5%|▍         | 182/4000 [02:26<49:48,  1.28it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.28it/s]
overall:   5%|▍         | 183/4000 [02:26<49:57,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:   5%|▍         | 184/4000 [02:27<49:50,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:   5%|▍         | 185/4000 [02:28<48:02,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:   5%|▍         | 186/4000 [02:29<1:01:57,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.38s/it]
overall:   5%|▍         | 187/4000 [02:30<58:18,  1.09it/s]  

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:   5%|▍         | 188/4000 [02:31<55:59,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:   5%|▍         | 189/4000 [02:32<54:10,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]train loss: 2.223191499710083
train acc: 0.39777326583862305

overall:   5%|▍         | 190/4000 [02:32<52:52,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:   5%|▍         | 191/4000 [02:33<52:02,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:   5%|▍         | 192/4000 [02:34<51:34,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:   5%|▍         | 193/4000 [02:35<51:00,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:   5%|▍         | 194/4000 [02:36<50:31,  1.26it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.26it/s]
overall:   5%|▍         | 195/4000 [02:36<50:15,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:   5%|▍         | 196/4000 [02:37<50:00,  1.27it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.27it/s]
overall:   5%|▍         | 197/4000 [02:38<49:52,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:   5%|▍         | 198/4000 [02:39<49:57,  1.27it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.27it/s]
overall:   5%|▍         | 199/4000 [02:40<49:52,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]train loss: 2.219931125640869
train acc: 0.40799999237060547

overall:   5%|▌         | 200/4000 [02:40<50:03,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:   5%|▌         | 201/4000 [02:41<49:48,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:   5%|▌         | 202/4000 [02:42<49:40,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:   5%|▌         | 203/4000 [02:43<49:32,  1.28it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.28it/s]
overall:   5%|▌         | 204/4000 [02:43<49:46,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:   5%|▌         | 205/4000 [02:44<49:44,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:   5%|▌         | 206/4000 [02:45<49:31,  1.28it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.28it/s]
overall:   5%|▌         | 207/4000 [02:46<49:35,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:   5%|▌         | 208/4000 [02:47<49:26,  1.28it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.28it/s]
overall:   5%|▌         | 209/4000 [02:47<49:24,  1.28it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.28it/s]train loss: 2.180952310562134
train acc: 0.41499999165534973

overall:   5%|▌         | 210/4000 [02:48<49:31,  1.28it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.28it/s]
overall:   5%|▌         | 211/4000 [02:49<49:34,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:   5%|▌         | 212/4000 [02:50<49:40,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:   5%|▌         | 213/4000 [02:50<49:38,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:   5%|▌         | 214/4000 [02:51<49:31,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:   5%|▌         | 215/4000 [02:52<49:24,  1.28it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.28it/s]
overall:   5%|▌         | 216/4000 [02:53<49:36,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.27it/s]
overall:   5%|▌         | 217/4000 [02:54<49:30,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:   5%|▌         | 218/4000 [02:54<49:28,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:   5%|▌         | 219/4000 [02:55<49:21,  1.28it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.28it/s]train loss: 2.1718151569366455
train acc: 0.4189999997615814

overall:   6%|▌         | 220/4000 [02:56<49:19,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:   6%|▌         | 221/4000 [02:57<49:12,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:   6%|▌         | 222/4000 [02:57<47:29,  1.33it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.33it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:   6%|▌         | 223/4000 [02:59<1:01:59,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:   6%|▌         | 224/4000 [03:00<58:27,  1.08it/s]  

train:   5%|▌         | 2/37 [00:02<00:36,  1.06s/it]
overall:   6%|▌         | 225/4000 [03:01<55:44,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:   6%|▌         | 226/4000 [03:01<53:40,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:   6%|▌         | 227/4000 [03:02<52:21,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:   6%|▌         | 228/4000 [03:03<51:21,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:   6%|▌         | 229/4000 [03:04<50:41,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.24it/s]train loss: 2.1427693367004395
train acc: 0.40080970525741577

overall:   6%|▌         | 230/4000 [03:04<50:25,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:   6%|▌         | 231/4000 [03:05<50:04,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:   6%|▌         | 232/4000 [03:06<49:51,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:   6%|▌         | 233/4000 [03:07<49:43,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:   6%|▌         | 234/4000 [03:08<49:31,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:   6%|▌         | 235/4000 [03:08<49:37,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.26it/s]
overall:   6%|▌         | 236/4000 [03:09<49:34,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:   6%|▌         | 237/4000 [03:10<49:24,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:   6%|▌         | 238/4000 [03:11<49:15,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:   6%|▌         | 239/4000 [03:12<49:13,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]train loss: 2.1025094985961914
train acc: 0.43799999356269836

overall:   6%|▌         | 240/4000 [03:12<49:16,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:   6%|▌         | 241/4000 [03:13<49:08,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:   6%|▌         | 242/4000 [03:14<49:06,  1.28it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.28it/s]
overall:   6%|▌         | 243/4000 [03:15<49:02,  1.28it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.28it/s]
overall:   6%|▌         | 244/4000 [03:15<49:10,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:   6%|▌         | 245/4000 [03:16<49:09,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:   6%|▌         | 246/4000 [03:17<49:03,  1.28it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.28it/s]
overall:   6%|▌         | 247/4000 [03:18<49:15,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:   6%|▌         | 248/4000 [03:19<49:08,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:   6%|▌         | 249/4000 [03:19<49:02,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]train loss: 2.0740864276885986
train acc: 0.46000000834465027

overall:   6%|▋         | 250/4000 [03:20<48:59,  1.28it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.28it/s]
overall:   6%|▋         | 251/4000 [03:21<48:58,  1.28it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.28it/s]
overall:   6%|▋         | 252/4000 [03:22<49:00,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:   6%|▋         | 253/4000 [03:23<49:01,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.27it/s]
overall:   6%|▋         | 254/4000 [03:23<48:59,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:   6%|▋         | 255/4000 [03:24<49:01,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:   6%|▋         | 256/4000 [03:25<49:00,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:   6%|▋         | 257/4000 [03:26<49:04,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:   6%|▋         | 258/4000 [03:26<49:05,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:   6%|▋         | 259/4000 [03:27<47:18,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]train loss: 2.064149856567383
train acc: 0.4554655849933624

overall:   6%|▋         | 260/4000 [03:29<1:01:15,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:50,  1.41s/it]
overall:   7%|▋         | 261/4000 [03:29<57:42,  1.08it/s]  

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:   7%|▋         | 262/4000 [03:30<55:07,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:   7%|▋         | 263/4000 [03:31<53:18,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:   7%|▋         | 264/4000 [03:32<52:20,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:   7%|▋         | 265/4000 [03:33<51:17,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:   7%|▋         | 266/4000 [03:33<50:37,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:   7%|▋         | 267/4000 [03:34<50:11,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:   7%|▋         | 268/4000 [03:35<49:42,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:   7%|▋         | 269/4000 [03:36<49:33,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]train loss: 2.0248913764953613
train acc: 0.4519999921321869

overall:   7%|▋         | 270/4000 [03:37<49:18,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:   7%|▋         | 271/4000 [03:37<49:05,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:   7%|▋         | 272/4000 [03:38<49:02,  1.27it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.27it/s]
overall:   7%|▋         | 273/4000 [03:39<49:05,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:   7%|▋         | 274/4000 [03:40<49:07,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:   7%|▋         | 275/4000 [03:40<49:00,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:   7%|▋         | 276/4000 [03:41<49:08,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:   7%|▋         | 277/4000 [03:42<49:02,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.27it/s]
overall:   7%|▋         | 278/4000 [03:43<48:54,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:   7%|▋         | 279/4000 [03:44<48:52,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]train loss: 2.006133556365967
train acc: 0.44699999690055847

overall:   7%|▋         | 280/4000 [03:44<48:46,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:   7%|▋         | 281/4000 [03:45<48:48,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:   7%|▋         | 282/4000 [03:46<48:47,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:   7%|▋         | 283/4000 [03:47<48:39,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:   7%|▋         | 284/4000 [03:48<48:38,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:   7%|▋         | 285/4000 [03:48<48:46,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:   7%|▋         | 286/4000 [03:49<48:39,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:   7%|▋         | 287/4000 [03:50<48:35,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:   7%|▋         | 288/4000 [03:51<48:49,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:   7%|▋         | 289/4000 [03:52<48:42,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]train loss: 1.9926176071166992
train acc: 0.46399998664855957

overall:   7%|▋         | 290/4000 [03:52<48:45,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:   7%|▋         | 291/4000 [03:53<48:36,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:   7%|▋         | 292/4000 [03:54<48:32,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:   7%|▋         | 293/4000 [03:55<48:37,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:   7%|▋         | 294/4000 [03:55<48:34,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:   7%|▋         | 295/4000 [03:56<48:28,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:   7%|▋         | 296/4000 [03:57<46:42,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:   7%|▋         | 297/4000 [03:58<1:00:56,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:   7%|▋         | 298/4000 [03:59<57:26,  1.07it/s]  

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:   7%|▋         | 299/4000 [04:00<54:43,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]train loss: 1.9597642421722412
train acc: 0.4787449538707733

overall:   8%|▊         | 300/4000 [04:01<52:57,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:   8%|▊         | 301/4000 [04:02<51:39,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:   8%|▊         | 302/4000 [04:02<50:43,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:   8%|▊         | 303/4000 [04:03<49:58,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:   8%|▊         | 304/4000 [04:04<49:29,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:   8%|▊         | 305/4000 [04:05<49:17,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:   8%|▊         | 306/4000 [04:06<48:57,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:   8%|▊         | 307/4000 [04:06<48:42,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:   8%|▊         | 308/4000 [04:07<48:37,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:   8%|▊         | 309/4000 [04:08<48:53,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]train loss: 1.9286766052246094
train acc: 0.4880000054836273

overall:   8%|▊         | 310/4000 [04:09<48:51,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:   8%|▊         | 311/4000 [04:10<48:42,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:   8%|▊         | 312/4000 [04:10<48:33,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:   8%|▊         | 313/4000 [04:11<48:26,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:   8%|▊         | 314/4000 [04:12<48:32,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.27it/s]
overall:   8%|▊         | 315/4000 [04:13<48:21,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:   8%|▊         | 316/4000 [04:13<48:17,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:   8%|▊         | 317/4000 [04:14<48:23,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:   8%|▊         | 318/4000 [04:15<48:15,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:   8%|▊         | 319/4000 [04:16<48:10,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]train loss: 1.9202302694320679
train acc: 0.503000020980835

overall:   8%|▊         | 320/4000 [04:17<48:17,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:   8%|▊         | 321/4000 [04:17<48:28,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:   8%|▊         | 322/4000 [04:18<48:22,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:   8%|▊         | 323/4000 [04:19<48:13,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:   8%|▊         | 324/4000 [04:20<48:08,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:   8%|▊         | 325/4000 [04:21<48:04,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:   8%|▊         | 326/4000 [04:21<48:07,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:   8%|▊         | 327/4000 [04:22<48:01,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:   8%|▊         | 328/4000 [04:23<47:57,  1.28it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.28it/s]
overall:   8%|▊         | 329/4000 [04:24<48:08,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]train loss: 1.9333356618881226
train acc: 0.46700000762939453

overall:   8%|▊         | 330/4000 [04:24<48:04,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:   8%|▊         | 331/4000 [04:25<47:56,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:   8%|▊         | 332/4000 [04:26<47:56,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:   8%|▊         | 333/4000 [04:27<46:22,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:   8%|▊         | 334/4000 [04:28<1:00:10,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:50,  1.42s/it]
overall:   8%|▊         | 335/4000 [04:29<56:45,  1.08it/s]  

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:   8%|▊         | 336/4000 [04:30<54:02,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:   8%|▊         | 337/4000 [04:31<52:09,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:   8%|▊         | 338/4000 [04:31<50:53,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:   8%|▊         | 339/4000 [04:32<50:05,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]train loss: 1.8459765911102295
train acc: 0.5121457576751709

overall:   8%|▊         | 340/4000 [04:33<49:42,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:   9%|▊         | 341/4000 [04:34<49:09,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:   9%|▊         | 342/4000 [04:35<48:50,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:   9%|▊         | 343/4000 [04:35<48:33,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:   9%|▊         | 344/4000 [04:36<48:15,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:   9%|▊         | 345/4000 [04:37<48:08,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:   9%|▊         | 346/4000 [04:38<48:05,  1.27it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.27it/s]
overall:   9%|▊         | 347/4000 [04:38<48:10,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:   9%|▊         | 348/4000 [04:39<48:04,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:   9%|▊         | 349/4000 [04:40<47:56,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]train loss: 1.8928149938583374
train acc: 0.503000020980835

overall:   9%|▉         | 350/4000 [04:41<48:00,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:   9%|▉         | 351/4000 [04:42<47:52,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:   9%|▉         | 352/4000 [04:42<48:05,  1.26it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.26it/s]
overall:   9%|▉         | 353/4000 [04:43<47:51,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:   9%|▉         | 354/4000 [04:44<47:46,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:   9%|▉         | 355/4000 [04:45<47:51,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:   9%|▉         | 356/4000 [04:46<47:47,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:   9%|▉         | 357/4000 [04:46<47:39,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:   9%|▉         | 358/4000 [04:47<47:41,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:   9%|▉         | 359/4000 [04:48<47:45,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]train loss: 1.8133199214935303
train acc: 0.515999972820282

overall:   9%|▉         | 360/4000 [04:49<47:38,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:   9%|▉         | 361/4000 [04:49<47:33,  1.28it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.28it/s]
overall:   9%|▉         | 362/4000 [04:50<47:41,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:   9%|▉         | 363/4000 [04:51<47:37,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:   9%|▉         | 364/4000 [04:52<47:49,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:   9%|▉         | 365/4000 [04:53<47:43,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:   9%|▉         | 366/4000 [04:53<47:39,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:   9%|▉         | 367/4000 [04:54<47:35,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:   9%|▉         | 368/4000 [04:55<47:36,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:   9%|▉         | 369/4000 [04:56<47:30,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]train loss: 1.855062484741211
train acc: 0.4858299493789673

overall:   9%|▉         | 370/4000 [04:56<45:51,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:   9%|▉         | 371/4000 [04:58<59:40,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.42s/it]
overall:   9%|▉         | 372/4000 [04:59<56:03,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:   9%|▉         | 373/4000 [05:00<53:40,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:   9%|▉         | 374/4000 [05:00<51:46,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:   9%|▉         | 375/4000 [05:01<50:35,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:   9%|▉         | 376/4000 [05:02<49:37,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:   9%|▉         | 377/4000 [05:03<49:03,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:   9%|▉         | 378/4000 [05:04<48:34,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:   9%|▉         | 379/4000 [05:04<48:26,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]train loss: 1.8343219757080078
train acc: 0.5220000147819519

overall:  10%|▉         | 380/4000 [05:05<48:07,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  10%|▉         | 381/4000 [05:06<48:02,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  10%|▉         | 382/4000 [05:07<47:44,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  10%|▉         | 383/4000 [05:07<47:52,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  10%|▉         | 384/4000 [05:08<47:43,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  10%|▉         | 385/4000 [05:09<47:42,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  10%|▉         | 386/4000 [05:10<47:29,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  10%|▉         | 387/4000 [05:11<47:27,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  10%|▉         | 388/4000 [05:11<47:33,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.27it/s]
overall:  10%|▉         | 389/4000 [05:12<47:26,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]train loss: 1.7813411951065063
train acc: 0.5329999923706055

overall:  10%|▉         | 390/4000 [05:13<47:29,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  10%|▉         | 391/4000 [05:14<49:46,  1.21it/s]

train:  57%|█████▋    | 21/37 [00:17<00:13,  1.21it/s]
overall:  10%|▉         | 392/4000 [05:15<49:02,  1.23it/s]

train:  59%|█████▉    | 22/37 [00:18<00:12,  1.23it/s]
overall:  10%|▉         | 393/4000 [05:15<48:21,  1.24it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.24it/s]
overall:  10%|▉         | 394/4000 [05:16<47:57,  1.25it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.25it/s]
overall:  10%|▉         | 395/4000 [05:17<47:49,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  10%|▉         | 396/4000 [05:18<47:38,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  10%|▉         | 397/4000 [05:19<47:28,  1.26it/s]

train:  73%|███████▎  | 27/37 [00:22<00:07,  1.26it/s]
overall:  10%|▉         | 398/4000 [05:19<47:22,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  10%|▉         | 399/4000 [05:20<47:15,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]train loss: 1.7517616748809814
train acc: 0.5460000038146973

overall:  10%|█         | 400/4000 [05:21<47:10,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  10%|█         | 401/4000 [05:22<47:04,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  10%|█         | 402/4000 [05:23<47:05,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  10%|█         | 403/4000 [05:23<47:09,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  10%|█         | 404/4000 [05:24<47:04,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  10%|█         | 405/4000 [05:25<46:57,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:  10%|█         | 406/4000 [05:26<47:06,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  10%|█         | 407/4000 [05:26<45:24,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  10%|█         | 408/4000 [05:28<58:55,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.42s/it]
overall:  10%|█         | 409/4000 [05:29<55:23,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]train loss: 1.767795205116272
train acc: 0.5141700506210327

overall:  10%|█         | 410/4000 [05:29<52:56,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.07it/s]
overall:  10%|█         | 411/4000 [05:30<51:06,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  10%|█         | 412/4000 [05:31<49:56,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  10%|█         | 413/4000 [05:32<49:02,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  10%|█         | 414/4000 [05:33<48:23,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  10%|█         | 415/4000 [05:33<47:52,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  10%|█         | 416/4000 [05:34<47:39,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  10%|█         | 417/4000 [05:35<47:24,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  10%|█         | 418/4000 [05:36<47:26,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  10%|█         | 419/4000 [05:37<47:10,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]train loss: 1.6703230142593384
train acc: 0.5830000042915344

overall:  10%|█         | 420/4000 [05:37<47:25,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  11%|█         | 421/4000 [05:38<47:13,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  11%|█         | 422/4000 [05:39<47:10,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  11%|█         | 423/4000 [05:40<46:56,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  11%|█         | 424/4000 [05:40<46:53,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  11%|█         | 425/4000 [05:41<46:45,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  11%|█         | 426/4000 [05:42<46:45,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  11%|█         | 427/4000 [05:43<46:44,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  11%|█         | 428/4000 [05:44<46:41,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  11%|█         | 429/4000 [05:44<46:42,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]train loss: 1.7344284057617188
train acc: 0.5450000166893005

overall:  11%|█         | 430/4000 [05:45<46:50,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  11%|█         | 431/4000 [05:46<46:43,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  11%|█         | 432/4000 [05:47<46:58,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  11%|█         | 433/4000 [05:48<46:52,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  11%|█         | 434/4000 [05:48<46:44,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  11%|█         | 435/4000 [05:49<46:42,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  11%|█         | 436/4000 [05:50<46:39,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  11%|█         | 437/4000 [05:51<46:36,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  11%|█         | 438/4000 [05:51<46:33,  1.28it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.28it/s]
overall:  11%|█         | 439/4000 [05:52<46:27,  1.28it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.28it/s]train loss: 1.732084035873413
train acc: 0.5070000290870667

overall:  11%|█         | 440/4000 [05:53<46:27,  1.28it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.28it/s]
overall:  11%|█         | 441/4000 [05:54<46:32,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  11%|█         | 442/4000 [05:55<46:31,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  11%|█         | 443/4000 [05:55<46:27,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:  11%|█         | 444/4000 [05:56<44:49,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  11%|█         | 445/4000 [05:58<58:39,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  11%|█         | 446/4000 [05:58<56:01,  1.06it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.09s/it]
overall:  11%|█         | 447/4000 [05:59<53:09,  1.11it/s]

train:   8%|▊         | 3/37 [00:03<00:32,  1.05it/s]
overall:  11%|█         | 448/4000 [06:00<51:09,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:29,  1.13it/s]
overall:  11%|█         | 449/4000 [06:01<49:51,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]train loss: 1.6827234029769897
train acc: 0.5657894611358643

overall:  11%|█▏        | 450/4000 [06:02<48:54,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.20it/s]
overall:  11%|█▏        | 451/4000 [06:02<48:10,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  11%|█▏        | 452/4000 [06:03<47:39,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  11%|█▏        | 453/4000 [06:04<47:14,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  11%|█▏        | 454/4000 [06:05<47:03,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  11%|█▏        | 455/4000 [06:06<46:52,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  11%|█▏        | 456/4000 [06:06<46:44,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  11%|█▏        | 457/4000 [06:07<46:43,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.26it/s]
overall:  11%|█▏        | 458/4000 [06:08<46:44,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  11%|█▏        | 459/4000 [06:09<46:36,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]train loss: 1.6808359622955322
train acc: 0.5339999794960022

overall:  12%|█▏        | 460/4000 [06:09<46:30,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  12%|█▏        | 461/4000 [06:10<46:33,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  12%|█▏        | 462/4000 [06:11<46:34,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.27it/s]
overall:  12%|█▏        | 463/4000 [06:12<46:33,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  12%|█▏        | 464/4000 [06:13<46:28,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  12%|█▏        | 465/4000 [06:13<46:22,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  12%|█▏        | 466/4000 [06:14<46:31,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  12%|█▏        | 467/4000 [06:15<46:22,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  12%|█▏        | 468/4000 [06:16<46:24,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  12%|█▏        | 469/4000 [06:17<46:20,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]train loss: 1.6357734203338623
train acc: 0.578000009059906

overall:  12%|█▏        | 470/4000 [06:17<46:24,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  12%|█▏        | 471/4000 [06:18<46:23,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  12%|█▏        | 472/4000 [06:19<46:13,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  12%|█▏        | 473/4000 [06:20<46:19,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  12%|█▏        | 474/4000 [06:21<46:17,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  12%|█▏        | 475/4000 [06:21<46:14,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  12%|█▏        | 476/4000 [06:22<46:08,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  12%|█▏        | 477/4000 [06:23<46:11,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  12%|█▏        | 478/4000 [06:24<46:06,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  12%|█▏        | 479/4000 [06:24<46:08,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]train loss: 1.6380889415740967
train acc: 0.5809999704360962

overall:  12%|█▏        | 480/4000 [06:25<46:13,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  12%|█▏        | 481/4000 [06:26<44:30,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  12%|█▏        | 482/4000 [06:27<58:06,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  12%|█▏        | 483/4000 [06:28<54:37,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.06s/it]
overall:  12%|█▏        | 484/4000 [06:29<51:54,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.08it/s]
overall:  12%|█▏        | 485/4000 [06:30<50:09,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  12%|█▏        | 486/4000 [06:31<48:54,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  12%|█▏        | 487/4000 [06:31<48:02,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  12%|█▏        | 488/4000 [06:32<47:39,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  12%|█▏        | 489/4000 [06:33<47:08,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]train loss: 1.6290632486343384
train acc: 0.5880566835403442

overall:  12%|█▏        | 490/4000 [06:34<46:54,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  12%|█▏        | 491/4000 [06:35<46:34,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  12%|█▏        | 492/4000 [06:35<46:29,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  12%|█▏        | 493/4000 [06:36<46:18,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  12%|█▏        | 494/4000 [06:37<46:23,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  12%|█▏        | 495/4000 [06:38<46:27,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  12%|█▏        | 496/4000 [06:39<46:13,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  12%|█▏        | 497/4000 [06:39<46:03,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  12%|█▏        | 498/4000 [06:40<46:07,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  12%|█▏        | 499/4000 [06:41<45:59,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]train loss: 1.6168912649154663
train acc: 0.5630000233650208

overall:  12%|█▎        | 500/4000 [06:42<45:53,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  13%|█▎        | 501/4000 [06:42<45:46,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  13%|█▎        | 502/4000 [06:43<45:43,  1.28it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.28it/s]
overall:  13%|█▎        | 503/4000 [06:44<45:46,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  13%|█▎        | 504/4000 [06:45<45:43,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  13%|█▎        | 505/4000 [06:46<45:39,  1.28it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.28it/s]
overall:  13%|█▎        | 506/4000 [06:46<45:51,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  13%|█▎        | 507/4000 [06:47<45:50,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  13%|█▎        | 508/4000 [06:48<45:42,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  13%|█▎        | 509/4000 [06:49<45:40,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]train loss: 1.5872976779937744
train acc: 0.5740000009536743

overall:  13%|█▎        | 510/4000 [06:50<45:45,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  13%|█▎        | 511/4000 [06:50<45:37,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  13%|█▎        | 512/4000 [06:51<45:30,  1.28it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.28it/s]
overall:  13%|█▎        | 513/4000 [06:52<45:28,  1.28it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.28it/s]
overall:  13%|█▎        | 514/4000 [06:53<45:23,  1.28it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.28it/s]
overall:  13%|█▎        | 515/4000 [06:53<45:24,  1.28it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.28it/s]
overall:  13%|█▎        | 516/4000 [06:54<45:31,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:  13%|█▎        | 517/4000 [06:55<45:29,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:  13%|█▎        | 518/4000 [06:56<43:53,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  13%|█▎        | 519/4000 [06:57<57:14,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]train loss: 1.6257283687591553
train acc: 0.5526315569877625

overall:  13%|█▎        | 520/4000 [06:58<53:57,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.06s/it]
overall:  13%|█▎        | 521/4000 [06:59<51:22,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  13%|█▎        | 522/4000 [07:00<49:34,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  13%|█▎        | 523/4000 [07:00<48:22,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  13%|█▎        | 524/4000 [07:01<47:41,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  13%|█▎        | 525/4000 [07:02<47:05,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  13%|█▎        | 526/4000 [07:03<46:34,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  13%|█▎        | 527/4000 [07:04<46:08,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  13%|█▎        | 528/4000 [07:04<45:59,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  13%|█▎        | 529/4000 [07:05<45:54,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]train loss: 1.5596257448196411
train acc: 0.593999981880188

overall:  13%|█▎        | 530/4000 [07:06<45:41,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  13%|█▎        | 531/4000 [07:07<45:40,  1.27it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.27it/s]
overall:  13%|█▎        | 532/4000 [07:07<45:45,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  13%|█▎        | 533/4000 [07:08<45:37,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  13%|█▎        | 534/4000 [07:09<45:31,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  13%|█▎        | 535/4000 [07:10<45:23,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  13%|█▎        | 536/4000 [07:11<45:24,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  13%|█▎        | 537/4000 [07:11<45:27,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  13%|█▎        | 538/4000 [07:12<45:23,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  13%|█▎        | 539/4000 [07:13<45:24,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]train loss: 1.5313136577606201
train acc: 0.574999988079071

overall:  14%|█▎        | 540/4000 [07:14<45:28,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  14%|█▎        | 541/4000 [07:15<45:24,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  14%|█▎        | 542/4000 [07:15<45:27,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  14%|█▎        | 543/4000 [07:16<45:21,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  14%|█▎        | 544/4000 [07:17<45:30,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  14%|█▎        | 545/4000 [07:18<45:33,  1.26it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.26it/s]
overall:  14%|█▎        | 546/4000 [07:18<45:29,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  14%|█▎        | 547/4000 [07:19<45:24,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  14%|█▎        | 548/4000 [07:20<45:24,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  14%|█▎        | 549/4000 [07:21<45:17,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]train loss: 1.5942740440368652
train acc: 0.5640000104904175

overall:  14%|█▍        | 550/4000 [07:22<45:18,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  14%|█▍        | 551/4000 [07:22<45:15,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  14%|█▍        | 552/4000 [07:23<45:27,  1.26it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.26it/s]
overall:  14%|█▍        | 553/4000 [07:24<45:19,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  14%|█▍        | 554/4000 [07:25<45:12,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  14%|█▍        | 555/4000 [07:25<43:40,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  14%|█▍        | 556/4000 [07:27<56:52,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  14%|█▍        | 557/4000 [07:28<53:20,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  14%|█▍        | 558/4000 [07:29<50:58,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  14%|█▍        | 559/4000 [07:29<49:15,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]train loss: 1.5212074518203735
train acc: 0.6133603453636169

overall:  14%|█▍        | 560/4000 [07:30<48:12,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  14%|█▍        | 561/4000 [07:31<47:15,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  14%|█▍        | 562/4000 [07:32<46:39,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  14%|█▍        | 563/4000 [07:33<46:18,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  14%|█▍        | 564/4000 [07:33<45:53,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  14%|█▍        | 565/4000 [07:34<45:36,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  14%|█▍        | 566/4000 [07:35<45:29,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  14%|█▍        | 567/4000 [07:36<45:21,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  14%|█▍        | 568/4000 [07:37<45:34,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  14%|█▍        | 569/4000 [07:37<45:24,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]train loss: 1.525479793548584
train acc: 0.5849999785423279

overall:  14%|█▍        | 570/4000 [07:38<45:16,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  14%|█▍        | 571/4000 [07:39<45:12,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  14%|█▍        | 572/4000 [07:40<45:05,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  14%|█▍        | 573/4000 [07:40<44:53,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  14%|█▍        | 574/4000 [07:41<44:52,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  14%|█▍        | 575/4000 [07:42<45:00,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  14%|█▍        | 576/4000 [07:43<44:55,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  14%|█▍        | 577/4000 [07:44<44:50,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  14%|█▍        | 578/4000 [07:44<44:53,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  14%|█▍        | 579/4000 [07:45<44:50,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]train loss: 1.5030971765518188
train acc: 0.6079999804496765

overall:  14%|█▍        | 580/4000 [07:46<45:02,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  15%|█▍        | 581/4000 [07:47<44:51,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  15%|█▍        | 582/4000 [07:48<44:50,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  15%|█▍        | 583/4000 [07:48<44:48,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  15%|█▍        | 584/4000 [07:49<44:45,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  15%|█▍        | 585/4000 [07:50<44:43,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  15%|█▍        | 586/4000 [07:51<44:37,  1.28it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.28it/s]
overall:  15%|█▍        | 587/4000 [07:51<44:44,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  15%|█▍        | 588/4000 [07:52<44:37,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  15%|█▍        | 589/4000 [07:53<44:36,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]train loss: 1.5050855875015259
train acc: 0.6069999933242798

overall:  15%|█▍        | 590/4000 [07:54<44:37,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  15%|█▍        | 591/4000 [07:55<44:41,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  15%|█▍        | 592/4000 [07:55<43:04,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  15%|█▍        | 593/4000 [07:57<55:02,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:48,  1.36s/it]
overall:  15%|█▍        | 594/4000 [07:58<51:56,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.02s/it]
overall:  15%|█▍        | 595/4000 [07:58<49:53,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  15%|█▍        | 596/4000 [07:59<48:14,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]
overall:  15%|█▍        | 597/4000 [08:00<47:08,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  15%|█▍        | 598/4000 [08:01<46:26,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  15%|█▍        | 599/4000 [08:01<45:50,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.24it/s]train loss: 1.4940640926361084
train acc: 0.6163967847824097

overall:  15%|█▌        | 600/4000 [08:02<45:32,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  15%|█▌        | 601/4000 [08:03<45:20,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  15%|█▌        | 602/4000 [08:04<45:09,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  15%|█▌        | 603/4000 [08:05<44:55,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  15%|█▌        | 604/4000 [08:05<44:53,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  15%|█▌        | 605/4000 [08:06<44:54,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  15%|█▌        | 606/4000 [08:07<44:47,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  15%|█▌        | 607/4000 [08:08<44:38,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  15%|█▌        | 608/4000 [08:09<44:39,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  15%|█▌        | 609/4000 [08:09<44:32,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]train loss: 1.4758509397506714
train acc: 0.6019999980926514

overall:  15%|█▌        | 610/4000 [08:10<44:47,  1.26it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.26it/s]
overall:  15%|█▌        | 611/4000 [08:11<44:35,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  15%|█▌        | 612/4000 [08:12<44:26,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  15%|█▌        | 613/4000 [08:13<44:31,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  15%|█▌        | 614/4000 [08:13<44:19,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  15%|█▌        | 615/4000 [08:14<44:16,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  15%|█▌        | 616/4000 [08:15<44:09,  1.28it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.28it/s]
overall:  15%|█▌        | 617/4000 [08:16<44:22,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  15%|█▌        | 618/4000 [08:16<44:16,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  15%|█▌        | 619/4000 [08:17<44:09,  1.28it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.28it/s]train loss: 1.4591058492660522
train acc: 0.6169999837875366

overall:  16%|█▌        | 620/4000 [08:18<44:21,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  16%|█▌        | 621/4000 [08:19<44:15,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  16%|█▌        | 622/4000 [08:20<44:25,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  16%|█▌        | 623/4000 [08:20<44:19,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.27it/s]
overall:  16%|█▌        | 624/4000 [08:21<44:18,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  16%|█▌        | 625/4000 [08:22<44:22,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  16%|█▌        | 626/4000 [08:23<44:12,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  16%|█▌        | 627/4000 [08:24<44:05,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:  16%|█▌        | 628/4000 [08:24<44:01,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:  16%|█▌        | 629/4000 [08:25<42:32,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]train loss: 1.4675817489624023
train acc: 0.6052631735801697

overall:  16%|█▌        | 630/4000 [08:27<55:24,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.42s/it]
overall:  16%|█▌        | 631/4000 [08:27<52:04,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  16%|█▌        | 632/4000 [08:28<49:43,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  16%|█▌        | 633/4000 [08:29<48:04,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  16%|█▌        | 634/4000 [08:30<46:53,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  16%|█▌        | 635/4000 [08:30<46:01,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  16%|█▌        | 636/4000 [08:31<45:19,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.24it/s]
overall:  16%|█▌        | 637/4000 [08:32<45:06,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  16%|█▌        | 638/4000 [08:33<44:43,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  16%|█▌        | 639/4000 [08:34<44:26,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]train loss: 1.4750114679336548
train acc: 0.6110000014305115

overall:  16%|█▌        | 640/4000 [08:34<44:22,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  16%|█▌        | 641/4000 [08:35<44:14,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  16%|█▌        | 642/4000 [08:36<44:21,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  16%|█▌        | 643/4000 [08:37<44:10,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  16%|█▌        | 644/4000 [08:38<44:03,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  16%|█▌        | 645/4000 [08:38<44:08,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  16%|█▌        | 646/4000 [08:39<44:01,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  16%|█▌        | 647/4000 [08:40<43:52,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  16%|█▌        | 648/4000 [08:41<43:50,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  16%|█▌        | 649/4000 [08:41<43:59,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]train loss: 1.4060790538787842
train acc: 0.6259999871253967

overall:  16%|█▋        | 650/4000 [08:42<43:51,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  16%|█▋        | 651/4000 [08:43<43:45,  1.28it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.28it/s]
overall:  16%|█▋        | 652/4000 [08:44<43:41,  1.28it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.28it/s]
overall:  16%|█▋        | 653/4000 [08:45<43:45,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  16%|█▋        | 654/4000 [08:45<43:50,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  16%|█▋        | 655/4000 [08:46<43:50,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  16%|█▋        | 656/4000 [08:47<43:47,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  16%|█▋        | 657/4000 [08:48<43:50,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  16%|█▋        | 658/4000 [08:49<43:56,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  16%|█▋        | 659/4000 [08:49<43:50,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]train loss: 1.4373937845230103
train acc: 0.6000000238418579

overall:  16%|█▋        | 660/4000 [08:50<43:44,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  17%|█▋        | 661/4000 [08:51<43:50,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  17%|█▋        | 662/4000 [08:52<43:49,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  17%|█▋        | 663/4000 [08:52<43:41,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  17%|█▋        | 664/4000 [08:53<43:46,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  17%|█▋        | 665/4000 [08:54<43:44,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  17%|█▋        | 666/4000 [08:55<42:20,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  17%|█▋        | 667/4000 [08:56<54:30,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.38s/it]
overall:  17%|█▋        | 668/4000 [08:57<51:12,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  17%|█▋        | 669/4000 [08:58<49:02,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]train loss: 1.3889232873916626
train acc: 0.6315789222717285

overall:  17%|█▋        | 670/4000 [08:59<47:57,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:29,  1.13it/s]
overall:  17%|█▋        | 671/4000 [08:59<46:39,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  17%|█▋        | 672/4000 [09:00<45:39,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  17%|█▋        | 673/4000 [09:01<45:02,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  17%|█▋        | 674/4000 [09:02<44:38,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  17%|█▋        | 675/4000 [09:03<44:14,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  17%|█▋        | 676/4000 [09:03<43:59,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  17%|█▋        | 677/4000 [09:04<43:49,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  17%|█▋        | 678/4000 [09:05<43:45,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  17%|█▋        | 679/4000 [09:06<43:37,  1.27it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.27it/s]train loss: 1.404840111732483
train acc: 0.6370000243186951

overall:  17%|█▋        | 680/4000 [09:07<43:43,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  17%|█▋        | 681/4000 [09:07<43:32,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  17%|█▋        | 682/4000 [09:08<43:52,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  17%|█▋        | 683/4000 [09:09<43:50,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  17%|█▋        | 684/4000 [09:10<43:37,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  17%|█▋        | 685/4000 [09:10<43:35,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  17%|█▋        | 686/4000 [09:11<43:29,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  17%|█▋        | 687/4000 [09:12<43:26,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  17%|█▋        | 688/4000 [09:13<43:19,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  17%|█▋        | 689/4000 [09:14<43:18,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]train loss: 1.4191248416900635
train acc: 0.621999979019165

overall:  17%|█▋        | 690/4000 [09:14<43:24,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  17%|█▋        | 691/4000 [09:15<43:20,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  17%|█▋        | 692/4000 [09:16<43:19,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  17%|█▋        | 693/4000 [09:17<43:24,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  17%|█▋        | 694/4000 [09:18<43:38,  1.26it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.26it/s]
overall:  17%|█▋        | 695/4000 [09:18<43:30,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  17%|█▋        | 696/4000 [09:19<43:25,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  17%|█▋        | 697/4000 [09:20<43:22,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  17%|█▋        | 698/4000 [09:21<43:15,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  17%|█▋        | 699/4000 [09:21<43:15,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]train loss: 1.390333890914917
train acc: 0.6399999856948853

overall:  18%|█▊        | 700/4000 [09:22<43:18,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  18%|█▊        | 701/4000 [09:23<43:12,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  18%|█▊        | 702/4000 [09:24<43:11,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  18%|█▊        | 703/4000 [09:25<41:38,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  18%|█▊        | 704/4000 [09:26<52:52,  1.04it/s]

train:   3%|▎         | 1/37 [00:01<00:48,  1.35s/it]
overall:  18%|█▊        | 705/4000 [09:27<50:12,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.03s/it]
overall:  18%|█▊        | 706/4000 [09:28<48:05,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  18%|█▊        | 707/4000 [09:28<46:36,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  18%|█▊        | 708/4000 [09:29<45:32,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.20it/s]
overall:  18%|█▊        | 709/4000 [09:30<44:52,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]train loss: 1.3639509677886963
train acc: 0.6214575171470642

overall:  18%|█▊        | 710/4000 [09:31<44:33,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  18%|█▊        | 711/4000 [09:32<44:07,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  18%|█▊        | 712/4000 [09:32<43:44,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  18%|█▊        | 713/4000 [09:33<43:31,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  18%|█▊        | 714/4000 [09:34<43:26,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  18%|█▊        | 715/4000 [09:35<43:16,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  18%|█▊        | 716/4000 [09:35<43:22,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  18%|█▊        | 717/4000 [09:36<43:21,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  18%|█▊        | 718/4000 [09:37<43:15,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  18%|█▊        | 719/4000 [09:38<43:05,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]train loss: 1.397361397743225
train acc: 0.6399999856948853

overall:  18%|█▊        | 720/4000 [09:39<43:03,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  18%|█▊        | 721/4000 [09:39<43:02,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  18%|█▊        | 722/4000 [09:40<43:06,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  18%|█▊        | 723/4000 [09:41<43:05,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  18%|█▊        | 724/4000 [09:42<42:59,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  18%|█▊        | 725/4000 [09:43<42:53,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  18%|█▊        | 726/4000 [09:43<42:53,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  18%|█▊        | 727/4000 [09:44<42:53,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  18%|█▊        | 728/4000 [09:45<42:58,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  18%|█▊        | 729/4000 [09:46<42:58,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]train loss: 1.3629578351974487
train acc: 0.6349999904632568

overall:  18%|█▊        | 730/4000 [09:46<42:53,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  18%|█▊        | 731/4000 [09:47<42:46,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  18%|█▊        | 732/4000 [09:48<42:44,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  18%|█▊        | 733/4000 [09:49<42:46,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  18%|█▊        | 734/4000 [09:50<42:50,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.27it/s]
overall:  18%|█▊        | 735/4000 [09:50<42:49,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  18%|█▊        | 736/4000 [09:51<42:49,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  18%|█▊        | 737/4000 [09:52<42:39,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  18%|█▊        | 738/4000 [09:53<42:38,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:  18%|█▊        | 739/4000 [09:54<42:33,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]train loss: 1.3298704624176025
train acc: 0.6558704376220703

overall:  18%|█▊        | 740/4000 [09:54<41:08,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  19%|█▊        | 741/4000 [09:56<53:37,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  19%|█▊        | 742/4000 [09:57<50:21,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  19%|█▊        | 743/4000 [09:57<48:01,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.08it/s]
overall:  19%|█▊        | 744/4000 [09:58<46:23,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  19%|█▊        | 745/4000 [09:59<45:12,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  19%|█▊        | 746/4000 [10:00<44:29,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  19%|█▊        | 747/4000 [10:00<44:03,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  19%|█▊        | 748/4000 [10:01<43:36,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  19%|█▊        | 749/4000 [10:02<43:17,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]train loss: 1.3850204944610596
train acc: 0.625

overall:  19%|█▉        | 750/4000 [10:03<43:09,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  19%|█▉        | 751/4000 [10:04<42:54,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  19%|█▉        | 752/4000 [10:04<42:47,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  19%|█▉        | 753/4000 [10:05<43:02,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  19%|█▉        | 754/4000 [10:06<42:53,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  19%|█▉        | 755/4000 [10:07<42:40,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  19%|█▉        | 756/4000 [10:08<42:36,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  19%|█▉        | 757/4000 [10:08<42:37,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  19%|█▉        | 758/4000 [10:09<42:32,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  19%|█▉        | 759/4000 [10:10<42:31,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]train loss: 1.2988922595977783
train acc: 0.6600000262260437

overall:  19%|█▉        | 760/4000 [10:11<42:35,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  19%|█▉        | 761/4000 [10:12<42:31,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  19%|█▉        | 762/4000 [10:12<42:33,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  19%|█▉        | 763/4000 [10:13<42:26,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  19%|█▉        | 764/4000 [10:14<42:25,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  19%|█▉        | 765/4000 [10:15<42:41,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  19%|█▉        | 766/4000 [10:15<42:35,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  19%|█▉        | 767/4000 [10:16<42:27,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  19%|█▉        | 768/4000 [10:17<42:23,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  19%|█▉        | 769/4000 [10:18<42:30,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]train loss: 1.3336749076843262
train acc: 0.6439999938011169

overall:  19%|█▉        | 770/4000 [10:19<42:23,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  19%|█▉        | 771/4000 [10:19<42:24,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  19%|█▉        | 772/4000 [10:20<42:21,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  19%|█▉        | 773/4000 [10:21<42:26,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  19%|█▉        | 774/4000 [10:22<42:22,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  19%|█▉        | 775/4000 [10:23<42:14,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  19%|█▉        | 776/4000 [10:23<42:16,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  19%|█▉        | 777/4000 [10:24<40:50,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  19%|█▉        | 778/4000 [10:25<52:24,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.38s/it]
overall:  19%|█▉        | 779/4000 [10:26<51:26,  1.04it/s]

train:   5%|▌         | 2/37 [00:02<00:38,  1.11s/it]train loss: 1.3123674392700195
train acc: 0.647773265838623

overall:  20%|█▉        | 780/4000 [10:27<48:48,  1.10it/s]

train:   8%|▊         | 3/37 [00:03<00:32,  1.04it/s]
overall:  20%|█▉        | 781/4000 [10:28<46:59,  1.14it/s]

train:  11%|█         | 4/37 [00:03<00:29,  1.11it/s]
overall:  20%|█▉        | 782/4000 [10:29<45:31,  1.18it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.16it/s]
overall:  20%|█▉        | 783/4000 [10:30<44:34,  1.20it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.20it/s]
overall:  20%|█▉        | 784/4000 [10:30<43:56,  1.22it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.22it/s]
overall:  20%|█▉        | 785/4000 [10:31<43:28,  1.23it/s]

train:  22%|██▏       | 8/37 [00:07<00:23,  1.23it/s]
overall:  20%|█▉        | 786/4000 [10:32<42:58,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  20%|█▉        | 787/4000 [10:33<42:43,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  20%|█▉        | 788/4000 [10:34<42:33,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  20%|█▉        | 789/4000 [10:34<42:20,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]train loss: 1.3375743627548218
train acc: 0.6299999952316284

overall:  20%|█▉        | 790/4000 [10:35<42:24,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  20%|█▉        | 791/4000 [10:36<42:16,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  20%|█▉        | 792/4000 [10:37<42:17,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  20%|█▉        | 793/4000 [10:37<42:20,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  20%|█▉        | 794/4000 [10:38<42:19,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  20%|█▉        | 795/4000 [10:39<42:07,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  20%|█▉        | 796/4000 [10:40<42:12,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  20%|█▉        | 797/4000 [10:41<42:03,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  20%|█▉        | 798/4000 [10:41<41:55,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  20%|█▉        | 799/4000 [10:42<41:50,  1.28it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.28it/s]train loss: 1.2860479354858398
train acc: 0.6610000133514404

overall:  20%|██        | 800/4000 [10:43<41:47,  1.28it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.28it/s]
overall:  20%|██        | 801/4000 [10:44<41:47,  1.28it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.28it/s]
overall:  20%|██        | 802/4000 [10:45<41:52,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  20%|██        | 803/4000 [10:45<41:49,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  20%|██        | 804/4000 [10:46<41:47,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  20%|██        | 805/4000 [10:47<41:54,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  20%|██        | 806/4000 [10:48<41:50,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  20%|██        | 807/4000 [10:48<41:46,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  20%|██        | 808/4000 [10:49<41:53,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  20%|██        | 809/4000 [10:50<41:46,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]train loss: 1.2958824634552002
train acc: 0.6639999747276306

overall:  20%|██        | 810/4000 [10:51<41:45,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  20%|██        | 811/4000 [10:52<41:43,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  20%|██        | 812/4000 [10:52<41:44,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  20%|██        | 813/4000 [10:53<41:44,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  20%|██        | 814/4000 [10:54<40:13,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  20%|██        | 815/4000 [10:55<52:20,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  20%|██        | 816/4000 [10:56<49:08,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  20%|██        | 817/4000 [10:57<47:01,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  20%|██        | 818/4000 [10:58<45:35,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:29,  1.14it/s]
overall:  20%|██        | 819/4000 [10:59<44:25,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]train loss: 1.3204410076141357
train acc: 0.6447368264198303

overall:  20%|██        | 820/4000 [10:59<43:37,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  21%|██        | 821/4000 [11:00<43:04,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  21%|██        | 822/4000 [11:01<42:35,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  21%|██        | 823/4000 [11:02<42:25,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  21%|██        | 824/4000 [11:02<42:05,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  21%|██        | 825/4000 [11:03<41:57,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  21%|██        | 826/4000 [11:04<41:53,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  21%|██        | 827/4000 [11:05<41:59,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  21%|██        | 828/4000 [11:06<41:53,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  21%|██        | 829/4000 [11:06<41:46,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]train loss: 1.2660835981369019
train acc: 0.6620000004768372

overall:  21%|██        | 830/4000 [11:07<41:48,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  21%|██        | 831/4000 [11:08<41:48,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  21%|██        | 832/4000 [11:09<41:44,  1.26it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.27it/s]
overall:  21%|██        | 833/4000 [11:10<41:43,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  21%|██        | 834/4000 [11:10<41:38,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  21%|██        | 835/4000 [11:11<41:39,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  21%|██        | 836/4000 [11:12<41:33,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  21%|██        | 837/4000 [11:13<41:31,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  21%|██        | 838/4000 [11:14<41:31,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  21%|██        | 839/4000 [11:14<41:37,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]train loss: 1.305337905883789
train acc: 0.6370000243186951

overall:  21%|██        | 840/4000 [11:15<41:36,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  21%|██        | 841/4000 [11:16<41:28,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  21%|██        | 842/4000 [11:17<41:35,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  21%|██        | 843/4000 [11:17<41:25,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  21%|██        | 844/4000 [11:18<41:27,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  21%|██        | 845/4000 [11:19<41:25,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  21%|██        | 846/4000 [11:20<41:19,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  21%|██        | 847/4000 [11:21<41:17,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  21%|██        | 848/4000 [11:21<41:18,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  21%|██        | 849/4000 [11:22<41:19,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]train loss: 1.2620164155960083
train acc: 0.6639999747276306

overall:  21%|██▏       | 850/4000 [11:23<41:15,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  21%|██▏       | 851/4000 [11:24<39:52,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  21%|██▏       | 852/4000 [11:25<52:18,  1.00it/s]

train:   3%|▎         | 1/37 [00:01<00:52,  1.45s/it]
overall:  21%|██▏       | 853/4000 [11:26<48:56,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:  21%|██▏       | 854/4000 [11:27<46:36,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  21%|██▏       | 855/4000 [11:28<45:05,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  21%|██▏       | 856/4000 [11:28<43:48,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  21%|██▏       | 857/4000 [11:29<42:59,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  21%|██▏       | 858/4000 [11:30<42:30,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  21%|██▏       | 859/4000 [11:31<42:04,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]train loss: 1.2700865268707275
train acc: 0.6771255135536194

overall:  22%|██▏       | 860/4000 [11:32<41:59,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  22%|██▏       | 861/4000 [11:32<41:39,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  22%|██▏       | 862/4000 [11:33<41:29,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  22%|██▏       | 863/4000 [11:34<41:25,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  22%|██▏       | 864/4000 [11:35<41:37,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  22%|██▏       | 865/4000 [11:36<41:31,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  22%|██▏       | 866/4000 [11:36<41:19,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  22%|██▏       | 867/4000 [11:37<41:09,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  22%|██▏       | 868/4000 [11:38<40:59,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  22%|██▏       | 869/4000 [11:39<40:57,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]train loss: 1.2629115581512451
train acc: 0.6600000262260437

overall:  22%|██▏       | 870/4000 [11:39<40:56,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  22%|██▏       | 871/4000 [11:40<40:57,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  22%|██▏       | 872/4000 [11:41<40:58,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  22%|██▏       | 873/4000 [11:42<40:57,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  22%|██▏       | 874/4000 [11:43<40:56,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  22%|██▏       | 875/4000 [11:43<40:52,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  22%|██▏       | 876/4000 [11:44<41:06,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  22%|██▏       | 877/4000 [11:45<41:02,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  22%|██▏       | 878/4000 [11:46<40:55,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  22%|██▏       | 879/4000 [11:47<40:55,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]train loss: 1.283066987991333
train acc: 0.6510000228881836

overall:  22%|██▏       | 880/4000 [11:47<40:52,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  22%|██▏       | 881/4000 [11:48<40:44,  1.28it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.28it/s]
overall:  22%|██▏       | 882/4000 [11:49<40:44,  1.28it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.28it/s]
overall:  22%|██▏       | 883/4000 [11:50<40:40,  1.28it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.28it/s]
overall:  22%|██▏       | 884/4000 [11:50<40:45,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  22%|██▏       | 885/4000 [11:51<40:44,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  22%|██▏       | 886/4000 [11:52<40:38,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:  22%|██▏       | 887/4000 [11:53<40:36,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:  22%|██▏       | 888/4000 [11:53<39:11,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  22%|██▏       | 889/4000 [11:55<50:46,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:50,  1.40s/it]train loss: 1.1843702793121338
train acc: 0.6943320035934448

overall:  22%|██▏       | 890/4000 [11:56<47:48,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.04s/it]
overall:  22%|██▏       | 891/4000 [11:57<45:44,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  22%|██▏       | 892/4000 [11:57<44:24,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  22%|██▏       | 893/4000 [11:58<43:30,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  22%|██▏       | 894/4000 [11:59<42:39,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  22%|██▏       | 895/4000 [12:00<41:59,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  22%|██▏       | 896/4000 [12:00<41:33,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  22%|██▏       | 897/4000 [12:01<41:21,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  22%|██▏       | 898/4000 [12:02<41:06,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  22%|██▏       | 899/4000 [12:03<40:56,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]train loss: 1.1771713495254517
train acc: 0.6890000104904175

overall:  22%|██▎       | 900/4000 [12:04<40:55,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  23%|██▎       | 901/4000 [12:04<40:50,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.27it/s]
overall:  23%|██▎       | 902/4000 [12:05<40:49,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  23%|██▎       | 903/4000 [12:06<40:38,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  23%|██▎       | 904/4000 [12:07<40:35,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  23%|██▎       | 905/4000 [12:08<40:38,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  23%|██▎       | 906/4000 [12:08<40:34,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  23%|██▎       | 907/4000 [12:09<40:29,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  23%|██▎       | 908/4000 [12:10<40:31,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  23%|██▎       | 909/4000 [12:11<40:30,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]train loss: 1.2729259729385376
train acc: 0.6399999856948853

overall:  23%|██▎       | 910/4000 [12:12<40:31,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  23%|██▎       | 911/4000 [12:12<40:26,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  23%|██▎       | 912/4000 [12:13<40:29,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  23%|██▎       | 913/4000 [12:14<40:26,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  23%|██▎       | 914/4000 [12:15<40:31,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  23%|██▎       | 915/4000 [12:15<40:26,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  23%|██▎       | 916/4000 [12:16<40:20,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  23%|██▎       | 917/4000 [12:17<40:29,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  23%|██▎       | 918/4000 [12:18<40:25,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  23%|██▎       | 919/4000 [12:19<40:17,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]train loss: 1.2526134252548218
train acc: 0.6610000133514404

overall:  23%|██▎       | 920/4000 [12:19<40:14,  1.28it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.28it/s]
overall:  23%|██▎       | 921/4000 [12:20<40:18,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  23%|██▎       | 922/4000 [12:21<40:21,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  23%|██▎       | 923/4000 [12:22<40:13,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  23%|██▎       | 924/4000 [12:22<40:12,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:  23%|██▎       | 925/4000 [12:23<38:49,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  23%|██▎       | 926/4000 [12:25<49:32,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  23%|██▎       | 927/4000 [12:25<46:51,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  23%|██▎       | 928/4000 [12:26<44:55,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  23%|██▎       | 929/4000 [12:27<43:54,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]train loss: 1.21896231174469
train acc: 0.6751012206077576

overall:  23%|██▎       | 930/4000 [12:28<42:52,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  23%|██▎       | 931/4000 [12:29<42:03,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  23%|██▎       | 932/4000 [12:29<41:33,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  23%|██▎       | 933/4000 [12:30<41:05,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  23%|██▎       | 934/4000 [12:31<40:53,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  23%|██▎       | 935/4000 [12:32<40:36,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  23%|██▎       | 936/4000 [12:33<40:29,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  23%|██▎       | 937/4000 [12:33<40:24,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  23%|██▎       | 938/4000 [12:34<40:32,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  23%|██▎       | 939/4000 [12:35<40:25,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]train loss: 1.2287135124206543
train acc: 0.6679999828338623

overall:  24%|██▎       | 940/4000 [12:36<40:24,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  24%|██▎       | 941/4000 [12:37<40:22,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  24%|██▎       | 942/4000 [12:37<40:16,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  24%|██▎       | 943/4000 [12:38<40:11,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  24%|██▎       | 944/4000 [12:39<40:08,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  24%|██▎       | 945/4000 [12:40<40:09,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  24%|██▎       | 946/4000 [12:40<40:02,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  24%|██▎       | 947/4000 [12:41<39:57,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  24%|██▎       | 948/4000 [12:42<39:53,  1.28it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.28it/s]
overall:  24%|██▎       | 949/4000 [12:43<39:58,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]train loss: 1.2117793560028076
train acc: 0.6909999847412109

overall:  24%|██▍       | 950/4000 [12:44<40:14,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  24%|██▍       | 951/4000 [12:44<40:07,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  24%|██▍       | 952/4000 [12:45<40:10,  1.26it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.26it/s]
overall:  24%|██▍       | 953/4000 [12:46<40:09,  1.26it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.26it/s]
overall:  24%|██▍       | 954/4000 [12:47<40:08,  1.26it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.26it/s]
overall:  24%|██▍       | 955/4000 [12:48<39:57,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  24%|██▍       | 956/4000 [12:48<39:58,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  24%|██▍       | 957/4000 [12:49<39:58,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  24%|██▍       | 958/4000 [12:50<39:53,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  24%|██▍       | 959/4000 [12:51<39:59,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]train loss: 1.208984613418579
train acc: 0.6570000052452087

overall:  24%|██▍       | 960/4000 [12:51<39:56,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  24%|██▍       | 961/4000 [12:52<40:00,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  24%|██▍       | 962/4000 [12:53<38:29,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  24%|██▍       | 963/4000 [12:55<50:07,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.42s/it]
overall:  24%|██▍       | 964/4000 [12:55<46:56,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  24%|██▍       | 965/4000 [12:56<44:54,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  24%|██▍       | 966/4000 [12:57<43:31,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  24%|██▍       | 967/4000 [12:58<42:25,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  24%|██▍       | 968/4000 [12:58<41:38,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  24%|██▍       | 969/4000 [12:59<41:05,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]train loss: 1.1992665529251099
train acc: 0.6831983923912048

overall:  24%|██▍       | 970/4000 [13:00<40:50,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  24%|██▍       | 971/4000 [13:01<40:26,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  24%|██▍       | 972/4000 [13:02<40:08,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  24%|██▍       | 973/4000 [13:02<40:00,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  24%|██▍       | 974/4000 [13:03<39:57,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  24%|██▍       | 975/4000 [13:04<40:01,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  24%|██▍       | 976/4000 [13:05<39:48,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  24%|██▍       | 977/4000 [13:06<39:39,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  24%|██▍       | 978/4000 [13:06<39:34,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  24%|██▍       | 979/4000 [13:07<39:31,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]train loss: 1.1691780090332031
train acc: 0.6859999895095825

overall:  24%|██▍       | 980/4000 [13:08<39:36,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  25%|██▍       | 981/4000 [13:09<39:31,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  25%|██▍       | 982/4000 [13:09<39:41,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  25%|██▍       | 983/4000 [13:10<39:32,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  25%|██▍       | 984/4000 [13:11<39:29,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  25%|██▍       | 985/4000 [13:12<39:27,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  25%|██▍       | 986/4000 [13:13<39:31,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  25%|██▍       | 987/4000 [13:13<39:41,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  25%|██▍       | 988/4000 [13:14<39:33,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  25%|██▍       | 989/4000 [13:15<39:26,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]train loss: 1.2031831741333008
train acc: 0.6600000262260437

overall:  25%|██▍       | 990/4000 [13:16<39:27,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  25%|██▍       | 991/4000 [13:17<39:26,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  25%|██▍       | 992/4000 [13:17<39:21,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  25%|██▍       | 993/4000 [13:18<39:21,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  25%|██▍       | 994/4000 [13:19<39:26,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  25%|██▍       | 995/4000 [13:20<39:24,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  25%|██▍       | 996/4000 [13:20<39:19,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  25%|██▍       | 997/4000 [13:21<39:26,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  25%|██▍       | 998/4000 [13:22<39:21,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  25%|██▍       | 999/4000 [13:23<37:57,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]train loss: 1.2191879749298096
train acc: 0.6781376600265503

overall:  25%|██▌       | 1000/4000 [13:24<49:37,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  25%|██▌       | 1001/4000 [13:25<46:28,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  25%|██▌       | 1002/4000 [13:26<44:32,  1.12it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  25%|██▌       | 1003/4000 [13:27<43:19,  1.15it/s]

train:  11%|█         | 4/37 [00:03<00:29,  1.13it/s]
overall:  25%|██▌       | 1004/4000 [13:27<42:03,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  25%|██▌       | 1005/4000 [13:28<41:08,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  25%|██▌       | 1006/4000 [13:29<40:29,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  25%|██▌       | 1007/4000 [13:30<40:04,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  25%|██▌       | 1008/4000 [13:31<39:56,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  25%|██▌       | 1009/4000 [13:31<39:43,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]train loss: 1.1296987533569336
train acc: 0.6990000009536743

overall:  25%|██▌       | 1010/4000 [13:32<39:37,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  25%|██▌       | 1011/4000 [13:33<39:24,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  25%|██▌       | 1012/4000 [13:34<39:26,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  25%|██▌       | 1013/4000 [13:35<39:21,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  25%|██▌       | 1014/4000 [13:35<39:24,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  25%|██▌       | 1015/4000 [13:36<39:15,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  25%|██▌       | 1016/4000 [13:37<39:15,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  25%|██▌       | 1017/4000 [13:38<39:12,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  25%|██▌       | 1018/4000 [13:38<39:11,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  25%|██▌       | 1019/4000 [13:39<39:10,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]train loss: 1.2381954193115234
train acc: 0.6359999775886536

overall:  26%|██▌       | 1020/4000 [13:40<39:17,  1.26it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.26it/s]
overall:  26%|██▌       | 1021/4000 [13:41<39:09,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  26%|██▌       | 1022/4000 [13:42<39:03,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  26%|██▌       | 1023/4000 [13:42<39:00,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  26%|██▌       | 1024/4000 [13:43<39:02,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  26%|██▌       | 1025/4000 [13:44<39:01,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  26%|██▌       | 1026/4000 [13:45<39:04,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  26%|██▌       | 1027/4000 [13:46<38:59,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  26%|██▌       | 1028/4000 [13:46<38:54,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  26%|██▌       | 1029/4000 [13:47<38:47,  1.28it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.28it/s]train loss: 1.1395117044448853
train acc: 0.7129999995231628

overall:  26%|██▌       | 1030/4000 [13:48<38:46,  1.28it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.28it/s]
overall:  26%|██▌       | 1031/4000 [13:49<38:43,  1.28it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.28it/s]
overall:  26%|██▌       | 1032/4000 [13:50<38:52,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  26%|██▌       | 1033/4000 [13:50<38:47,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  26%|██▌       | 1034/4000 [13:51<38:44,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:  26%|██▌       | 1035/4000 [13:52<38:39,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:  26%|██▌       | 1036/4000 [13:53<37:19,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  26%|██▌       | 1037/4000 [13:54<47:46,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  26%|██▌       | 1038/4000 [13:55<45:03,  1.10it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.03s/it]
overall:  26%|██▌       | 1039/4000 [13:56<43:17,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]train loss: 1.1677449941635132
train acc: 0.6902834177017212

overall:  26%|██▌       | 1040/4000 [13:56<41:57,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  26%|██▌       | 1041/4000 [13:57<40:58,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  26%|██▌       | 1042/4000 [13:58<40:21,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  26%|██▌       | 1043/4000 [13:59<39:46,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.24it/s]
overall:  26%|██▌       | 1044/4000 [14:00<39:35,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  26%|██▌       | 1045/4000 [14:00<39:21,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  26%|██▌       | 1046/4000 [14:01<39:12,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  26%|██▌       | 1047/4000 [14:02<38:59,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  26%|██▌       | 1048/4000 [14:03<38:48,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  26%|██▌       | 1049/4000 [14:03<38:54,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.26it/s]train loss: 1.1254551410675049
train acc: 0.6930000185966492

overall:  26%|██▋       | 1050/4000 [14:04<38:53,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  26%|██▋       | 1051/4000 [14:05<38:52,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  26%|██▋       | 1052/4000 [14:06<38:45,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  26%|██▋       | 1053/4000 [14:07<38:37,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  26%|██▋       | 1054/4000 [14:07<38:35,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  26%|██▋       | 1055/4000 [14:08<38:33,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  26%|██▋       | 1056/4000 [14:09<38:32,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  26%|██▋       | 1057/4000 [14:10<38:35,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  26%|██▋       | 1058/4000 [14:11<38:34,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  26%|██▋       | 1059/4000 [14:11<38:33,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]train loss: 1.1708787679672241
train acc: 0.671999990940094

overall:  26%|██▋       | 1060/4000 [14:12<38:28,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  27%|██▋       | 1061/4000 [14:13<38:35,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  27%|██▋       | 1062/4000 [14:14<38:32,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  27%|██▋       | 1063/4000 [14:14<38:39,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  27%|██▋       | 1064/4000 [14:15<38:34,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  27%|██▋       | 1065/4000 [14:16<38:33,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  27%|██▋       | 1066/4000 [14:17<38:33,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  27%|██▋       | 1067/4000 [14:18<38:24,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.27it/s]
overall:  27%|██▋       | 1068/4000 [14:18<38:31,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  27%|██▋       | 1069/4000 [14:19<38:32,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]train loss: 1.1522175073623657
train acc: 0.6800000071525574

overall:  27%|██▋       | 1070/4000 [14:20<38:31,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  27%|██▋       | 1071/4000 [14:21<38:25,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  27%|██▋       | 1072/4000 [14:22<38:19,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  27%|██▋       | 1073/4000 [14:22<36:57,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  27%|██▋       | 1074/4000 [14:24<48:24,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  27%|██▋       | 1075/4000 [14:25<45:18,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  27%|██▋       | 1076/4000 [14:25<43:11,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  27%|██▋       | 1077/4000 [14:26<41:48,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  27%|██▋       | 1078/4000 [14:27<40:52,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  27%|██▋       | 1079/4000 [14:28<40:05,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]train loss: 1.149838924407959
train acc: 0.6963562965393066

overall:  27%|██▋       | 1080/4000 [14:29<39:44,  1.22it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.22it/s]
overall:  27%|██▋       | 1081/4000 [14:29<39:13,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  27%|██▋       | 1082/4000 [14:30<38:55,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  27%|██▋       | 1083/4000 [14:31<38:47,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  27%|██▋       | 1084/4000 [14:32<38:36,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  27%|██▋       | 1085/4000 [14:32<38:35,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  27%|██▋       | 1086/4000 [14:33<38:37,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  27%|██▋       | 1087/4000 [14:34<38:30,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  27%|██▋       | 1088/4000 [14:35<38:26,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  27%|██▋       | 1089/4000 [14:36<38:17,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]train loss: 1.154023289680481
train acc: 0.6800000071525574

overall:  27%|██▋       | 1090/4000 [14:36<38:21,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  27%|██▋       | 1091/4000 [14:37<38:10,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  27%|██▋       | 1092/4000 [14:38<38:18,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  27%|██▋       | 1093/4000 [14:39<38:12,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  27%|██▋       | 1094/4000 [14:40<38:05,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  27%|██▋       | 1095/4000 [14:40<38:02,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  27%|██▋       | 1096/4000 [14:41<38:07,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  27%|██▋       | 1097/4000 [14:42<38:01,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  27%|██▋       | 1098/4000 [14:43<38:09,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  27%|██▋       | 1099/4000 [14:44<38:02,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]train loss: 1.1210987567901611
train acc: 0.699999988079071

overall:  28%|██▊       | 1100/4000 [14:44<38:01,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  28%|██▊       | 1101/4000 [14:45<37:54,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  28%|██▊       | 1102/4000 [14:46<37:59,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  28%|██▊       | 1103/4000 [14:47<37:57,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  28%|██▊       | 1104/4000 [14:47<38:03,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  28%|██▊       | 1105/4000 [14:48<38:00,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  28%|██▊       | 1106/4000 [14:49<37:50,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  28%|██▊       | 1107/4000 [14:50<37:47,  1.28it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.28it/s]
overall:  28%|██▊       | 1108/4000 [14:51<37:43,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:  28%|██▊       | 1109/4000 [14:51<37:44,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.28it/s]train loss: 1.160747766494751
train acc: 0.6801619529724121

overall:  28%|██▊       | 1110/4000 [14:52<36:34,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  28%|██▊       | 1111/4000 [14:54<46:41,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  28%|██▊       | 1112/4000 [14:54<44:02,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.03s/it]
overall:  28%|██▊       | 1113/4000 [14:55<42:06,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  28%|██▊       | 1114/4000 [14:56<40:53,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  28%|██▊       | 1115/4000 [14:57<40:14,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  28%|██▊       | 1116/4000 [14:57<39:32,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  28%|██▊       | 1117/4000 [14:58<39:03,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  28%|██▊       | 1118/4000 [14:59<38:34,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  28%|██▊       | 1119/4000 [15:00<38:29,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]train loss: 1.136109471321106
train acc: 0.6930000185966492

overall:  28%|██▊       | 1120/4000 [15:01<38:14,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  28%|██▊       | 1121/4000 [15:01<38:01,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  28%|██▊       | 1122/4000 [15:02<37:58,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  28%|██▊       | 1123/4000 [15:03<37:55,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.26it/s]
overall:  28%|██▊       | 1124/4000 [15:04<37:59,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  28%|██▊       | 1125/4000 [15:05<37:47,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  28%|██▊       | 1126/4000 [15:05<37:53,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  28%|██▊       | 1127/4000 [15:06<37:44,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  28%|██▊       | 1128/4000 [15:07<37:39,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  28%|██▊       | 1129/4000 [15:08<37:44,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]train loss: 1.1190741062164307
train acc: 0.7009999752044678

overall:  28%|██▊       | 1130/4000 [15:09<37:40,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  28%|██▊       | 1131/4000 [15:09<37:45,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  28%|██▊       | 1132/4000 [15:10<37:43,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  28%|██▊       | 1133/4000 [15:11<37:44,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  28%|██▊       | 1134/4000 [15:12<37:37,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  28%|██▊       | 1135/4000 [15:12<37:42,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  28%|██▊       | 1136/4000 [15:13<37:34,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  28%|██▊       | 1137/4000 [15:14<37:32,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  28%|██▊       | 1138/4000 [15:15<37:38,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  28%|██▊       | 1139/4000 [15:16<37:30,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]train loss: 1.0897413492202759
train acc: 0.7089999914169312

overall:  28%|██▊       | 1140/4000 [15:16<37:29,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  29%|██▊       | 1141/4000 [15:17<37:28,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  29%|██▊       | 1142/4000 [15:18<37:27,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  29%|██▊       | 1143/4000 [15:19<37:33,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  29%|██▊       | 1144/4000 [15:20<37:29,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  29%|██▊       | 1145/4000 [15:20<37:23,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  29%|██▊       | 1146/4000 [15:21<37:20,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  29%|██▊       | 1147/4000 [15:22<36:03,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  29%|██▊       | 1148/4000 [15:23<48:42,  1.02s/it]

train:   3%|▎         | 1/37 [00:01<00:56,  1.56s/it]
overall:  29%|██▊       | 1149/4000 [15:24<45:21,  1.05it/s]

train:   5%|▌         | 2/37 [00:02<00:38,  1.11s/it]train loss: 1.1648733615875244
train acc: 0.6862348318099976

overall:  29%|██▉       | 1150/4000 [15:25<42:52,  1.11it/s]

train:   8%|▊         | 3/37 [00:03<00:32,  1.04it/s]
overall:  29%|██▉       | 1151/4000 [15:26<41:18,  1.15it/s]

train:  11%|█         | 4/37 [00:03<00:29,  1.12it/s]
overall:  29%|██▉       | 1152/4000 [15:27<40:04,  1.18it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.17it/s]
overall:  29%|██▉       | 1153/4000 [15:27<39:12,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.20it/s]
overall:  29%|██▉       | 1154/4000 [15:28<38:34,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  29%|██▉       | 1155/4000 [15:29<38:19,  1.24it/s]

train:  22%|██▏       | 8/37 [00:07<00:23,  1.24it/s]
overall:  29%|██▉       | 1156/4000 [15:30<38:01,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  29%|██▉       | 1157/4000 [15:31<37:47,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  29%|██▉       | 1158/4000 [15:31<37:34,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  29%|██▉       | 1159/4000 [15:32<37:25,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]train loss: 1.118196964263916
train acc: 0.6869999766349792

overall:  29%|██▉       | 1160/4000 [15:33<37:41,  1.26it/s]

train:  35%|███▌      | 13/37 [00:11<00:19,  1.26it/s]
overall:  29%|██▉       | 1161/4000 [15:34<37:28,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  29%|██▉       | 1162/4000 [15:34<37:18,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  29%|██▉       | 1163/4000 [15:35<37:12,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  29%|██▉       | 1164/4000 [15:36<37:09,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  29%|██▉       | 1165/4000 [15:37<37:12,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  29%|██▉       | 1166/4000 [15:38<37:06,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  29%|██▉       | 1167/4000 [15:38<37:13,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  29%|██▉       | 1168/4000 [15:39<37:08,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  29%|██▉       | 1169/4000 [15:40<37:04,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]train loss: 1.0895822048187256
train acc: 0.7099999785423279

overall:  29%|██▉       | 1170/4000 [15:41<37:05,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  29%|██▉       | 1171/4000 [15:42<37:06,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  29%|██▉       | 1172/4000 [15:42<37:16,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  29%|██▉       | 1173/4000 [15:43<37:12,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  29%|██▉       | 1174/4000 [15:44<37:09,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:22<00:07,  1.27it/s]
overall:  29%|██▉       | 1175/4000 [15:45<37:04,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  29%|██▉       | 1176/4000 [15:45<36:57,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  29%|██▉       | 1177/4000 [15:46<36:57,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  29%|██▉       | 1178/4000 [15:47<36:59,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  29%|██▉       | 1179/4000 [15:48<37:11,  1.26it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.26it/s]train loss: 1.1177091598510742
train acc: 0.6850000023841858

overall:  30%|██▉       | 1180/4000 [15:49<37:07,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  30%|██▉       | 1181/4000 [15:49<37:04,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  30%|██▉       | 1182/4000 [15:50<37:03,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  30%|██▉       | 1183/4000 [15:51<36:55,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  30%|██▉       | 1184/4000 [15:52<35:39,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  30%|██▉       | 1185/4000 [15:53<45:47,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.38s/it]
overall:  30%|██▉       | 1186/4000 [15:54<43:06,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  30%|██▉       | 1187/4000 [15:55<41:14,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  30%|██▉       | 1188/4000 [15:56<39:54,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  30%|██▉       | 1189/4000 [15:56<39:13,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]train loss: 1.1173142194747925
train acc: 0.6973684430122375

overall:  30%|██▉       | 1190/4000 [15:57<38:36,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  30%|██▉       | 1191/4000 [15:58<37:59,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  30%|██▉       | 1192/4000 [15:59<37:37,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  30%|██▉       | 1193/4000 [16:00<37:26,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  30%|██▉       | 1194/4000 [16:00<37:14,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  30%|██▉       | 1195/4000 [16:01<37:11,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  30%|██▉       | 1196/4000 [16:02<37:02,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  30%|██▉       | 1197/4000 [16:03<37:15,  1.25it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.25it/s]
overall:  30%|██▉       | 1198/4000 [16:03<37:05,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  30%|██▉       | 1199/4000 [16:04<36:54,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]train loss: 1.0863659381866455
train acc: 0.7149999737739563

overall:  30%|███       | 1200/4000 [16:05<36:53,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  30%|███       | 1201/4000 [16:06<36:47,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  30%|███       | 1202/4000 [16:07<36:44,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  30%|███       | 1203/4000 [16:07<36:38,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  30%|███       | 1204/4000 [16:08<36:39,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  30%|███       | 1205/4000 [16:09<36:35,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  30%|███       | 1206/4000 [16:10<36:37,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  30%|███       | 1207/4000 [16:11<36:40,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  30%|███       | 1208/4000 [16:11<36:33,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  30%|███       | 1209/4000 [16:12<36:47,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]train loss: 1.071415662765503
train acc: 0.7039999961853027

overall:  30%|███       | 1210/4000 [16:13<36:42,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  30%|███       | 1211/4000 [16:14<36:38,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  30%|███       | 1212/4000 [16:14<36:33,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  30%|███       | 1213/4000 [16:15<36:28,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  30%|███       | 1214/4000 [16:16<36:24,  1.28it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.28it/s]
overall:  30%|███       | 1215/4000 [16:17<36:22,  1.28it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.28it/s]
overall:  30%|███       | 1216/4000 [16:18<36:23,  1.28it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.28it/s]
overall:  30%|███       | 1217/4000 [16:18<36:24,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  30%|███       | 1218/4000 [16:19<36:27,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  30%|███       | 1219/4000 [16:20<36:29,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]train loss: 1.1100672483444214
train acc: 0.6919999718666077

overall:  30%|███       | 1220/4000 [16:21<36:24,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  31%|███       | 1221/4000 [16:21<35:05,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  31%|███       | 1222/4000 [16:23<45:57,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:52,  1.44s/it]
overall:  31%|███       | 1223/4000 [16:24<43:00,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  31%|███       | 1224/4000 [16:25<40:59,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  31%|███       | 1225/4000 [16:25<39:32,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  31%|███       | 1226/4000 [16:26<38:37,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  31%|███       | 1227/4000 [16:27<37:55,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  31%|███       | 1228/4000 [16:28<37:33,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  31%|███       | 1229/4000 [16:28<37:08,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]train loss: 1.0548579692840576
train acc: 0.7257084846496582

overall:  31%|███       | 1230/4000 [16:29<36:49,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  31%|███       | 1231/4000 [16:30<36:38,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  31%|███       | 1232/4000 [16:31<36:30,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  31%|███       | 1233/4000 [16:32<36:23,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  31%|███       | 1234/4000 [16:32<36:26,  1.27it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.27it/s]
overall:  31%|███       | 1235/4000 [16:33<36:23,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  31%|███       | 1236/4000 [16:34<36:16,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  31%|███       | 1237/4000 [16:35<36:13,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  31%|███       | 1238/4000 [16:36<36:14,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  31%|███       | 1239/4000 [16:36<36:09,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]train loss: 1.094351053237915
train acc: 0.7059999704360962

overall:  31%|███       | 1240/4000 [16:37<36:16,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  31%|███       | 1241/4000 [16:38<36:16,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  31%|███       | 1242/4000 [16:39<36:08,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  31%|███       | 1243/4000 [16:40<36:04,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  31%|███       | 1244/4000 [16:40<36:03,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  31%|███       | 1245/4000 [16:41<36:03,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  31%|███       | 1246/4000 [16:42<36:06,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  31%|███       | 1247/4000 [16:43<36:10,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  31%|███       | 1248/4000 [16:43<36:05,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  31%|███       | 1249/4000 [16:44<36:03,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]train loss: 1.096986174583435
train acc: 0.7089999914169312

overall:  31%|███▏      | 1250/4000 [16:45<36:03,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  31%|███▏      | 1251/4000 [16:46<35:57,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  31%|███▏      | 1252/4000 [16:47<35:59,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  31%|███▏      | 1253/4000 [16:47<36:03,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  31%|███▏      | 1254/4000 [16:48<36:02,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  31%|███▏      | 1255/4000 [16:49<36:02,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  31%|███▏      | 1256/4000 [16:50<35:58,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  31%|███▏      | 1257/4000 [16:51<35:59,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  31%|███▏      | 1258/4000 [16:51<34:47,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  31%|███▏      | 1259/4000 [16:53<44:25,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]train loss: 1.0891649723052979
train acc: 0.692307710647583

overall:  32%|███▏      | 1260/4000 [16:53<41:51,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.03s/it]
overall:  32%|███▏      | 1261/4000 [16:54<39:59,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  32%|███▏      | 1262/4000 [16:55<38:44,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]
overall:  32%|███▏      | 1263/4000 [16:56<37:54,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  32%|███▏      | 1264/4000 [16:57<37:21,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  32%|███▏      | 1265/4000 [16:57<36:57,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  32%|███▏      | 1266/4000 [16:58<36:47,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  32%|███▏      | 1267/4000 [16:59<36:31,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  32%|███▏      | 1268/4000 [17:00<36:19,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  32%|███▏      | 1269/4000 [17:01<36:06,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]train loss: 1.111004114151001
train acc: 0.6930000185966492

overall:  32%|███▏      | 1270/4000 [17:01<36:01,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  32%|███▏      | 1271/4000 [17:02<36:14,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  32%|███▏      | 1272/4000 [17:03<36:05,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  32%|███▏      | 1273/4000 [17:04<35:59,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  32%|███▏      | 1274/4000 [17:05<35:54,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  32%|███▏      | 1275/4000 [17:05<35:46,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  32%|███▏      | 1276/4000 [17:06<35:42,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  32%|███▏      | 1277/4000 [17:07<35:41,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  32%|███▏      | 1278/4000 [17:08<35:47,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  32%|███▏      | 1279/4000 [17:08<35:47,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]train loss: 1.079041838645935
train acc: 0.7139999866485596

overall:  32%|███▏      | 1280/4000 [17:09<35:40,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  32%|███▏      | 1281/4000 [17:10<35:36,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  32%|███▏      | 1282/4000 [17:11<35:34,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  32%|███▏      | 1283/4000 [17:12<35:43,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  32%|███▏      | 1284/4000 [17:12<35:44,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  32%|███▏      | 1285/4000 [17:13<35:38,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  32%|███▏      | 1286/4000 [17:14<35:30,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  32%|███▏      | 1287/4000 [17:15<35:28,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  32%|███▏      | 1288/4000 [17:16<35:22,  1.28it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.28it/s]
overall:  32%|███▏      | 1289/4000 [17:16<35:22,  1.28it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.28it/s]train loss: 1.02561616897583
train acc: 0.7170000076293945

overall:  32%|███▏      | 1290/4000 [17:17<35:30,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  32%|███▏      | 1291/4000 [17:18<35:26,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  32%|███▏      | 1292/4000 [17:19<35:20,  1.28it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.28it/s]
overall:  32%|███▏      | 1293/4000 [17:19<35:16,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:  32%|███▏      | 1294/4000 [17:20<35:15,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:  32%|███▏      | 1295/4000 [17:21<34:00,  1.33it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.33it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  32%|███▏      | 1296/4000 [17:22<43:21,  1.04it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.36s/it]
overall:  32%|███▏      | 1297/4000 [17:23<41:06,  1.10it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  32%|███▏      | 1298/4000 [17:24<39:24,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  32%|███▏      | 1299/4000 [17:25<38:10,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]train loss: 1.0798466205596924
train acc: 0.6862348318099976

overall:  32%|███▎      | 1300/4000 [17:26<37:23,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  33%|███▎      | 1301/4000 [17:26<36:53,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  33%|███▎      | 1302/4000 [17:27<36:23,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  33%|███▎      | 1303/4000 [17:28<36:11,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  33%|███▎      | 1304/4000 [17:29<35:53,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  33%|███▎      | 1305/4000 [17:29<35:38,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  33%|███▎      | 1306/4000 [17:30<35:36,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  33%|███▎      | 1307/4000 [17:31<35:29,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  33%|███▎      | 1308/4000 [17:32<35:33,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  33%|███▎      | 1309/4000 [17:33<35:26,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]train loss: 1.0306035280227661
train acc: 0.7250000238418579

overall:  33%|███▎      | 1310/4000 [17:33<35:30,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  33%|███▎      | 1311/4000 [17:34<35:21,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  33%|███▎      | 1312/4000 [17:35<35:15,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  33%|███▎      | 1313/4000 [17:36<35:15,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  33%|███▎      | 1314/4000 [17:37<35:15,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  33%|███▎      | 1315/4000 [17:37<35:18,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  33%|███▎      | 1316/4000 [17:38<35:16,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  33%|███▎      | 1317/4000 [17:39<35:10,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  33%|███▎      | 1318/4000 [17:40<35:09,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  33%|███▎      | 1319/4000 [17:40<35:09,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]train loss: 1.0540274381637573
train acc: 0.7260000109672546

overall:  33%|███▎      | 1320/4000 [17:41<35:10,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  33%|███▎      | 1321/4000 [17:42<35:12,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  33%|███▎      | 1322/4000 [17:43<35:08,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  33%|███▎      | 1323/4000 [17:44<35:07,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  33%|███▎      | 1324/4000 [17:44<35:04,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  33%|███▎      | 1325/4000 [17:45<35:02,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  33%|███▎      | 1326/4000 [17:46<35:00,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.27it/s]
overall:  33%|███▎      | 1327/4000 [17:47<35:07,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  33%|███▎      | 1328/4000 [17:48<35:00,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  33%|███▎      | 1329/4000 [17:48<34:55,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]train loss: 1.0382145643234253
train acc: 0.7149999737739563

overall:  33%|███▎      | 1330/4000 [17:49<34:57,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  33%|███▎      | 1331/4000 [17:50<34:59,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  33%|███▎      | 1332/4000 [17:51<33:47,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  33%|███▎      | 1333/4000 [17:52<43:33,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:50,  1.40s/it]
overall:  33%|███▎      | 1334/4000 [17:53<40:59,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.04s/it]
overall:  33%|███▎      | 1335/4000 [17:54<39:05,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  33%|███▎      | 1336/4000 [17:54<37:47,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  33%|███▎      | 1337/4000 [17:55<36:58,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  33%|███▎      | 1338/4000 [17:56<36:29,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  33%|███▎      | 1339/4000 [17:57<35:59,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]train loss: 1.0875998735427856
train acc: 0.6963562965393066

overall:  34%|███▎      | 1340/4000 [17:58<35:36,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  34%|███▎      | 1341/4000 [17:58<35:26,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  34%|███▎      | 1342/4000 [17:59<35:15,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  34%|███▎      | 1343/4000 [18:00<35:04,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  34%|███▎      | 1344/4000 [18:01<35:01,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  34%|███▎      | 1345/4000 [18:02<35:09,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  34%|███▎      | 1346/4000 [18:02<35:01,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  34%|███▎      | 1347/4000 [18:03<34:58,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  34%|███▎      | 1348/4000 [18:04<34:51,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  34%|███▎      | 1349/4000 [18:05<34:46,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]train loss: 1.0220822095870972
train acc: 0.7350000143051147

overall:  34%|███▍      | 1350/4000 [18:06<34:45,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  34%|███▍      | 1351/4000 [18:06<34:45,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  34%|███▍      | 1352/4000 [18:07<34:43,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  34%|███▍      | 1353/4000 [18:08<34:47,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  34%|███▍      | 1354/4000 [18:09<34:44,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  34%|███▍      | 1355/4000 [18:09<34:38,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  34%|███▍      | 1356/4000 [18:10<34:44,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  34%|███▍      | 1357/4000 [18:11<34:52,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  34%|███▍      | 1358/4000 [18:12<34:49,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  34%|███▍      | 1359/4000 [18:13<34:41,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]train loss: 1.0742422342300415
train acc: 0.7009999752044678

overall:  34%|███▍      | 1360/4000 [18:13<34:40,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  34%|███▍      | 1361/4000 [18:14<34:36,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  34%|███▍      | 1362/4000 [18:15<34:32,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  34%|███▍      | 1363/4000 [18:16<34:30,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  34%|███▍      | 1364/4000 [18:17<34:30,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  34%|███▍      | 1365/4000 [18:17<34:29,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  34%|███▍      | 1366/4000 [18:18<34:26,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  34%|███▍      | 1367/4000 [18:19<34:23,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:  34%|███▍      | 1368/4000 [18:20<34:23,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:  34%|███▍      | 1369/4000 [18:20<33:12,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]train loss: 1.0230305194854736
train acc: 0.7236841917037964

overall:  34%|███▍      | 1370/4000 [18:22<42:30,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  34%|███▍      | 1371/4000 [18:23<40:06,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.03s/it]
overall:  34%|███▍      | 1372/4000 [18:23<38:20,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  34%|███▍      | 1373/4000 [18:24<37:15,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  34%|███▍      | 1374/4000 [18:25<36:23,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  34%|███▍      | 1375/4000 [18:26<35:49,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  34%|███▍      | 1376/4000 [18:27<35:24,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  34%|███▍      | 1377/4000 [18:27<35:06,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  34%|███▍      | 1378/4000 [18:28<34:55,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  34%|███▍      | 1379/4000 [18:29<34:41,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]train loss: 1.0280075073242188
train acc: 0.7099999785423279

overall:  34%|███▍      | 1380/4000 [18:30<34:42,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  35%|███▍      | 1381/4000 [18:30<34:31,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  35%|███▍      | 1382/4000 [18:31<34:36,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  35%|███▍      | 1383/4000 [18:32<34:26,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  35%|███▍      | 1384/4000 [18:33<34:17,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  35%|███▍      | 1385/4000 [18:34<34:24,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  35%|███▍      | 1386/4000 [18:34<34:17,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  35%|███▍      | 1387/4000 [18:35<34:12,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  35%|███▍      | 1388/4000 [18:36<34:07,  1.28it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.28it/s]
overall:  35%|███▍      | 1389/4000 [18:37<34:10,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]train loss: 1.0518896579742432
train acc: 0.7020000219345093

overall:  35%|███▍      | 1390/4000 [18:38<34:17,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  35%|███▍      | 1391/4000 [18:38<34:17,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  35%|███▍      | 1392/4000 [18:39<34:15,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  35%|███▍      | 1393/4000 [18:40<34:13,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  35%|███▍      | 1394/4000 [18:41<34:28,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  35%|███▍      | 1395/4000 [18:42<34:21,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  35%|███▍      | 1396/4000 [18:42<34:16,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  35%|███▍      | 1397/4000 [18:43<34:15,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  35%|███▍      | 1398/4000 [18:44<34:08,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  35%|███▍      | 1399/4000 [18:45<34:05,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]train loss: 1.043064832687378
train acc: 0.7239999771118164

overall:  35%|███▌      | 1400/4000 [18:45<34:11,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  35%|███▌      | 1401/4000 [18:46<34:09,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  35%|███▌      | 1402/4000 [18:47<34:09,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  35%|███▌      | 1403/4000 [18:48<34:02,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  35%|███▌      | 1404/4000 [18:49<34:03,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  35%|███▌      | 1405/4000 [18:49<34:05,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  35%|███▌      | 1406/4000 [18:50<32:55,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  35%|███▌      | 1407/4000 [18:52<42:01,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  35%|███▌      | 1408/4000 [18:52<39:44,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  35%|███▌      | 1409/4000 [18:53<38:07,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]train loss: 1.0145082473754883
train acc: 0.7297570705413818

overall:  35%|███▌      | 1410/4000 [18:54<36:52,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  35%|███▌      | 1411/4000 [18:55<36:01,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  35%|███▌      | 1412/4000 [18:56<35:21,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  35%|███▌      | 1413/4000 [18:56<34:56,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  35%|███▌      | 1414/4000 [18:57<34:40,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  35%|███▌      | 1415/4000 [18:58<34:33,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  35%|███▌      | 1416/4000 [18:59<34:19,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  35%|███▌      | 1417/4000 [18:59<34:07,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  35%|███▌      | 1418/4000 [19:00<34:06,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  35%|███▌      | 1419/4000 [19:01<33:59,  1.27it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.27it/s]train loss: 1.004067301750183
train acc: 0.7289999723434448

overall:  36%|███▌      | 1420/4000 [19:02<34:04,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  36%|███▌      | 1421/4000 [19:03<34:09,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  36%|███▌      | 1422/4000 [19:03<34:02,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  36%|███▌      | 1423/4000 [19:04<34:00,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  36%|███▌      | 1424/4000 [19:05<33:52,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  36%|███▌      | 1425/4000 [19:06<33:47,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  36%|███▌      | 1426/4000 [19:07<33:45,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  36%|███▌      | 1427/4000 [19:07<33:44,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  36%|███▌      | 1428/4000 [19:08<33:42,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  36%|███▌      | 1429/4000 [19:09<33:35,  1.28it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.28it/s]train loss: 1.0616748332977295
train acc: 0.7020000219345093

overall:  36%|███▌      | 1430/4000 [19:10<33:36,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  36%|███▌      | 1431/4000 [19:11<33:39,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  36%|███▌      | 1432/4000 [19:11<33:36,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  36%|███▌      | 1433/4000 [19:12<33:44,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  36%|███▌      | 1434/4000 [19:13<33:39,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  36%|███▌      | 1435/4000 [19:14<33:36,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  36%|███▌      | 1436/4000 [19:14<33:33,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  36%|███▌      | 1437/4000 [19:15<33:34,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  36%|███▌      | 1438/4000 [19:16<33:31,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  36%|███▌      | 1439/4000 [19:17<33:32,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]train loss: 1.01677668094635
train acc: 0.7379999756813049

overall:  36%|███▌      | 1440/4000 [19:18<33:36,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  36%|███▌      | 1441/4000 [19:18<33:29,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  36%|███▌      | 1442/4000 [19:19<33:31,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  36%|███▌      | 1443/4000 [19:20<32:19,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  36%|███▌      | 1444/4000 [19:21<41:17,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  36%|███▌      | 1445/4000 [19:22<38:56,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.03s/it]
overall:  36%|███▌      | 1446/4000 [19:23<37:18,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  36%|███▌      | 1447/4000 [19:24<36:08,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]
overall:  36%|███▌      | 1448/4000 [19:24<35:20,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  36%|███▌      | 1449/4000 [19:25<35:00,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]train loss: 1.0130903720855713
train acc: 0.7155870199203491

overall:  36%|███▋      | 1450/4000 [19:26<34:35,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  36%|███▋      | 1451/4000 [19:27<34:11,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  36%|███▋      | 1452/4000 [19:28<33:53,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  36%|███▋      | 1453/4000 [19:28<33:42,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  36%|███▋      | 1454/4000 [19:29<33:39,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  36%|███▋      | 1455/4000 [19:30<33:31,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  36%|███▋      | 1456/4000 [19:31<33:35,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  36%|███▋      | 1457/4000 [19:32<33:26,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  36%|███▋      | 1458/4000 [19:32<33:20,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  36%|███▋      | 1459/4000 [19:33<33:16,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]train loss: 1.034232258796692
train acc: 0.7120000123977661

overall:  36%|███▋      | 1460/4000 [19:34<33:15,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  37%|███▋      | 1461/4000 [19:35<33:24,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  37%|███▋      | 1462/4000 [19:35<33:22,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  37%|███▋      | 1463/4000 [19:36<33:17,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  37%|███▋      | 1464/4000 [19:37<33:11,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  37%|███▋      | 1465/4000 [19:38<33:10,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  37%|███▋      | 1466/4000 [19:39<33:10,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  37%|███▋      | 1467/4000 [19:39<33:09,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  37%|███▋      | 1468/4000 [19:40<33:16,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  37%|███▋      | 1469/4000 [19:41<33:14,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]train loss: 1.0059404373168945
train acc: 0.7279999852180481

overall:  37%|███▋      | 1470/4000 [19:42<33:11,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  37%|███▋      | 1471/4000 [19:43<33:09,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  37%|███▋      | 1472/4000 [19:43<33:07,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  37%|███▋      | 1473/4000 [19:44<33:14,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  37%|███▋      | 1474/4000 [19:45<33:09,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.27it/s]
overall:  37%|███▋      | 1475/4000 [19:46<33:03,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  37%|███▋      | 1476/4000 [19:46<33:02,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  37%|███▋      | 1477/4000 [19:47<33:01,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  37%|███▋      | 1478/4000 [19:48<33:01,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  37%|███▋      | 1479/4000 [19:49<33:00,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]train loss: 1.0419632196426392
train acc: 0.7155870199203491

overall:  37%|███▋      | 1480/4000 [19:50<31:55,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  37%|███▋      | 1481/4000 [19:51<41:26,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  37%|███▋      | 1482/4000 [19:52<39:03,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:  37%|███▋      | 1483/4000 [19:53<37:15,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  37%|███▋      | 1484/4000 [19:53<35:58,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  37%|███▋      | 1485/4000 [19:54<34:59,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  37%|███▋      | 1486/4000 [19:55<34:22,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  37%|███▋      | 1487/4000 [19:56<34:01,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  37%|███▋      | 1488/4000 [19:57<33:36,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  37%|███▋      | 1489/4000 [19:57<33:32,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]train loss: 0.9842764139175415
train acc: 0.7289999723434448

overall:  37%|███▋      | 1490/4000 [19:58<33:19,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  37%|███▋      | 1491/4000 [19:59<33:13,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  37%|███▋      | 1492/4000 [20:00<33:06,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  37%|███▋      | 1493/4000 [20:01<33:10,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  37%|███▋      | 1494/4000 [20:01<33:07,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  37%|███▋      | 1495/4000 [20:02<33:03,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  37%|███▋      | 1496/4000 [20:03<32:56,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  37%|███▋      | 1497/4000 [20:04<32:56,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  37%|███▋      | 1498/4000 [20:04<32:51,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  37%|███▋      | 1499/4000 [20:05<32:51,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]train loss: 1.003835916519165
train acc: 0.718999981880188

overall:  38%|███▊      | 1500/4000 [20:06<32:54,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  38%|███▊      | 1501/4000 [20:07<32:55,  1.26it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.26it/s]
overall:  38%|███▊      | 1502/4000 [20:08<32:52,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  38%|███▊      | 1503/4000 [20:08<32:48,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  38%|███▊      | 1504/4000 [20:09<32:49,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  38%|███▊      | 1505/4000 [20:10<32:50,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  38%|███▊      | 1506/4000 [20:11<32:49,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  38%|███▊      | 1507/4000 [20:12<32:45,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  38%|███▊      | 1508/4000 [20:12<32:43,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  38%|███▊      | 1509/4000 [20:13<32:40,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]train loss: 1.0072907209396362
train acc: 0.7250000238418579

overall:  38%|███▊      | 1510/4000 [20:14<32:36,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  38%|███▊      | 1511/4000 [20:15<32:32,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  38%|███▊      | 1512/4000 [20:16<32:29,  1.28it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.28it/s]
overall:  38%|███▊      | 1513/4000 [20:16<32:34,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  38%|███▊      | 1514/4000 [20:17<32:32,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  38%|███▊      | 1515/4000 [20:18<32:27,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:  38%|███▊      | 1516/4000 [20:19<32:27,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.28it/s]
overall:  38%|███▊      | 1517/4000 [20:19<31:20,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  38%|███▊      | 1518/4000 [20:21<40:13,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.38s/it]
overall:  38%|███▊      | 1519/4000 [20:22<37:53,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]train loss: 1.0405912399291992
train acc: 0.72773277759552

overall:  38%|███▊      | 1520/4000 [20:22<36:22,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  38%|███▊      | 1521/4000 [20:23<35:08,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  38%|███▊      | 1522/4000 [20:24<34:19,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  38%|███▊      | 1523/4000 [20:25<33:51,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  38%|███▊      | 1524/4000 [20:26<33:24,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  38%|███▊      | 1525/4000 [20:26<33:11,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  38%|███▊      | 1526/4000 [20:27<32:52,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  38%|███▊      | 1527/4000 [20:28<32:41,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  38%|███▊      | 1528/4000 [20:29<32:37,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  38%|███▊      | 1529/4000 [20:29<32:33,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]train loss: 0.9731117486953735
train acc: 0.7360000014305115

overall:  38%|███▊      | 1530/4000 [20:30<32:39,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  38%|███▊      | 1531/4000 [20:31<32:30,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  38%|███▊      | 1532/4000 [20:32<32:22,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  38%|███▊      | 1533/4000 [20:33<32:18,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  38%|███▊      | 1534/4000 [20:33<32:17,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  38%|███▊      | 1535/4000 [20:34<32:18,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  38%|███▊      | 1536/4000 [20:35<32:15,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  38%|███▊      | 1537/4000 [20:36<32:19,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  38%|███▊      | 1538/4000 [20:37<32:15,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  38%|███▊      | 1539/4000 [20:37<32:10,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]train loss: 1.0235427618026733
train acc: 0.7089999914169312

overall:  38%|███▊      | 1540/4000 [20:38<33:46,  1.21it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.21it/s]
overall:  39%|███▊      | 1541/4000 [20:39<33:15,  1.23it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.23it/s]
overall:  39%|███▊      | 1542/4000 [20:40<33:02,  1.24it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.24it/s]
overall:  39%|███▊      | 1543/4000 [20:41<32:45,  1.25it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.25it/s]
overall:  39%|███▊      | 1544/4000 [20:41<32:28,  1.26it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.26it/s]
overall:  39%|███▊      | 1545/4000 [20:42<32:21,  1.26it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.26it/s]
overall:  39%|███▊      | 1546/4000 [20:43<32:13,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  39%|███▊      | 1547/4000 [20:44<32:15,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  39%|███▊      | 1548/4000 [20:45<32:11,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  39%|███▊      | 1549/4000 [20:45<32:10,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]train loss: 1.0075085163116455
train acc: 0.7269999980926514

overall:  39%|███▉      | 1550/4000 [20:46<32:09,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  39%|███▉      | 1551/4000 [20:47<32:05,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  39%|███▉      | 1552/4000 [20:48<32:01,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  39%|███▉      | 1553/4000 [20:48<32:00,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  39%|███▉      | 1554/4000 [20:49<30:55,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  39%|███▉      | 1555/4000 [20:51<40:27,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  39%|███▉      | 1556/4000 [20:51<37:58,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:  39%|███▉      | 1557/4000 [20:52<36:10,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  39%|███▉      | 1558/4000 [20:53<34:54,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  39%|███▉      | 1559/4000 [20:54<34:01,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]train loss: 0.9673290252685547
train acc: 0.7378542423248291

overall:  39%|███▉      | 1560/4000 [20:55<33:25,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  39%|███▉      | 1561/4000 [20:55<33:05,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  39%|███▉      | 1562/4000 [20:56<32:48,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  39%|███▉      | 1563/4000 [20:57<32:33,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  39%|███▉      | 1564/4000 [20:58<32:16,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  39%|███▉      | 1565/4000 [20:59<32:09,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  39%|███▉      | 1566/4000 [20:59<32:08,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  39%|███▉      | 1567/4000 [21:00<32:15,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  39%|███▉      | 1568/4000 [21:01<32:06,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  39%|███▉      | 1569/4000 [21:02<32:02,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]train loss: 0.9839408993721008
train acc: 0.7329999804496765

overall:  39%|███▉      | 1570/4000 [21:03<31:57,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  39%|███▉      | 1571/4000 [21:03<31:52,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  39%|███▉      | 1572/4000 [21:04<31:48,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  39%|███▉      | 1573/4000 [21:05<31:50,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  39%|███▉      | 1574/4000 [21:06<31:56,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  39%|███▉      | 1575/4000 [21:06<31:53,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  39%|███▉      | 1576/4000 [21:07<31:49,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  39%|███▉      | 1577/4000 [21:08<31:49,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  39%|███▉      | 1578/4000 [21:09<31:47,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  39%|███▉      | 1579/4000 [21:10<31:55,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]train loss: 0.9978383779525757
train acc: 0.7450000047683716

overall:  40%|███▉      | 1580/4000 [21:10<31:51,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  40%|███▉      | 1581/4000 [21:11<31:46,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  40%|███▉      | 1582/4000 [21:12<31:44,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  40%|███▉      | 1583/4000 [21:13<31:37,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  40%|███▉      | 1584/4000 [21:14<31:36,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  40%|███▉      | 1585/4000 [21:14<31:40,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  40%|███▉      | 1586/4000 [21:15<31:41,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  40%|███▉      | 1587/4000 [21:16<31:40,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  40%|███▉      | 1588/4000 [21:17<31:39,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  40%|███▉      | 1589/4000 [21:17<31:33,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]train loss: 1.0183459520339966
train acc: 0.7059999704360962

overall:  40%|███▉      | 1590/4000 [21:18<31:33,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  40%|███▉      | 1591/4000 [21:19<30:28,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  40%|███▉      | 1592/4000 [21:21<39:50,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  40%|███▉      | 1593/4000 [21:21<37:26,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:  40%|███▉      | 1594/4000 [21:22<35:35,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  40%|███▉      | 1595/4000 [21:23<34:23,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  40%|███▉      | 1596/4000 [21:24<33:34,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  40%|███▉      | 1597/4000 [21:24<32:59,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  40%|███▉      | 1598/4000 [21:25<32:33,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  40%|███▉      | 1599/4000 [21:26<32:20,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]train loss: 0.9194039106369019
train acc: 0.7580971717834473

overall:  40%|████      | 1600/4000 [21:27<32:02,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  40%|████      | 1601/4000 [21:28<31:49,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  40%|████      | 1602/4000 [21:28<31:42,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  40%|████      | 1603/4000 [21:29<31:36,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  40%|████      | 1604/4000 [21:30<31:39,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  40%|████      | 1605/4000 [21:31<31:37,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  40%|████      | 1606/4000 [21:32<31:32,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  40%|████      | 1607/4000 [21:32<31:28,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  40%|████      | 1608/4000 [21:33<31:23,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  40%|████      | 1609/4000 [21:34<31:22,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]train loss: 0.982042670249939
train acc: 0.7289999723434448

overall:  40%|████      | 1610/4000 [21:35<31:20,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  40%|████      | 1611/4000 [21:35<31:26,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  40%|████      | 1612/4000 [21:36<31:20,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  40%|████      | 1613/4000 [21:37<31:16,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  40%|████      | 1614/4000 [21:38<31:11,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  40%|████      | 1615/4000 [21:39<31:10,  1.28it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.28it/s]
overall:  40%|████      | 1616/4000 [21:39<31:17,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  40%|████      | 1617/4000 [21:40<31:17,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  40%|████      | 1618/4000 [21:41<31:14,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  40%|████      | 1619/4000 [21:42<31:12,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]train loss: 1.0074225664138794
train acc: 0.722000002861023

overall:  40%|████      | 1620/4000 [21:43<31:08,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  41%|████      | 1621/4000 [21:43<31:05,  1.28it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.28it/s]
overall:  41%|████      | 1622/4000 [21:44<31:02,  1.28it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.28it/s]
overall:  41%|████      | 1623/4000 [21:45<31:08,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  41%|████      | 1624/4000 [21:46<31:06,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  41%|████      | 1625/4000 [21:46<31:01,  1.28it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.28it/s]
overall:  41%|████      | 1626/4000 [21:47<30:57,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:  41%|████      | 1627/4000 [21:48<30:59,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:  41%|████      | 1628/4000 [21:49<29:57,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  41%|████      | 1629/4000 [21:50<39:18,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]train loss: 0.9986674189567566
train acc: 0.7246963381767273

overall:  41%|████      | 1630/4000 [21:51<36:49,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.06s/it]
overall:  41%|████      | 1631/4000 [21:52<35:04,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  41%|████      | 1632/4000 [21:53<33:49,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  41%|████      | 1633/4000 [21:53<32:58,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  41%|████      | 1634/4000 [21:54<32:21,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  41%|████      | 1635/4000 [21:55<31:54,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  41%|████      | 1636/4000 [21:56<31:42,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  41%|████      | 1637/4000 [21:57<31:33,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  41%|████      | 1638/4000 [21:57<31:20,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  41%|████      | 1639/4000 [21:58<31:10,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]train loss: 0.9823722839355469
train acc: 0.7310000061988831

overall:  41%|████      | 1640/4000 [21:59<31:05,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  41%|████      | 1641/4000 [22:00<31:17,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  41%|████      | 1642/4000 [22:01<31:06,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  41%|████      | 1643/4000 [22:01<31:02,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  41%|████      | 1644/4000 [22:02<31:02,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  41%|████      | 1645/4000 [22:03<30:53,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  41%|████      | 1646/4000 [22:04<30:58,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.27it/s]
overall:  41%|████      | 1647/4000 [22:04<30:53,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  41%|████      | 1648/4000 [22:05<30:48,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  41%|████      | 1649/4000 [22:06<30:53,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]train loss: 0.9732677340507507
train acc: 0.7269999980926514

overall:  41%|████▏     | 1650/4000 [22:07<30:51,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  41%|████▏     | 1651/4000 [22:08<30:47,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  41%|████▏     | 1652/4000 [22:08<30:44,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  41%|████▏     | 1653/4000 [22:09<30:55,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  41%|████▏     | 1654/4000 [22:10<30:52,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  41%|████▏     | 1655/4000 [22:11<30:44,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  41%|████▏     | 1656/4000 [22:12<30:41,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  41%|████▏     | 1657/4000 [22:12<30:36,  1.28it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.28it/s]
overall:  41%|████▏     | 1658/4000 [22:13<30:35,  1.28it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.28it/s]
overall:  41%|████▏     | 1659/4000 [22:14<30:34,  1.28it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.28it/s]train loss: 0.9618740081787109
train acc: 0.7409999966621399

overall:  42%|████▏     | 1660/4000 [22:15<30:38,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  42%|████▏     | 1661/4000 [22:15<30:36,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  42%|████▏     | 1662/4000 [22:16<30:42,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  42%|████▏     | 1663/4000 [22:17<30:36,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  42%|████▏     | 1664/4000 [22:18<30:34,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  42%|████▏     | 1665/4000 [22:19<29:30,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  42%|████▏     | 1666/4000 [22:20<37:34,  1.04it/s]

train:   3%|▎         | 1/37 [00:01<00:48,  1.36s/it]
overall:  42%|████▏     | 1667/4000 [22:21<35:28,  1.10it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.02s/it]
overall:  42%|████▏     | 1668/4000 [22:22<33:55,  1.15it/s]

train:   8%|▊         | 3/37 [00:02<00:30,  1.10it/s]
overall:  42%|████▏     | 1669/4000 [22:22<33:00,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]train loss: 0.9741202592849731
train acc: 0.7155870199203491

overall:  42%|████▏     | 1670/4000 [22:23<32:15,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  42%|████▏     | 1671/4000 [22:24<31:43,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  42%|████▏     | 1672/4000 [22:25<31:24,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  42%|████▏     | 1673/4000 [22:25<31:03,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  42%|████▏     | 1674/4000 [22:26<31:01,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  42%|████▏     | 1675/4000 [22:27<30:49,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  42%|████▏     | 1676/4000 [22:28<30:46,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  42%|████▏     | 1677/4000 [22:29<30:38,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  42%|████▏     | 1678/4000 [22:29<30:38,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  42%|████▏     | 1679/4000 [22:30<30:36,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]train loss: 1.000287652015686
train acc: 0.7329999804496765

overall:  42%|████▏     | 1680/4000 [22:31<30:35,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  42%|████▏     | 1681/4000 [22:32<30:39,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  42%|████▏     | 1682/4000 [22:33<30:32,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  42%|████▏     | 1683/4000 [22:33<30:28,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  42%|████▏     | 1684/4000 [22:34<30:30,  1.26it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.26it/s]
overall:  42%|████▏     | 1685/4000 [22:35<30:22,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  42%|████▏     | 1686/4000 [22:36<30:27,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  42%|████▏     | 1687/4000 [22:37<30:22,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  42%|████▏     | 1688/4000 [22:37<30:27,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  42%|████▏     | 1689/4000 [22:38<30:23,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]train loss: 0.982506275177002
train acc: 0.7310000061988831

overall:  42%|████▏     | 1690/4000 [22:39<30:26,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  42%|████▏     | 1691/4000 [22:40<30:25,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  42%|████▏     | 1692/4000 [22:40<30:20,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  42%|████▏     | 1693/4000 [22:41<30:21,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  42%|████▏     | 1694/4000 [22:42<30:20,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  42%|████▏     | 1695/4000 [22:43<30:16,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  42%|████▏     | 1696/4000 [22:44<30:18,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  42%|████▏     | 1697/4000 [22:44<30:18,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  42%|████▏     | 1698/4000 [22:45<30:17,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  42%|████▏     | 1699/4000 [22:46<30:15,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]train loss: 0.9481980204582214
train acc: 0.7300000190734863

overall:  42%|████▎     | 1700/4000 [22:47<30:15,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  43%|████▎     | 1701/4000 [22:48<30:13,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  43%|████▎     | 1702/4000 [22:48<29:06,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  43%|████▎     | 1703/4000 [22:50<37:11,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  43%|████▎     | 1704/4000 [22:51<35:11,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  43%|████▎     | 1705/4000 [22:51<33:40,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  43%|████▎     | 1706/4000 [22:52<32:35,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  43%|████▎     | 1707/4000 [22:53<31:51,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  43%|████▎     | 1708/4000 [22:54<31:16,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  43%|████▎     | 1709/4000 [22:54<30:58,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]train loss: 0.9450367093086243
train acc: 0.7489878535270691

overall:  43%|████▎     | 1710/4000 [22:55<30:48,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  43%|████▎     | 1711/4000 [22:56<30:32,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  43%|████▎     | 1712/4000 [22:57<30:26,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  43%|████▎     | 1713/4000 [22:58<30:15,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  43%|████▎     | 1714/4000 [22:58<30:08,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  43%|████▎     | 1715/4000 [22:59<30:14,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  43%|████▎     | 1716/4000 [23:00<30:14,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  43%|████▎     | 1717/4000 [23:01<30:05,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  43%|████▎     | 1718/4000 [23:02<29:59,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  43%|████▎     | 1719/4000 [23:02<29:57,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]train loss: 0.9420889616012573
train acc: 0.7419999837875366

overall:  43%|████▎     | 1720/4000 [23:03<29:55,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  43%|████▎     | 1721/4000 [23:04<29:49,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  43%|████▎     | 1722/4000 [23:05<29:47,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  43%|████▎     | 1723/4000 [23:06<29:46,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  43%|████▎     | 1724/4000 [23:06<29:49,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  43%|████▎     | 1725/4000 [23:07<29:52,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  43%|████▎     | 1726/4000 [23:08<29:51,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  43%|████▎     | 1727/4000 [23:09<29:59,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  43%|████▎     | 1728/4000 [23:09<29:58,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  43%|████▎     | 1729/4000 [23:10<29:54,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]train loss: 0.9839800596237183
train acc: 0.7360000014305115

overall:  43%|████▎     | 1730/4000 [23:11<29:48,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  43%|████▎     | 1731/4000 [23:12<29:43,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  43%|████▎     | 1732/4000 [23:13<29:47,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  43%|████▎     | 1733/4000 [23:13<29:42,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  43%|████▎     | 1734/4000 [23:14<29:39,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  43%|████▎     | 1735/4000 [23:15<29:39,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  43%|████▎     | 1736/4000 [23:16<29:41,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  43%|████▎     | 1737/4000 [23:17<29:36,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  43%|████▎     | 1738/4000 [23:17<29:32,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:  43%|████▎     | 1739/4000 [23:18<28:29,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]train loss: 0.954929530620575
train acc: 0.7246963381767273

overall:  44%|████▎     | 1740/4000 [23:20<37:12,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  44%|████▎     | 1741/4000 [23:20<34:58,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.06s/it]
overall:  44%|████▎     | 1742/4000 [23:21<33:20,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  44%|████▎     | 1743/4000 [23:22<32:08,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  44%|████▎     | 1744/4000 [23:23<31:21,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  44%|████▎     | 1745/4000 [23:23<30:49,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  44%|████▎     | 1746/4000 [23:24<30:23,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  44%|████▎     | 1747/4000 [23:25<30:06,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  44%|████▎     | 1748/4000 [23:26<29:54,  1.26it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.26it/s]
overall:  44%|████▎     | 1749/4000 [23:27<29:49,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]train loss: 0.9813885688781738
train acc: 0.7260000109672546

overall:  44%|████▍     | 1750/4000 [23:27<29:41,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  44%|████▍     | 1751/4000 [23:28<29:39,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  44%|████▍     | 1752/4000 [23:29<29:45,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  44%|████▍     | 1753/4000 [23:30<29:39,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  44%|████▍     | 1754/4000 [23:31<29:33,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  44%|████▍     | 1755/4000 [23:31<29:28,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  44%|████▍     | 1756/4000 [23:32<29:24,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  44%|████▍     | 1757/4000 [23:33<29:23,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  44%|████▍     | 1758/4000 [23:34<29:21,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  44%|████▍     | 1759/4000 [23:35<29:26,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]train loss: 0.96002596616745
train acc: 0.7429999709129333

overall:  44%|████▍     | 1760/4000 [23:35<29:21,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  44%|████▍     | 1761/4000 [23:36<29:18,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  44%|████▍     | 1762/4000 [23:37<29:21,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  44%|████▍     | 1763/4000 [23:38<29:18,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  44%|████▍     | 1764/4000 [23:38<29:29,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  44%|████▍     | 1765/4000 [23:39<29:28,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  44%|████▍     | 1766/4000 [23:40<29:22,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  44%|████▍     | 1767/4000 [23:41<29:16,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  44%|████▍     | 1768/4000 [23:42<29:12,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  44%|████▍     | 1769/4000 [23:42<29:10,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]train loss: 0.9091349840164185
train acc: 0.7620000243186951

overall:  44%|████▍     | 1770/4000 [23:43<29:13,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  44%|████▍     | 1771/4000 [23:44<29:12,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  44%|████▍     | 1772/4000 [23:45<29:12,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  44%|████▍     | 1773/4000 [23:46<29:07,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  44%|████▍     | 1774/4000 [23:46<29:07,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  44%|████▍     | 1775/4000 [23:47<29:05,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  44%|████▍     | 1776/4000 [23:48<28:08,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  44%|████▍     | 1777/4000 [23:49<36:44,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  44%|████▍     | 1778/4000 [23:50<34:28,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  44%|████▍     | 1779/4000 [23:51<32:55,  1.12it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]train loss: 0.9679291844367981
train acc: 0.7236841917037964

overall:  44%|████▍     | 1780/4000 [23:52<31:46,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  45%|████▍     | 1781/4000 [23:52<30:58,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  45%|████▍     | 1782/4000 [23:53<30:25,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  45%|████▍     | 1783/4000 [23:54<30:01,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  45%|████▍     | 1784/4000 [23:55<29:42,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  45%|████▍     | 1785/4000 [23:56<29:36,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  45%|████▍     | 1786/4000 [23:56<29:24,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  45%|████▍     | 1787/4000 [23:57<29:13,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  45%|████▍     | 1788/4000 [23:58<29:11,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  45%|████▍     | 1789/4000 [23:59<29:15,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]train loss: 0.9095133543014526
train acc: 0.7570000290870667

overall:  45%|████▍     | 1790/4000 [24:00<29:12,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  45%|████▍     | 1791/4000 [24:00<29:04,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  45%|████▍     | 1792/4000 [24:01<29:02,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  45%|████▍     | 1793/4000 [24:02<29:00,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  45%|████▍     | 1794/4000 [24:03<28:56,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  45%|████▍     | 1795/4000 [24:04<28:57,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  45%|████▍     | 1796/4000 [24:04<28:54,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  45%|████▍     | 1797/4000 [24:05<29:01,  1.26it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.26it/s]
overall:  45%|████▍     | 1798/4000 [24:06<28:56,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  45%|████▍     | 1799/4000 [24:07<28:53,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]train loss: 0.9656535387039185
train acc: 0.7450000047683716

overall:  45%|████▌     | 1800/4000 [24:07<28:54,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  45%|████▌     | 1801/4000 [24:08<29:00,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  45%|████▌     | 1802/4000 [24:09<28:55,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  45%|████▌     | 1803/4000 [24:10<28:50,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  45%|████▌     | 1804/4000 [24:11<28:52,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  45%|████▌     | 1805/4000 [24:11<28:45,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  45%|████▌     | 1806/4000 [24:12<28:43,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  45%|████▌     | 1807/4000 [24:13<28:40,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  45%|████▌     | 1808/4000 [24:14<28:43,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  45%|████▌     | 1809/4000 [24:15<28:49,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]train loss: 0.9821141958236694
train acc: 0.718999981880188

overall:  45%|████▌     | 1810/4000 [24:15<28:48,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  45%|████▌     | 1811/4000 [24:16<28:46,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  45%|████▌     | 1812/4000 [24:17<28:44,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  45%|████▌     | 1813/4000 [24:18<27:41,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  45%|████▌     | 1814/4000 [24:19<35:48,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:50,  1.41s/it]
overall:  45%|████▌     | 1815/4000 [24:20<33:48,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  45%|████▌     | 1816/4000 [24:21<32:14,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.07it/s]
overall:  45%|████▌     | 1817/4000 [24:21<31:03,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  45%|████▌     | 1818/4000 [24:22<30:17,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  45%|████▌     | 1819/4000 [24:23<29:49,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]train loss: 0.9245592355728149
train acc: 0.7520242929458618

overall:  46%|████▌     | 1820/4000 [24:24<29:26,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  46%|████▌     | 1821/4000 [24:25<29:08,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  46%|████▌     | 1822/4000 [24:25<29:00,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  46%|████▌     | 1823/4000 [24:26<28:51,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  46%|████▌     | 1824/4000 [24:27<28:46,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  46%|████▌     | 1825/4000 [24:28<28:43,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  46%|████▌     | 1826/4000 [24:29<28:41,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  46%|████▌     | 1827/4000 [24:29<28:49,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  46%|████▌     | 1828/4000 [24:30<28:45,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  46%|████▌     | 1829/4000 [24:31<28:37,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]train loss: 0.9144874811172485
train acc: 0.7450000047683716

overall:  46%|████▌     | 1830/4000 [24:32<28:32,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  46%|████▌     | 1831/4000 [24:33<28:32,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.27it/s]
overall:  46%|████▌     | 1832/4000 [24:33<28:28,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  46%|████▌     | 1833/4000 [24:34<28:25,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  46%|████▌     | 1834/4000 [24:35<28:27,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  46%|████▌     | 1835/4000 [24:36<28:22,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  46%|████▌     | 1836/4000 [24:36<28:19,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  46%|████▌     | 1837/4000 [24:37<28:19,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  46%|████▌     | 1838/4000 [24:38<28:21,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  46%|████▌     | 1839/4000 [24:39<28:28,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]train loss: 0.9407733082771301
train acc: 0.7390000224113464

overall:  46%|████▌     | 1840/4000 [24:40<28:24,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  46%|████▌     | 1841/4000 [24:40<28:20,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  46%|████▌     | 1842/4000 [24:41<28:19,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  46%|████▌     | 1843/4000 [24:42<28:19,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  46%|████▌     | 1844/4000 [24:43<28:20,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  46%|████▌     | 1845/4000 [24:44<28:15,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  46%|████▌     | 1846/4000 [24:44<28:10,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  46%|████▌     | 1847/4000 [24:45<28:07,  1.28it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.28it/s]
overall:  46%|████▌     | 1848/4000 [24:46<28:11,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  46%|████▌     | 1849/4000 [24:47<28:11,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]train loss: 0.9887983202934265
train acc: 0.7186234593391418

overall:  46%|████▋     | 1850/4000 [24:47<27:11,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  46%|████▋     | 1851/4000 [24:49<34:42,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  46%|████▋     | 1852/4000 [24:50<32:41,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.02s/it]
overall:  46%|████▋     | 1853/4000 [24:50<31:14,  1.15it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.10it/s]
overall:  46%|████▋     | 1854/4000 [24:51<30:16,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]
overall:  46%|████▋     | 1855/4000 [24:52<29:37,  1.21it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.20it/s]
overall:  46%|████▋     | 1856/4000 [24:53<29:12,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  46%|████▋     | 1857/4000 [24:54<28:51,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.24it/s]
overall:  46%|████▋     | 1858/4000 [24:54<28:38,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  46%|████▋     | 1859/4000 [24:55<28:28,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]train loss: 0.9017260670661926
train acc: 0.746999979019165

overall:  46%|████▋     | 1860/4000 [24:56<28:26,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  47%|████▋     | 1861/4000 [24:57<28:19,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  47%|████▋     | 1862/4000 [24:57<28:16,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  47%|████▋     | 1863/4000 [24:58<28:29,  1.25it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.25it/s]
overall:  47%|████▋     | 1864/4000 [24:59<28:19,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  47%|████▋     | 1865/4000 [25:00<28:17,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  47%|████▋     | 1866/4000 [25:01<28:09,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  47%|████▋     | 1867/4000 [25:01<28:03,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  47%|████▋     | 1868/4000 [25:02<28:01,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  47%|████▋     | 1869/4000 [25:03<28:01,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]train loss: 0.9728196263313293
train acc: 0.7329999804496765

overall:  47%|████▋     | 1870/4000 [25:04<27:56,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  47%|████▋     | 1871/4000 [25:05<27:54,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  47%|████▋     | 1872/4000 [25:05<27:56,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  47%|████▋     | 1873/4000 [25:06<27:54,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  47%|████▋     | 1874/4000 [25:07<27:55,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  47%|████▋     | 1875/4000 [25:08<28:05,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  47%|████▋     | 1876/4000 [25:09<27:59,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  47%|████▋     | 1877/4000 [25:09<27:54,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  47%|████▋     | 1878/4000 [25:10<27:52,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  47%|████▋     | 1879/4000 [25:11<27:48,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]train loss: 0.9477141499519348
train acc: 0.7300000190734863

overall:  47%|████▋     | 1880/4000 [25:12<27:48,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  47%|████▋     | 1881/4000 [25:12<27:47,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  47%|████▋     | 1882/4000 [25:13<27:47,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  47%|████▋     | 1883/4000 [25:14<27:45,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  47%|████▋     | 1884/4000 [25:15<27:45,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  47%|████▋     | 1885/4000 [25:16<27:42,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  47%|████▋     | 1886/4000 [25:16<27:41,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  47%|████▋     | 1887/4000 [25:17<26:44,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  47%|████▋     | 1888/4000 [25:19<36:36,  1.04s/it]

train:   3%|▎         | 1/37 [00:01<00:57,  1.59s/it]
overall:  47%|████▋     | 1889/4000 [25:20<33:52,  1.04it/s]

train:   5%|▌         | 2/37 [00:02<00:39,  1.11s/it]train loss: 0.8978236317634583
train acc: 0.7601214647293091

overall:  47%|████▋     | 1890/4000 [25:20<32:04,  1.10it/s]

train:   8%|▊         | 3/37 [00:03<00:32,  1.03it/s]
overall:  47%|████▋     | 1891/4000 [25:21<30:46,  1.14it/s]

train:  11%|█         | 4/37 [00:03<00:29,  1.11it/s]
overall:  47%|████▋     | 1892/4000 [25:22<29:50,  1.18it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.17it/s]
overall:  47%|████▋     | 1893/4000 [25:23<29:07,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.20it/s]
overall:  47%|████▋     | 1894/4000 [25:24<28:40,  1.22it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.22it/s]
overall:  47%|████▋     | 1895/4000 [25:24<28:24,  1.23it/s]

train:  22%|██▏       | 8/37 [00:07<00:23,  1.23it/s]
overall:  47%|████▋     | 1896/4000 [25:25<28:11,  1.24it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.24it/s]
overall:  47%|████▋     | 1897/4000 [25:26<27:58,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  47%|████▋     | 1898/4000 [25:27<27:47,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  47%|████▋     | 1899/4000 [25:27<27:42,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]train loss: 0.9456837773323059
train acc: 0.7480000257492065

overall:  48%|████▊     | 1900/4000 [25:28<27:42,  1.26it/s]

train:  35%|███▌      | 13/37 [00:11<00:19,  1.26it/s]
overall:  48%|████▊     | 1901/4000 [25:29<27:36,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  48%|████▊     | 1902/4000 [25:30<27:36,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  48%|████▊     | 1903/4000 [25:31<27:34,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  48%|████▊     | 1904/4000 [25:31<27:30,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  48%|████▊     | 1905/4000 [25:32<27:28,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  48%|████▊     | 1906/4000 [25:33<27:28,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  48%|████▊     | 1907/4000 [25:34<27:30,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  48%|████▊     | 1908/4000 [25:35<27:28,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  48%|████▊     | 1909/4000 [25:35<27:23,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]train loss: 0.9509231448173523
train acc: 0.7329999804496765

overall:  48%|████▊     | 1910/4000 [25:36<27:21,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  48%|████▊     | 1911/4000 [25:37<27:23,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  48%|████▊     | 1912/4000 [25:38<27:21,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  48%|████▊     | 1913/4000 [25:38<27:18,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  48%|████▊     | 1914/4000 [25:39<27:16,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:22<00:07,  1.27it/s]
overall:  48%|████▊     | 1915/4000 [25:40<27:22,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  48%|████▊     | 1916/4000 [25:41<27:16,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  48%|████▊     | 1917/4000 [25:42<27:13,  1.28it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.28it/s]
overall:  48%|████▊     | 1918/4000 [25:42<27:13,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  48%|████▊     | 1919/4000 [25:43<27:16,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]train loss: 0.9398124814033508
train acc: 0.7300000190734863

overall:  48%|████▊     | 1920/4000 [25:44<27:13,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  48%|████▊     | 1921/4000 [25:45<27:10,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  48%|████▊     | 1922/4000 [25:46<27:12,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  48%|████▊     | 1923/4000 [25:46<27:10,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  48%|████▊     | 1924/4000 [25:47<26:11,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  48%|████▊     | 1925/4000 [25:48<33:15,  1.04it/s]

train:   3%|▎         | 1/37 [00:01<00:48,  1.35s/it]
overall:  48%|████▊     | 1926/4000 [25:49<31:37,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.03s/it]
overall:  48%|████▊     | 1927/4000 [25:50<30:16,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  48%|████▊     | 1928/4000 [25:51<29:17,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]
overall:  48%|████▊     | 1929/4000 [25:52<28:39,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]train loss: 0.8673237562179565
train acc: 0.7773279547691345

overall:  48%|████▊     | 1930/4000 [25:52<28:13,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  48%|████▊     | 1931/4000 [25:53<27:52,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.24it/s]
overall:  48%|████▊     | 1932/4000 [25:54<27:39,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  48%|████▊     | 1933/4000 [25:55<27:26,  1.26it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.26it/s]
overall:  48%|████▊     | 1934/4000 [25:56<27:17,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  48%|████▊     | 1935/4000 [25:56<27:11,  1.27it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.27it/s]
overall:  48%|████▊     | 1936/4000 [25:57<27:11,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  48%|████▊     | 1937/4000 [25:58<27:12,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.26it/s]
overall:  48%|████▊     | 1938/4000 [25:59<27:21,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  48%|████▊     | 1939/4000 [26:00<27:12,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]train loss: 0.9185382723808289
train acc: 0.7440000176429749

overall:  48%|████▊     | 1940/4000 [26:00<27:11,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  49%|████▊     | 1941/4000 [26:01<27:03,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  49%|████▊     | 1942/4000 [26:02<27:00,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  49%|████▊     | 1943/4000 [26:03<26:54,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  49%|████▊     | 1944/4000 [26:03<26:59,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  49%|████▊     | 1945/4000 [26:04<26:54,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  49%|████▊     | 1946/4000 [26:05<26:51,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  49%|████▊     | 1947/4000 [26:06<26:47,  1.28it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.28it/s]
overall:  49%|████▊     | 1948/4000 [26:07<26:46,  1.28it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.28it/s]
overall:  49%|████▊     | 1949/4000 [26:07<26:45,  1.28it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.28it/s]train loss: 0.9195948839187622
train acc: 0.7599999904632568

overall:  49%|████▉     | 1950/4000 [26:08<26:56,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  49%|████▉     | 1951/4000 [26:09<26:55,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  49%|████▉     | 1952/4000 [26:10<26:51,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  49%|████▉     | 1953/4000 [26:11<26:46,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  49%|████▉     | 1954/4000 [26:11<26:42,  1.28it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.28it/s]
overall:  49%|████▉     | 1955/4000 [26:12<26:40,  1.28it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.28it/s]
overall:  49%|████▉     | 1956/4000 [26:13<26:46,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  49%|████▉     | 1957/4000 [26:14<26:43,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  49%|████▉     | 1958/4000 [26:14<26:40,  1.28it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.28it/s]
overall:  49%|████▉     | 1959/4000 [26:15<26:37,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]train loss: 0.9701743125915527
train acc: 0.7120000123977661

overall:  49%|████▉     | 1960/4000 [26:16<26:37,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:  49%|████▉     | 1961/4000 [26:17<25:38,  1.33it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.33it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  49%|████▉     | 1962/4000 [26:18<33:05,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:50,  1.40s/it]
overall:  49%|████▉     | 1963/4000 [26:19<31:08,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.04s/it]
overall:  49%|████▉     | 1964/4000 [26:20<29:47,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  49%|████▉     | 1965/4000 [26:21<28:51,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  49%|████▉     | 1966/4000 [26:21<28:09,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  49%|████▉     | 1967/4000 [26:22<27:40,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  49%|████▉     | 1968/4000 [26:23<27:24,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  49%|████▉     | 1969/4000 [26:24<27:10,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]train loss: 0.9129800796508789
train acc: 0.7550607323646545

overall:  49%|████▉     | 1970/4000 [26:24<27:04,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  49%|████▉     | 1971/4000 [26:25<26:58,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  49%|████▉     | 1972/4000 [26:26<26:51,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  49%|████▉     | 1973/4000 [26:27<26:42,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  49%|████▉     | 1974/4000 [26:28<26:48,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  49%|████▉     | 1975/4000 [26:28<26:47,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  49%|████▉     | 1976/4000 [26:29<26:38,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  49%|████▉     | 1977/4000 [26:30<26:34,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  49%|████▉     | 1978/4000 [26:31<26:31,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  49%|████▉     | 1979/4000 [26:32<26:32,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]train loss: 0.9193045496940613
train acc: 0.753000020980835

overall:  50%|████▉     | 1980/4000 [26:32<26:36,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.26it/s]
overall:  50%|████▉     | 1981/4000 [26:33<26:31,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  50%|████▉     | 1982/4000 [26:34<26:27,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  50%|████▉     | 1983/4000 [26:35<26:25,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  50%|████▉     | 1984/4000 [26:35<26:24,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  50%|████▉     | 1985/4000 [26:36<26:23,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  50%|████▉     | 1986/4000 [26:37<26:30,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  50%|████▉     | 1987/4000 [26:38<26:27,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  50%|████▉     | 1988/4000 [26:39<26:21,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  50%|████▉     | 1989/4000 [26:39<26:20,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]train loss: 0.9311115145683289
train acc: 0.7350000143051147

overall:  50%|████▉     | 1990/4000 [26:40<26:21,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  50%|████▉     | 1991/4000 [26:41<26:14,  1.28it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.28it/s]
overall:  50%|████▉     | 1992/4000 [26:42<26:19,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  50%|████▉     | 1993/4000 [26:43<26:18,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  50%|████▉     | 1994/4000 [26:43<26:16,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  50%|████▉     | 1995/4000 [26:44<26:12,  1.28it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.28it/s]
overall:  50%|████▉     | 1996/4000 [26:45<26:08,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:  50%|████▉     | 1997/4000 [26:46<26:08,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:  50%|████▉     | 1998/4000 [26:46<25:14,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  50%|████▉     | 1999/4000 [26:48<32:46,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.42s/it]train loss: 0.9275500178337097
train acc: 0.7348178029060364



test:   0%|          | 0/15 [00:00<?, ?it/s]


test:   7%|▋         | 1/15 [00:01<00:19,  1.38s/it]


test:  13%|█▎        | 2/15 [00:02<00:13,  1.02s/it]


test:  20%|██        | 3/15 [00:02<00:10,  1.12it/s]


test:  27%|██▋       | 4/15 [00:03<00:09,  1.19it/s]


test:  33%|███▎      | 5/15 [00:04<00:08,  1.23it/s]


test:  40%|████      | 6/15 [00:05<00:07,  1.26it/s]


test:  47%|████▋     | 7/15 [00:05<00:06,  1.28it/s]


test:  53%|█████▎    | 8/15 [00:06<00:05,  1.29it/s]


test:  60%|██████    | 9/15 [00:07<00:04,  1.31it/s]


test:  67%|██████▋   | 10/15 [00:08<00:03,  1.31it/s]


test:  73%|███████▎  | 11/15 [00:08<00:03,  1.31it/s]


test:  80%|████████  | 12/15 [00:09<00:02,  1.31it/s]


test:  87%|████████▋ | 13/15 [00:10<00:01,  1.31it/s]


test:  93%|█████████▎| 14/15 [00:11<00:00,  1.31it/s]
test: 100%|██████████| 15/15 [00:11<00:00,  1.32it/s]
test loss: 1.2606076002120972
test acc: 0.6640569567680359

overall:  50%|█████     | 2000/4000 [27:00<2:26:11,  4.39s/it]

train:   5%|▌         | 2/37 [00:13<04:34,  7.84s/it]
overall:  50%|█████     | 2001/4000 [27:01<1:52:54,  3.39s/it]

train:   8%|▊         | 3/37 [00:14<02:41,  4.74s/it]
overall:  50%|█████     | 2002/4000 [27:02<1:26:50,  2.61s/it]

train:  11%|█         | 4/37 [00:15<01:44,  3.18s/it]
overall:  50%|█████     | 2003/4000 [27:03<1:08:36,  2.06s/it]

train:  14%|█▎        | 5/37 [00:16<01:14,  2.32s/it]
overall:  50%|█████     | 2004/4000 [27:04<55:48,  1.68s/it]  

train:  16%|█▌        | 6/37 [00:17<00:55,  1.80s/it]
overall:  50%|█████     | 2005/4000 [27:04<46:55,  1.41s/it]

train:  19%|█▉        | 7/37 [00:17<00:44,  1.47s/it]
overall:  50%|█████     | 2006/4000 [27:05<40:42,  1.22s/it]

train:  22%|██▏       | 8/37 [00:18<00:36,  1.25s/it]
overall:  50%|█████     | 2007/4000 [27:06<36:22,  1.10s/it]

train:  24%|██▍       | 9/37 [00:19<00:31,  1.11s/it]
overall:  50%|█████     | 2008/4000 [27:07<33:26,  1.01s/it]

train:  27%|██▋       | 10/37 [00:20<00:27,  1.01s/it]
overall:  50%|█████     | 2009/4000 [27:08<31:18,  1.06it/s]

train:  30%|██▉       | 11/37 [00:21<00:24,  1.06it/s]train loss: 0.9006773829460144
train acc: 0.7639999985694885

overall:  50%|█████     | 2010/4000 [27:08<29:44,  1.12it/s]

train:  32%|███▏      | 12/37 [00:21<00:22,  1.11it/s]
overall:  50%|█████     | 2011/4000 [27:09<28:45,  1.15it/s]

train:  35%|███▌      | 13/37 [00:22<00:20,  1.15it/s]
overall:  50%|█████     | 2012/4000 [27:10<27:54,  1.19it/s]

train:  38%|███▊      | 14/37 [00:23<00:19,  1.19it/s]
overall:  50%|█████     | 2013/4000 [27:11<27:18,  1.21it/s]

train:  41%|████      | 15/37 [00:24<00:18,  1.21it/s]
overall:  50%|█████     | 2014/4000 [27:12<26:52,  1.23it/s]

train:  43%|████▎     | 16/37 [00:25<00:17,  1.23it/s]
overall:  50%|█████     | 2015/4000 [27:12<26:36,  1.24it/s]

train:  46%|████▌     | 17/37 [00:25<00:16,  1.24it/s]
overall:  50%|█████     | 2016/4000 [27:13<26:24,  1.25it/s]

train:  49%|████▊     | 18/37 [00:26<00:15,  1.25it/s]
overall:  50%|█████     | 2017/4000 [27:14<26:17,  1.26it/s]

train:  51%|█████▏    | 19/37 [00:27<00:14,  1.26it/s]
overall:  50%|█████     | 2018/4000 [27:15<26:14,  1.26it/s]

train:  54%|█████▍    | 20/37 [00:28<00:13,  1.26it/s]
overall:  50%|█████     | 2019/4000 [27:16<26:14,  1.26it/s]

train:  57%|█████▋    | 21/37 [00:29<00:12,  1.26it/s]train loss: 0.9287691116333008
train acc: 0.7440000176429749

overall:  50%|█████     | 2020/4000 [27:16<26:09,  1.26it/s]

train:  59%|█████▉    | 22/37 [00:29<00:11,  1.26it/s]
overall:  51%|█████     | 2021/4000 [27:17<26:08,  1.26it/s]

train:  62%|██████▏   | 23/37 [00:30<00:11,  1.26it/s]
overall:  51%|█████     | 2022/4000 [27:18<26:10,  1.26it/s]

train:  65%|██████▍   | 24/37 [00:31<00:10,  1.26it/s]
overall:  51%|█████     | 2023/4000 [27:19<26:14,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:32<00:09,  1.26it/s]
overall:  51%|█████     | 2024/4000 [27:19<26:10,  1.26it/s]

train:  70%|███████   | 26/37 [00:32<00:08,  1.26it/s]
overall:  51%|█████     | 2025/4000 [27:20<26:07,  1.26it/s]

train:  73%|███████▎  | 27/37 [00:33<00:07,  1.26it/s]
overall:  51%|█████     | 2026/4000 [27:21<26:07,  1.26it/s]

train:  76%|███████▌  | 28/37 [00:34<00:07,  1.26it/s]
overall:  51%|█████     | 2027/4000 [27:22<26:01,  1.26it/s]

train:  78%|███████▊  | 29/37 [00:35<00:06,  1.26it/s]
overall:  51%|█████     | 2028/4000 [27:23<26:25,  1.24it/s]

train:  81%|████████  | 30/37 [00:36<00:05,  1.24it/s]
overall:  51%|█████     | 2029/4000 [27:23<26:14,  1.25it/s]

train:  84%|████████▍ | 31/37 [00:36<00:04,  1.25it/s]train loss: 0.9561229944229126
train acc: 0.7139999866485596

overall:  51%|█████     | 2030/4000 [27:24<26:16,  1.25it/s]

train:  86%|████████▋ | 32/37 [00:37<00:04,  1.25it/s]
overall:  51%|█████     | 2031/4000 [27:25<26:08,  1.26it/s]

train:  89%|████████▉ | 33/37 [00:38<00:03,  1.26it/s]
overall:  51%|█████     | 2032/4000 [27:26<26:01,  1.26it/s]

train:  92%|█████████▏| 34/37 [00:39<00:02,  1.26it/s]
overall:  51%|█████     | 2033/4000 [27:27<25:53,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:40<00:01,  1.27it/s]
overall:  51%|█████     | 2034/4000 [27:27<25:49,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:40<00:00,  1.27it/s]
overall:  51%|█████     | 2035/4000 [27:28<24:54,  1.31it/s]

train: 100%|██████████| 37/37 [00:41<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:41<00:00,  1.13s/it]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  51%|█████     | 2036/4000 [27:30<32:04,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:50,  1.40s/it]
overall:  51%|█████     | 2037/4000 [27:30<30:10,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.04s/it]
overall:  51%|█████     | 2038/4000 [27:31<28:53,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  51%|█████     | 2039/4000 [27:32<27:53,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]train loss: 0.8628915548324585
train acc: 0.7823886871337891

overall:  51%|█████     | 2040/4000 [27:33<27:10,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  51%|█████     | 2041/4000 [27:34<26:47,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  51%|█████     | 2042/4000 [27:34<26:25,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  51%|█████     | 2043/4000 [27:35<26:10,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  51%|█████     | 2044/4000 [27:36<26:02,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  51%|█████     | 2045/4000 [27:37<25:57,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  51%|█████     | 2046/4000 [27:37<25:53,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  51%|█████     | 2047/4000 [27:38<25:43,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  51%|█████     | 2048/4000 [27:39<25:46,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  51%|█████     | 2049/4000 [27:40<25:46,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]train loss: 0.9114216566085815
train acc: 0.7490000128746033

overall:  51%|█████▏    | 2050/4000 [27:41<25:44,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  51%|█████▏    | 2051/4000 [27:41<25:37,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  51%|█████▏    | 2052/4000 [27:42<25:35,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  51%|█████▏    | 2053/4000 [27:43<25:35,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  51%|█████▏    | 2054/4000 [27:44<25:31,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  51%|█████▏    | 2055/4000 [27:45<26:04,  1.24it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.24it/s]
overall:  51%|█████▏    | 2056/4000 [27:45<25:55,  1.25it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.25it/s]
overall:  51%|█████▏    | 2057/4000 [27:46<26:23,  1.23it/s]

train:  59%|█████▉    | 22/37 [00:18<00:12,  1.23it/s]
overall:  51%|█████▏    | 2058/4000 [27:47<26:08,  1.24it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.24it/s]
overall:  51%|█████▏    | 2059/4000 [27:48<25:57,  1.25it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.25it/s]train loss: 0.8918130993843079
train acc: 0.7559999823570251

overall:  52%|█████▏    | 2060/4000 [27:49<26:06,  1.24it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.24it/s]
overall:  52%|█████▏    | 2061/4000 [27:49<26:00,  1.24it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.24it/s]
overall:  52%|█████▏    | 2062/4000 [27:50<25:51,  1.25it/s]

train:  73%|███████▎  | 27/37 [00:22<00:08,  1.25it/s]
overall:  52%|█████▏    | 2063/4000 [27:51<25:45,  1.25it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.25it/s]
overall:  52%|█████▏    | 2064/4000 [27:52<25:38,  1.26it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.26it/s]
overall:  52%|█████▏    | 2065/4000 [27:53<25:33,  1.26it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.26it/s]
overall:  52%|█████▏    | 2066/4000 [27:53<25:32,  1.26it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.26it/s]
overall:  52%|█████▏    | 2067/4000 [27:54<25:26,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  52%|█████▏    | 2068/4000 [27:55<25:29,  1.26it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.26it/s]
overall:  52%|█████▏    | 2069/4000 [27:56<25:25,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]train loss: 0.9065659642219543
train acc: 0.7459999918937683

overall:  52%|█████▏    | 2070/4000 [27:57<25:25,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  52%|█████▏    | 2071/4000 [27:57<25:19,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  52%|█████▏    | 2072/4000 [27:58<24:23,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  52%|█████▏    | 2073/4000 [28:00<31:16,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.38s/it]
overall:  52%|█████▏    | 2074/4000 [28:00<29:27,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  52%|█████▏    | 2075/4000 [28:01<28:11,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  52%|█████▏    | 2076/4000 [28:02<27:17,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  52%|█████▏    | 2077/4000 [28:03<26:40,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  52%|█████▏    | 2078/4000 [28:03<26:11,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  52%|█████▏    | 2079/4000 [28:04<25:52,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.24it/s]train loss: 0.8764206767082214
train acc: 0.7753036618232727

overall:  52%|█████▏    | 2080/4000 [28:05<25:46,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  52%|█████▏    | 2081/4000 [28:06<25:35,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  52%|█████▏    | 2082/4000 [28:07<25:26,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  52%|█████▏    | 2083/4000 [28:07<25:24,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  52%|█████▏    | 2084/4000 [28:08<25:21,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  52%|█████▏    | 2085/4000 [28:09<25:23,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  52%|█████▏    | 2086/4000 [28:10<25:19,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  52%|█████▏    | 2087/4000 [28:11<25:13,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  52%|█████▏    | 2088/4000 [28:11<25:07,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  52%|█████▏    | 2089/4000 [28:12<25:05,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]train loss: 0.9240066409111023
train acc: 0.7400000095367432

overall:  52%|█████▏    | 2090/4000 [28:13<25:07,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  52%|█████▏    | 2091/4000 [28:14<25:07,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  52%|█████▏    | 2092/4000 [28:15<25:12,  1.26it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.26it/s]
overall:  52%|█████▏    | 2093/4000 [28:15<25:07,  1.26it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.26it/s]
overall:  52%|█████▏    | 2094/4000 [28:16<25:01,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  52%|█████▏    | 2095/4000 [28:17<24:59,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  52%|█████▏    | 2096/4000 [28:18<25:01,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  52%|█████▏    | 2097/4000 [28:18<25:01,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  52%|█████▏    | 2098/4000 [28:19<25:01,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  52%|█████▏    | 2099/4000 [28:20<24:56,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]train loss: 0.9136112928390503
train acc: 0.7570000290870667

overall:  52%|█████▎    | 2100/4000 [28:21<24:54,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  53%|█████▎    | 2101/4000 [28:22<24:52,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  53%|█████▎    | 2102/4000 [28:22<24:50,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  53%|█████▎    | 2103/4000 [28:23<24:48,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  53%|█████▎    | 2104/4000 [28:24<24:55,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  53%|█████▎    | 2105/4000 [28:25<24:52,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  53%|█████▎    | 2106/4000 [28:26<24:48,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  53%|█████▎    | 2107/4000 [28:26<24:49,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  53%|█████▎    | 2108/4000 [28:27<24:48,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  53%|█████▎    | 2109/4000 [28:28<23:53,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]train loss: 0.9252358675003052
train acc: 0.7297570705413818

overall:  53%|█████▎    | 2110/4000 [28:29<30:32,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  53%|█████▎    | 2111/4000 [28:30<28:45,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.02s/it]
overall:  53%|█████▎    | 2112/4000 [28:31<27:31,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  53%|█████▎    | 2113/4000 [28:32<26:39,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]
overall:  53%|█████▎    | 2114/4000 [28:32<26:05,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.20it/s]
overall:  53%|█████▎    | 2115/4000 [28:33<25:40,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  53%|█████▎    | 2116/4000 [28:34<25:26,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  53%|█████▎    | 2117/4000 [28:35<25:16,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  53%|█████▎    | 2118/4000 [28:36<25:04,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  53%|█████▎    | 2119/4000 [28:36<24:55,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]train loss: 0.866409957408905
train acc: 0.7609999775886536

overall:  53%|█████▎    | 2120/4000 [28:37<24:53,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  53%|█████▎    | 2121/4000 [28:38<24:45,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  53%|█████▎    | 2122/4000 [28:39<24:47,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  53%|█████▎    | 2123/4000 [28:40<24:45,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  53%|█████▎    | 2124/4000 [28:40<24:42,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  53%|█████▎    | 2125/4000 [28:41<24:39,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  53%|█████▎    | 2126/4000 [28:42<24:32,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  53%|█████▎    | 2127/4000 [28:43<24:29,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  53%|█████▎    | 2128/4000 [28:43<24:28,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  53%|█████▎    | 2129/4000 [28:44<24:34,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]train loss: 0.9034233093261719
train acc: 0.7549999952316284

overall:  53%|█████▎    | 2130/4000 [28:45<24:28,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  53%|█████▎    | 2131/4000 [28:46<24:26,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  53%|█████▎    | 2132/4000 [28:47<24:28,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  53%|█████▎    | 2133/4000 [28:47<24:26,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  53%|█████▎    | 2134/4000 [28:48<24:28,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  53%|█████▎    | 2135/4000 [28:49<24:28,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  53%|█████▎    | 2136/4000 [28:50<24:25,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  53%|█████▎    | 2137/4000 [28:51<24:20,  1.28it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.28it/s]
overall:  53%|█████▎    | 2138/4000 [28:51<24:18,  1.28it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.28it/s]
overall:  53%|█████▎    | 2139/4000 [28:52<24:16,  1.28it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.28it/s]train loss: 0.9332348108291626
train acc: 0.7300000190734863

overall:  54%|█████▎    | 2140/4000 [28:53<24:21,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.27it/s]
overall:  54%|█████▎    | 2141/4000 [28:54<24:21,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  54%|█████▎    | 2142/4000 [28:54<24:21,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  54%|█████▎    | 2143/4000 [28:55<24:19,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  54%|█████▎    | 2144/4000 [28:56<24:16,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  54%|█████▎    | 2145/4000 [28:57<24:16,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  54%|█████▎    | 2146/4000 [28:57<23:27,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  54%|█████▎    | 2147/4000 [28:59<30:35,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  54%|█████▎    | 2148/4000 [29:00<28:40,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  54%|█████▎    | 2149/4000 [29:01<27:19,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]train loss: 0.8862897157669067
train acc: 0.7540485858917236

overall:  54%|█████▍    | 2150/4000 [29:01<26:26,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  54%|█████▍    | 2151/4000 [29:02<25:48,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  54%|█████▍    | 2152/4000 [29:03<25:20,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  54%|█████▍    | 2153/4000 [29:04<25:01,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  54%|█████▍    | 2154/4000 [29:05<24:49,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  54%|█████▍    | 2155/4000 [29:05<24:42,  1.24it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.24it/s]
overall:  54%|█████▍    | 2156/4000 [29:06<24:44,  1.24it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.24it/s]
overall:  54%|█████▍    | 2157/4000 [29:07<24:34,  1.25it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.25it/s]
overall:  54%|█████▍    | 2158/4000 [29:08<24:26,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  54%|█████▍    | 2159/4000 [29:09<24:25,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]train loss: 0.8467034101486206
train acc: 0.7739999890327454

overall:  54%|█████▍    | 2160/4000 [29:09<24:20,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  54%|█████▍    | 2161/4000 [29:10<24:17,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  54%|█████▍    | 2162/4000 [29:11<24:10,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  54%|█████▍    | 2163/4000 [29:12<24:05,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  54%|█████▍    | 2164/4000 [29:12<24:02,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  54%|█████▍    | 2165/4000 [29:13<24:02,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  54%|█████▍    | 2166/4000 [29:14<24:03,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  54%|█████▍    | 2167/4000 [29:15<24:04,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  54%|█████▍    | 2168/4000 [29:16<24:03,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  54%|█████▍    | 2169/4000 [29:16<23:59,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]train loss: 0.8936862945556641
train acc: 0.7480000257492065

overall:  54%|█████▍    | 2170/4000 [29:17<23:59,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  54%|█████▍    | 2171/4000 [29:18<24:02,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  54%|█████▍    | 2172/4000 [29:19<24:02,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  54%|█████▍    | 2173/4000 [29:20<24:00,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  54%|█████▍    | 2174/4000 [29:20<23:55,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  54%|█████▍    | 2175/4000 [29:21<23:53,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  54%|█████▍    | 2176/4000 [29:22<23:50,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  54%|█████▍    | 2177/4000 [29:23<23:52,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  54%|█████▍    | 2178/4000 [29:23<23:50,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  54%|█████▍    | 2179/4000 [29:24<23:53,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]train loss: 0.9337930679321289
train acc: 0.7360000014305115

overall:  55%|█████▍    | 2180/4000 [29:25<23:53,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  55%|█████▍    | 2181/4000 [29:26<23:49,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  55%|█████▍    | 2182/4000 [29:27<23:53,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  55%|█████▍    | 2183/4000 [29:27<23:02,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  55%|█████▍    | 2184/4000 [29:29<29:43,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:50,  1.41s/it]
overall:  55%|█████▍    | 2185/4000 [29:30<27:54,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.04s/it]
overall:  55%|█████▍    | 2186/4000 [29:30<26:37,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  55%|█████▍    | 2187/4000 [29:31<25:45,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  55%|█████▍    | 2188/4000 [29:32<25:19,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  55%|█████▍    | 2189/4000 [29:33<24:50,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]train loss: 0.9131261110305786
train acc: 0.7489878535270691

overall:  55%|█████▍    | 2190/4000 [29:34<24:28,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  55%|█████▍    | 2191/4000 [29:34<24:17,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  55%|█████▍    | 2192/4000 [29:35<24:03,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  55%|█████▍    | 2193/4000 [29:36<23:57,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  55%|█████▍    | 2194/4000 [29:37<23:52,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  55%|█████▍    | 2195/4000 [29:37<23:48,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  55%|█████▍    | 2196/4000 [29:38<23:50,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  55%|█████▍    | 2197/4000 [29:39<23:45,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  55%|█████▍    | 2198/4000 [29:40<23:44,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  55%|█████▍    | 2199/4000 [29:41<23:38,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]train loss: 0.8972324132919312
train acc: 0.7620000243186951

overall:  55%|█████▌    | 2200/4000 [29:41<23:45,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  55%|█████▌    | 2201/4000 [29:42<23:41,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.27it/s]
overall:  55%|█████▌    | 2202/4000 [29:43<23:40,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  55%|█████▌    | 2203/4000 [29:44<23:35,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  55%|█████▌    | 2204/4000 [29:45<23:33,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  55%|█████▌    | 2205/4000 [29:45<23:31,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  55%|█████▌    | 2206/4000 [29:46<23:31,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  55%|█████▌    | 2207/4000 [29:47<23:27,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  55%|█████▌    | 2208/4000 [29:48<23:33,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  55%|█████▌    | 2209/4000 [29:49<23:29,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]train loss: 0.8361933827400208
train acc: 0.7910000085830688

overall:  55%|█████▌    | 2210/4000 [29:49<23:26,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  55%|█████▌    | 2211/4000 [29:50<23:26,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  55%|█████▌    | 2212/4000 [29:51<23:35,  1.26it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.26it/s]
overall:  55%|█████▌    | 2213/4000 [29:52<23:31,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  55%|█████▌    | 2214/4000 [29:52<23:29,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  55%|█████▌    | 2215/4000 [29:53<23:28,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  55%|█████▌    | 2216/4000 [29:54<23:23,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  55%|█████▌    | 2217/4000 [29:55<23:20,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  55%|█████▌    | 2218/4000 [29:56<23:19,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  55%|█████▌    | 2219/4000 [29:56<23:20,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]train loss: 0.936060905456543
train acc: 0.7246963381767273

overall:  56%|█████▌    | 2220/4000 [29:57<22:31,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  56%|█████▌    | 2221/4000 [29:59<28:45,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  56%|█████▌    | 2222/4000 [29:59<27:13,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  56%|█████▌    | 2223/4000 [30:00<26:00,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  56%|█████▌    | 2224/4000 [30:01<25:07,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]
overall:  56%|█████▌    | 2225/4000 [30:02<24:31,  1.21it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.20it/s]
overall:  56%|█████▌    | 2226/4000 [30:02<24:11,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  56%|█████▌    | 2227/4000 [30:03<23:52,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.24it/s]
overall:  56%|█████▌    | 2228/4000 [30:04<23:42,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  56%|█████▌    | 2229/4000 [30:05<23:35,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]train loss: 0.8752186894416809
train acc: 0.7450000047683716

overall:  56%|█████▌    | 2230/4000 [30:06<23:30,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  56%|█████▌    | 2231/4000 [30:06<23:23,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  56%|█████▌    | 2232/4000 [30:07<23:17,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  56%|█████▌    | 2233/4000 [30:08<23:19,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  56%|█████▌    | 2234/4000 [30:09<23:21,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  56%|█████▌    | 2235/4000 [30:10<23:15,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  56%|█████▌    | 2236/4000 [30:10<23:12,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  56%|█████▌    | 2237/4000 [30:11<23:11,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  56%|█████▌    | 2238/4000 [30:12<23:07,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  56%|█████▌    | 2239/4000 [30:13<23:06,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]train loss: 0.9029241800308228
train acc: 0.7540000081062317

overall:  56%|█████▌    | 2240/4000 [30:14<23:07,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  56%|█████▌    | 2241/4000 [30:14<23:10,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  56%|█████▌    | 2242/4000 [30:15<23:07,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  56%|█████▌    | 2243/4000 [30:16<23:01,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  56%|█████▌    | 2244/4000 [30:17<22:59,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  56%|█████▌    | 2245/4000 [30:18<24:07,  1.21it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.21it/s]
overall:  56%|█████▌    | 2246/4000 [30:18<23:51,  1.23it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.23it/s]
overall:  56%|█████▌    | 2247/4000 [30:19<23:33,  1.24it/s]

train:  73%|███████▎  | 27/37 [00:21<00:08,  1.24it/s]
overall:  56%|█████▌    | 2248/4000 [30:20<23:19,  1.25it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.25it/s]
overall:  56%|█████▌    | 2249/4000 [30:21<23:10,  1.26it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.26it/s]train loss: 0.9049564599990845
train acc: 0.7639999985694885

overall:  56%|█████▋    | 2250/4000 [30:22<23:06,  1.26it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.26it/s]
overall:  56%|█████▋    | 2251/4000 [30:22<23:02,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  56%|█████▋    | 2252/4000 [30:23<22:58,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  56%|█████▋    | 2253/4000 [30:24<23:02,  1.26it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.26it/s]
overall:  56%|█████▋    | 2254/4000 [30:25<22:55,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  56%|█████▋    | 2255/4000 [30:25<22:54,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  56%|█████▋    | 2256/4000 [30:26<22:52,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  56%|█████▋    | 2257/4000 [30:27<22:04,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  56%|█████▋    | 2258/4000 [30:28<28:05,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.36s/it]
overall:  56%|█████▋    | 2259/4000 [30:29<26:37,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]train loss: 0.8760193586349487
train acc: 0.7560728788375854

overall:  56%|█████▋    | 2260/4000 [30:30<25:32,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  57%|█████▋    | 2261/4000 [30:31<24:39,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  57%|█████▋    | 2262/4000 [30:32<24:05,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  57%|█████▋    | 2263/4000 [30:32<23:41,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  57%|█████▋    | 2264/4000 [30:33<23:24,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  57%|█████▋    | 2265/4000 [30:34<23:13,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  57%|█████▋    | 2266/4000 [30:35<23:07,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  57%|█████▋    | 2267/4000 [30:35<23:00,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  57%|█████▋    | 2268/4000 [30:36<22:52,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  57%|█████▋    | 2269/4000 [30:37<22:49,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]train loss: 0.8591569066047668
train acc: 0.765999972820282

overall:  57%|█████▋    | 2270/4000 [30:38<22:54,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  57%|█████▋    | 2271/4000 [30:39<22:51,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  57%|█████▋    | 2272/4000 [30:39<22:45,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  57%|█████▋    | 2273/4000 [30:40<22:40,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  57%|█████▋    | 2274/4000 [30:41<22:39,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  57%|█████▋    | 2275/4000 [30:42<22:38,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  57%|█████▋    | 2276/4000 [30:43<22:34,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  57%|█████▋    | 2277/4000 [30:43<22:32,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  57%|█████▋    | 2278/4000 [30:44<22:35,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  57%|█████▋    | 2279/4000 [30:45<22:30,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]train loss: 0.8902719616889954
train acc: 0.7599999904632568

overall:  57%|█████▋    | 2280/4000 [30:46<22:29,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  57%|█████▋    | 2281/4000 [30:46<22:30,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  57%|█████▋    | 2282/4000 [30:47<22:37,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  57%|█████▋    | 2283/4000 [30:48<22:35,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  57%|█████▋    | 2284/4000 [30:49<22:30,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  57%|█████▋    | 2285/4000 [30:50<22:27,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  57%|█████▋    | 2286/4000 [30:50<22:26,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  57%|█████▋    | 2287/4000 [30:51<22:26,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  57%|█████▋    | 2288/4000 [30:52<22:23,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.27it/s]
overall:  57%|█████▋    | 2289/4000 [30:53<22:21,  1.28it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.28it/s]train loss: 0.8900264501571655
train acc: 0.7549999952316284

overall:  57%|█████▋    | 2290/4000 [30:54<22:26,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  57%|█████▋    | 2291/4000 [30:54<22:23,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  57%|█████▋    | 2292/4000 [30:55<22:20,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  57%|█████▋    | 2293/4000 [30:56<22:21,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  57%|█████▋    | 2294/4000 [30:57<21:33,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  57%|█████▋    | 2295/4000 [30:58<27:29,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:48,  1.36s/it]
overall:  57%|█████▋    | 2296/4000 [30:59<25:58,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.02s/it]
overall:  57%|█████▋    | 2297/4000 [31:00<24:56,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  57%|█████▋    | 2298/4000 [31:00<24:10,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  57%|█████▋    | 2299/4000 [31:01<23:32,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.20it/s]train loss: 0.8476178050041199
train acc: 0.7763158082962036

overall:  57%|█████▊    | 2300/4000 [31:02<23:15,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  58%|█████▊    | 2301/4000 [31:03<23:02,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  58%|█████▊    | 2302/4000 [31:04<22:46,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  58%|█████▊    | 2303/4000 [31:04<22:34,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  58%|█████▊    | 2304/4000 [31:05<22:28,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  58%|█████▊    | 2305/4000 [31:06<22:24,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  58%|█████▊    | 2306/4000 [31:07<22:24,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  58%|█████▊    | 2307/4000 [31:08<22:21,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  58%|█████▊    | 2308/4000 [31:08<22:19,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  58%|█████▊    | 2309/4000 [31:09<22:20,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]train loss: 0.9308246374130249
train acc: 0.7409999966621399

overall:  58%|█████▊    | 2310/4000 [31:10<22:18,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  58%|█████▊    | 2311/4000 [31:11<22:12,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  58%|█████▊    | 2312/4000 [31:12<22:14,  1.26it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.26it/s]
overall:  58%|█████▊    | 2313/4000 [31:12<22:12,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  58%|█████▊    | 2314/4000 [31:13<22:10,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  58%|█████▊    | 2315/4000 [31:14<22:09,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  58%|█████▊    | 2316/4000 [31:15<22:08,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  58%|█████▊    | 2317/4000 [31:15<22:07,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  58%|█████▊    | 2318/4000 [31:16<22:07,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  58%|█████▊    | 2319/4000 [31:17<22:06,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]train loss: 0.8835646510124207
train acc: 0.7559999823570251

overall:  58%|█████▊    | 2320/4000 [31:18<22:05,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  58%|█████▊    | 2321/4000 [31:19<22:08,  1.26it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.26it/s]
overall:  58%|█████▊    | 2322/4000 [31:19<22:08,  1.26it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.26it/s]
overall:  58%|█████▊    | 2323/4000 [31:20<22:03,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  58%|█████▊    | 2324/4000 [31:21<22:07,  1.26it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.26it/s]
overall:  58%|█████▊    | 2325/4000 [31:22<22:08,  1.26it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.26it/s]
overall:  58%|█████▊    | 2326/4000 [31:23<22:02,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  58%|█████▊    | 2327/4000 [31:23<22:00,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  58%|█████▊    | 2328/4000 [31:24<21:59,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  58%|█████▊    | 2329/4000 [31:25<21:54,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]train loss: 0.8634689450263977
train acc: 0.7590000033378601

overall:  58%|█████▊    | 2330/4000 [31:26<21:58,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  58%|█████▊    | 2331/4000 [31:26<21:09,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  58%|█████▊    | 2332/4000 [31:28<27:37,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  58%|█████▊    | 2333/4000 [31:29<25:56,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:  58%|█████▊    | 2334/4000 [31:30<24:45,  1.12it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  58%|█████▊    | 2335/4000 [31:30<23:50,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  58%|█████▊    | 2336/4000 [31:31<23:12,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  58%|█████▊    | 2337/4000 [31:32<22:52,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  58%|█████▊    | 2338/4000 [31:33<22:36,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.22it/s]
overall:  58%|█████▊    | 2339/4000 [31:33<22:19,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]train loss: 0.8644779920578003
train acc: 0.7702429294586182

overall:  58%|█████▊    | 2340/4000 [31:34<22:12,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  59%|█████▊    | 2341/4000 [31:35<22:01,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  59%|█████▊    | 2342/4000 [31:36<21:56,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  59%|█████▊    | 2343/4000 [31:37<21:54,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  59%|█████▊    | 2344/4000 [31:37<21:50,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.26it/s]
overall:  59%|█████▊    | 2345/4000 [31:38<21:51,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  59%|█████▊    | 2346/4000 [31:39<21:45,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  59%|█████▊    | 2347/4000 [31:40<21:47,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  59%|█████▊    | 2348/4000 [31:41<21:41,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  59%|█████▊    | 2349/4000 [31:41<21:46,  1.26it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.26it/s]train loss: 0.8845608830451965
train acc: 0.7419999837875366

overall:  59%|█████▉    | 2350/4000 [31:42<21:43,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  59%|█████▉    | 2351/4000 [31:43<21:42,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  59%|█████▉    | 2352/4000 [31:44<21:39,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  59%|█████▉    | 2353/4000 [31:45<21:35,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  59%|█████▉    | 2354/4000 [31:45<21:37,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  59%|█████▉    | 2355/4000 [31:46<21:33,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  59%|█████▉    | 2356/4000 [31:47<21:32,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  59%|█████▉    | 2357/4000 [31:48<21:34,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  59%|█████▉    | 2358/4000 [31:48<21:31,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  59%|█████▉    | 2359/4000 [31:49<21:32,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]train loss: 0.8735299110412598
train acc: 0.7649999856948853

overall:  59%|█████▉    | 2360/4000 [31:50<21:32,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  59%|█████▉    | 2361/4000 [31:51<21:34,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  59%|█████▉    | 2362/4000 [31:52<21:32,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  59%|█████▉    | 2363/4000 [31:52<21:28,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  59%|█████▉    | 2364/4000 [31:53<21:26,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  59%|█████▉    | 2365/4000 [31:54<21:26,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  59%|█████▉    | 2366/4000 [31:55<21:23,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  59%|█████▉    | 2367/4000 [31:56<21:21,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  59%|█████▉    | 2368/4000 [31:56<20:33,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  59%|█████▉    | 2369/4000 [31:58<26:44,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:50,  1.41s/it]train loss: 0.8780620694160461
train acc: 0.7651821970939636

overall:  59%|█████▉    | 2370/4000 [31:59<25:10,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  59%|█████▉    | 2371/4000 [31:59<23:59,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  59%|█████▉    | 2372/4000 [32:00<23:14,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  59%|█████▉    | 2373/4000 [32:01<22:35,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  59%|█████▉    | 2374/4000 [32:02<22:17,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  59%|█████▉    | 2375/4000 [32:02<21:58,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  59%|█████▉    | 2376/4000 [32:03<21:47,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  59%|█████▉    | 2377/4000 [32:04<21:37,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  59%|█████▉    | 2378/4000 [32:05<21:36,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  59%|█████▉    | 2379/4000 [32:06<21:29,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]train loss: 0.8376971483230591
train acc: 0.7699999809265137

overall:  60%|█████▉    | 2380/4000 [32:06<21:28,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  60%|█████▉    | 2381/4000 [32:07<21:26,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  60%|█████▉    | 2382/4000 [32:08<21:23,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  60%|█████▉    | 2383/4000 [32:09<21:20,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  60%|█████▉    | 2384/4000 [32:10<21:16,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  60%|█████▉    | 2385/4000 [32:10<21:14,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  60%|█████▉    | 2386/4000 [32:11<21:14,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.27it/s]
overall:  60%|█████▉    | 2387/4000 [32:12<21:10,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  60%|█████▉    | 2388/4000 [32:13<21:07,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  60%|█████▉    | 2389/4000 [32:14<21:09,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]train loss: 0.8823555707931519
train acc: 0.7599999904632568

overall:  60%|█████▉    | 2390/4000 [32:14<21:07,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  60%|█████▉    | 2391/4000 [32:15<21:07,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  60%|█████▉    | 2392/4000 [32:16<21:06,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  60%|█████▉    | 2393/4000 [32:17<21:08,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  60%|█████▉    | 2394/4000 [32:17<21:05,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  60%|█████▉    | 2395/4000 [32:18<21:03,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  60%|█████▉    | 2396/4000 [32:19<21:06,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  60%|█████▉    | 2397/4000 [32:20<21:03,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  60%|█████▉    | 2398/4000 [32:21<21:06,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  60%|█████▉    | 2399/4000 [32:21<21:03,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]train loss: 0.8779155015945435
train acc: 0.7540000081062317

overall:  60%|██████    | 2400/4000 [32:22<21:04,  1.26it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.26it/s]
overall:  60%|██████    | 2401/4000 [32:23<21:05,  1.26it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.26it/s]
overall:  60%|██████    | 2402/4000 [32:24<21:03,  1.26it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.26it/s]
overall:  60%|██████    | 2403/4000 [32:25<21:00,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  60%|██████    | 2404/4000 [32:25<21:01,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  60%|██████    | 2405/4000 [32:26<20:11,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  60%|██████    | 2406/4000 [32:28<25:42,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.36s/it]
overall:  60%|██████    | 2407/4000 [32:28<24:16,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.03s/it]
overall:  60%|██████    | 2408/4000 [32:29<23:13,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  60%|██████    | 2409/4000 [32:30<22:29,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]train loss: 0.8718082308769226
train acc: 0.76113361120224

overall:  60%|██████    | 2410/4000 [32:31<22:09,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  60%|██████    | 2411/4000 [32:31<21:45,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  60%|██████    | 2412/4000 [32:32<21:25,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  60%|██████    | 2413/4000 [32:33<21:14,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  60%|██████    | 2414/4000 [32:34<21:07,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  60%|██████    | 2415/4000 [32:35<20:59,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  60%|██████    | 2416/4000 [32:35<20:56,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  60%|██████    | 2417/4000 [32:36<20:51,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  60%|██████    | 2418/4000 [32:37<20:53,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  60%|██████    | 2419/4000 [32:38<20:51,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]train loss: 0.8618108034133911
train acc: 0.7620000243186951

overall:  60%|██████    | 2420/4000 [32:39<20:51,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  61%|██████    | 2421/4000 [32:39<20:49,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  61%|██████    | 2422/4000 [32:40<20:56,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  61%|██████    | 2423/4000 [32:41<20:49,  1.26it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.26it/s]
overall:  61%|██████    | 2424/4000 [32:42<20:47,  1.26it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.26it/s]
overall:  61%|██████    | 2425/4000 [32:43<20:44,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  61%|██████    | 2426/4000 [32:43<20:40,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  61%|██████    | 2427/4000 [32:44<20:38,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  61%|██████    | 2428/4000 [32:45<20:38,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  61%|██████    | 2429/4000 [32:46<20:34,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]train loss: 0.854019284248352
train acc: 0.7609999775886536

overall:  61%|██████    | 2430/4000 [32:46<20:38,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  61%|██████    | 2431/4000 [32:47<20:38,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  61%|██████    | 2432/4000 [32:48<20:39,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  61%|██████    | 2433/4000 [32:49<20:34,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  61%|██████    | 2434/4000 [32:50<20:38,  1.26it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.26it/s]
overall:  61%|██████    | 2435/4000 [32:50<20:39,  1.26it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.26it/s]
overall:  61%|██████    | 2436/4000 [32:51<20:35,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  61%|██████    | 2437/4000 [32:52<20:31,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  61%|██████    | 2438/4000 [32:53<20:32,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  61%|██████    | 2439/4000 [32:54<20:30,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]train loss: 0.8992632031440735
train acc: 0.7559999823570251

overall:  61%|██████    | 2440/4000 [32:54<20:30,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  61%|██████    | 2441/4000 [32:55<20:29,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  61%|██████    | 2442/4000 [32:56<19:43,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  61%|██████    | 2443/4000 [32:57<25:24,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.38s/it]
overall:  61%|██████    | 2444/4000 [32:58<23:58,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.04s/it]
overall:  61%|██████    | 2445/4000 [32:59<22:53,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  61%|██████    | 2446/4000 [33:00<22:07,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  61%|██████    | 2447/4000 [33:00<21:36,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  61%|██████    | 2448/4000 [33:01<21:14,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  61%|██████    | 2449/4000 [33:02<20:56,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]train loss: 0.8527871370315552
train acc: 0.7631579041481018

overall:  61%|██████▏   | 2450/4000 [33:03<20:47,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  61%|██████▏   | 2451/4000 [33:04<20:39,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  61%|██████▏   | 2452/4000 [33:04<20:36,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  61%|██████▏   | 2453/4000 [33:05<20:30,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  61%|██████▏   | 2454/4000 [33:06<20:27,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  61%|██████▏   | 2455/4000 [33:07<20:28,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  61%|██████▏   | 2456/4000 [33:08<20:27,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  61%|██████▏   | 2457/4000 [33:08<20:24,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  61%|██████▏   | 2458/4000 [33:09<20:17,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  61%|██████▏   | 2459/4000 [33:10<20:14,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]train loss: 0.8838412165641785
train acc: 0.7549999952316284

overall:  62%|██████▏   | 2460/4000 [33:11<20:14,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  62%|██████▏   | 2461/4000 [33:12<20:11,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  62%|██████▏   | 2462/4000 [33:12<20:11,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  62%|██████▏   | 2463/4000 [33:13<20:11,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  62%|██████▏   | 2464/4000 [33:14<20:10,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  62%|██████▏   | 2465/4000 [33:15<20:13,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  62%|██████▏   | 2466/4000 [33:15<20:12,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  62%|██████▏   | 2467/4000 [33:16<20:16,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  62%|██████▏   | 2468/4000 [33:17<20:13,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  62%|██████▏   | 2469/4000 [33:18<20:10,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]train loss: 0.8265846371650696
train acc: 0.7829999923706055

overall:  62%|██████▏   | 2470/4000 [33:19<20:09,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  62%|██████▏   | 2471/4000 [33:19<20:04,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  62%|██████▏   | 2472/4000 [33:20<20:03,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  62%|██████▏   | 2473/4000 [33:21<20:03,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  62%|██████▏   | 2474/4000 [33:22<20:01,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  62%|██████▏   | 2475/4000 [33:23<20:03,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  62%|██████▏   | 2476/4000 [33:23<19:59,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  62%|██████▏   | 2477/4000 [33:24<20:00,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  62%|██████▏   | 2478/4000 [33:25<19:58,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  62%|██████▏   | 2479/4000 [33:26<19:14,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]train loss: 0.9041731953620911
train acc: 0.7479757070541382

overall:  62%|██████▏   | 2480/4000 [33:27<25:00,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  62%|██████▏   | 2481/4000 [33:28<23:28,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  62%|██████▏   | 2482/4000 [33:29<22:22,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  62%|██████▏   | 2483/4000 [33:30<21:37,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  62%|██████▏   | 2484/4000 [33:30<21:05,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  62%|██████▏   | 2485/4000 [33:31<20:43,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  62%|██████▏   | 2486/4000 [33:32<20:28,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  62%|██████▏   | 2487/4000 [33:33<20:14,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  62%|██████▏   | 2488/4000 [33:33<20:09,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  62%|██████▏   | 2489/4000 [33:34<20:02,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]train loss: 0.8903888463973999
train acc: 0.7490000128746033

overall:  62%|██████▏   | 2490/4000 [33:35<19:59,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  62%|██████▏   | 2491/4000 [33:36<19:55,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  62%|██████▏   | 2492/4000 [33:37<19:57,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  62%|██████▏   | 2493/4000 [33:37<19:54,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  62%|██████▏   | 2494/4000 [33:38<19:50,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  62%|██████▏   | 2495/4000 [33:39<19:49,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  62%|██████▏   | 2496/4000 [33:40<19:46,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  62%|██████▏   | 2497/4000 [33:41<19:46,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  62%|██████▏   | 2498/4000 [33:41<19:44,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  62%|██████▏   | 2499/4000 [33:42<19:43,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]train loss: 0.8628465533256531
train acc: 0.7720000147819519

overall:  62%|██████▎   | 2500/4000 [33:43<19:46,  1.26it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.26it/s]
overall:  63%|██████▎   | 2501/4000 [33:44<19:43,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  63%|██████▎   | 2502/4000 [33:45<19:40,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  63%|██████▎   | 2503/4000 [33:45<19:41,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  63%|██████▎   | 2504/4000 [33:46<19:44,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  63%|██████▎   | 2505/4000 [33:47<19:40,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  63%|██████▎   | 2506/4000 [33:48<19:35,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  63%|██████▎   | 2507/4000 [33:48<19:33,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  63%|██████▎   | 2508/4000 [33:49<19:31,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  63%|██████▎   | 2509/4000 [33:50<19:32,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]train loss: 0.8709559440612793
train acc: 0.7599999904632568

overall:  63%|██████▎   | 2510/4000 [33:51<19:31,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  63%|██████▎   | 2511/4000 [33:52<19:28,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  63%|██████▎   | 2512/4000 [33:52<19:31,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  63%|██████▎   | 2513/4000 [33:53<19:27,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  63%|██████▎   | 2514/4000 [33:54<19:25,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.28it/s]
overall:  63%|██████▎   | 2515/4000 [33:55<19:26,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  63%|██████▎   | 2516/4000 [33:55<18:43,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  63%|██████▎   | 2517/4000 [33:57<23:56,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  63%|██████▎   | 2518/4000 [33:58<22:35,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.03s/it]
overall:  63%|██████▎   | 2519/4000 [33:58<21:39,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]train loss: 0.8249682188034058
train acc: 0.7773279547691345

overall:  63%|██████▎   | 2520/4000 [33:59<21:00,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  63%|██████▎   | 2521/4000 [34:00<20:29,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  63%|██████▎   | 2522/4000 [34:01<20:13,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  63%|██████▎   | 2523/4000 [34:02<20:01,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  63%|██████▎   | 2524/4000 [34:02<19:49,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  63%|██████▎   | 2525/4000 [34:03<19:38,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  63%|██████▎   | 2526/4000 [34:04<19:35,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  63%|██████▎   | 2527/4000 [34:05<19:29,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  63%|██████▎   | 2528/4000 [34:06<19:27,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  63%|██████▎   | 2529/4000 [34:06<19:23,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.26it/s]train loss: 0.8509810566902161
train acc: 0.777999997138977

overall:  63%|██████▎   | 2530/4000 [34:07<19:26,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  63%|██████▎   | 2531/4000 [34:08<19:22,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  63%|██████▎   | 2532/4000 [34:09<19:20,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  63%|██████▎   | 2533/4000 [34:10<19:15,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  63%|██████▎   | 2534/4000 [34:10<19:19,  1.26it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.26it/s]
overall:  63%|██████▎   | 2535/4000 [34:11<19:15,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  63%|██████▎   | 2536/4000 [34:12<19:12,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  63%|██████▎   | 2537/4000 [34:13<19:13,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  63%|██████▎   | 2538/4000 [34:13<19:09,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  63%|██████▎   | 2539/4000 [34:14<19:07,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]train loss: 0.8585516810417175
train acc: 0.7609999775886536

overall:  64%|██████▎   | 2540/4000 [34:15<19:07,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  64%|██████▎   | 2541/4000 [34:16<19:07,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  64%|██████▎   | 2542/4000 [34:17<19:08,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  64%|██████▎   | 2543/4000 [34:17<19:08,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  64%|██████▎   | 2544/4000 [34:18<19:04,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  64%|██████▎   | 2545/4000 [34:19<19:00,  1.28it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.28it/s]
overall:  64%|██████▎   | 2546/4000 [34:20<19:04,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  64%|██████▎   | 2547/4000 [34:21<19:03,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  64%|██████▎   | 2548/4000 [34:21<19:00,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  64%|██████▎   | 2549/4000 [34:22<18:57,  1.28it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.28it/s]train loss: 0.8716220855712891
train acc: 0.7519999742507935

overall:  64%|██████▍   | 2550/4000 [34:23<18:59,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  64%|██████▍   | 2551/4000 [34:24<18:57,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  64%|██████▍   | 2552/4000 [34:24<18:55,  1.28it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.28it/s]
overall:  64%|██████▍   | 2553/4000 [34:25<18:15,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  64%|██████▍   | 2554/4000 [34:27<23:53,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  64%|██████▍   | 2555/4000 [34:27<22:25,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:  64%|██████▍   | 2556/4000 [34:28<21:21,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  64%|██████▍   | 2557/4000 [34:29<20:38,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  64%|██████▍   | 2558/4000 [34:30<20:05,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  64%|██████▍   | 2559/4000 [34:31<19:46,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]train loss: 0.8575493097305298
train acc: 0.7672064900398254

overall:  64%|██████▍   | 2560/4000 [34:31<19:29,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  64%|██████▍   | 2561/4000 [34:32<19:20,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  64%|██████▍   | 2562/4000 [34:33<19:10,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  64%|██████▍   | 2563/4000 [34:34<19:07,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  64%|██████▍   | 2564/4000 [34:35<19:06,  1.25it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.25it/s]
overall:  64%|██████▍   | 2565/4000 [34:35<19:00,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  64%|██████▍   | 2566/4000 [34:36<19:00,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  64%|██████▍   | 2567/4000 [34:37<18:56,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  64%|██████▍   | 2568/4000 [34:38<18:54,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  64%|██████▍   | 2569/4000 [34:39<18:49,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]train loss: 0.854433536529541
train acc: 0.7590000033378601

overall:  64%|██████▍   | 2570/4000 [34:39<18:48,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  64%|██████▍   | 2571/4000 [34:40<18:45,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  64%|██████▍   | 2572/4000 [34:41<18:44,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  64%|██████▍   | 2573/4000 [34:42<18:43,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  64%|██████▍   | 2574/4000 [34:42<18:44,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  64%|██████▍   | 2575/4000 [34:43<18:46,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  64%|██████▍   | 2576/4000 [34:44<18:48,  1.26it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.26it/s]
overall:  64%|██████▍   | 2577/4000 [34:45<18:44,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  64%|██████▍   | 2578/4000 [34:46<18:46,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  64%|██████▍   | 2579/4000 [34:46<18:41,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]train loss: 0.8488684892654419
train acc: 0.7730000019073486

overall:  64%|██████▍   | 2580/4000 [34:47<18:39,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  65%|██████▍   | 2581/4000 [34:48<18:39,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  65%|██████▍   | 2582/4000 [34:49<18:36,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  65%|██████▍   | 2583/4000 [34:50<18:35,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  65%|██████▍   | 2584/4000 [34:50<18:36,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  65%|██████▍   | 2585/4000 [34:51<18:33,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  65%|██████▍   | 2586/4000 [34:52<18:35,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  65%|██████▍   | 2587/4000 [34:53<18:31,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  65%|██████▍   | 2588/4000 [34:54<18:32,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  65%|██████▍   | 2589/4000 [34:54<18:32,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]train loss: 0.8722588419914246
train acc: 0.7560728788375854

overall:  65%|██████▍   | 2590/4000 [34:55<17:51,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  65%|██████▍   | 2591/4000 [34:56<22:56,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.39s/it]
overall:  65%|██████▍   | 2592/4000 [34:57<21:35,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  65%|██████▍   | 2593/4000 [34:58<20:40,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  65%|██████▍   | 2594/4000 [34:59<19:57,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  65%|██████▍   | 2595/4000 [35:00<19:28,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  65%|██████▍   | 2596/4000 [35:00<19:06,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  65%|██████▍   | 2597/4000 [35:01<18:55,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  65%|██████▍   | 2598/4000 [35:02<18:47,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  65%|██████▍   | 2599/4000 [35:03<18:40,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]train loss: 0.8658398389816284
train acc: 0.7450000047683716

overall:  65%|██████▌   | 2600/4000 [35:04<18:39,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  65%|██████▌   | 2601/4000 [35:04<18:36,  1.25it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.25it/s]
overall:  65%|██████▌   | 2602/4000 [35:05<18:30,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  65%|██████▌   | 2603/4000 [35:06<18:30,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  65%|██████▌   | 2604/4000 [35:07<18:31,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  65%|██████▌   | 2605/4000 [35:08<18:24,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  65%|██████▌   | 2606/4000 [35:08<18:19,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  65%|██████▌   | 2607/4000 [35:09<18:19,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  65%|██████▌   | 2608/4000 [35:10<18:17,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  65%|██████▌   | 2609/4000 [35:11<18:14,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]train loss: 0.7897646427154541
train acc: 0.7960000038146973

overall:  65%|██████▌   | 2610/4000 [35:11<18:16,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  65%|██████▌   | 2611/4000 [35:12<18:14,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  65%|██████▌   | 2612/4000 [35:13<18:13,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  65%|██████▌   | 2613/4000 [35:14<18:13,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  65%|██████▌   | 2614/4000 [35:15<18:09,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  65%|██████▌   | 2615/4000 [35:15<18:13,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  65%|██████▌   | 2616/4000 [35:16<18:13,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  65%|██████▌   | 2617/4000 [35:17<18:11,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  65%|██████▌   | 2618/4000 [35:18<18:08,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  65%|██████▌   | 2619/4000 [35:19<18:09,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]train loss: 0.8681521415710449
train acc: 0.7699999809265137

overall:  66%|██████▌   | 2620/4000 [35:19<18:08,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  66%|██████▌   | 2621/4000 [35:20<18:06,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  66%|██████▌   | 2622/4000 [35:21<18:06,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  66%|██████▌   | 2623/4000 [35:22<18:04,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  66%|██████▌   | 2624/4000 [35:23<18:05,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  66%|██████▌   | 2625/4000 [35:23<18:01,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  66%|██████▌   | 2626/4000 [35:24<18:00,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  66%|██████▌   | 2627/4000 [35:25<17:21,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  66%|██████▌   | 2628/4000 [35:26<23:13,  1.02s/it]

train:   3%|▎         | 1/37 [00:01<00:54,  1.52s/it]
overall:  66%|██████▌   | 2629/4000 [35:27<21:42,  1.05it/s]

train:   5%|▌         | 2/37 [00:02<00:38,  1.10s/it]train loss: 0.9010079503059387
train acc: 0.7520242929458618

overall:  66%|██████▌   | 2630/4000 [35:28<20:36,  1.11it/s]

train:   8%|▊         | 3/37 [00:03<00:32,  1.05it/s]
overall:  66%|██████▌   | 2631/4000 [35:29<19:46,  1.15it/s]

train:  11%|█         | 4/37 [00:03<00:29,  1.13it/s]
overall:  66%|██████▌   | 2632/4000 [35:30<19:14,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.17it/s]
overall:  66%|██████▌   | 2633/4000 [35:30<18:48,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  66%|██████▌   | 2634/4000 [35:31<18:36,  1.22it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.22it/s]
overall:  66%|██████▌   | 2635/4000 [35:32<18:20,  1.24it/s]

train:  22%|██▏       | 8/37 [00:07<00:23,  1.24it/s]
overall:  66%|██████▌   | 2636/4000 [35:33<18:14,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  66%|██████▌   | 2637/4000 [35:33<18:06,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  66%|██████▌   | 2638/4000 [35:34<18:02,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  66%|██████▌   | 2639/4000 [35:35<18:04,  1.25it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.25it/s]train loss: 0.8414839506149292
train acc: 0.7799999713897705

overall:  66%|██████▌   | 2640/4000 [35:36<17:59,  1.26it/s]

train:  35%|███▌      | 13/37 [00:11<00:19,  1.26it/s]
overall:  66%|██████▌   | 2641/4000 [35:37<17:55,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  66%|██████▌   | 2642/4000 [35:37<17:50,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  66%|██████▌   | 2643/4000 [35:38<17:47,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  66%|██████▌   | 2644/4000 [35:39<17:47,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  66%|██████▌   | 2645/4000 [35:40<17:45,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  66%|██████▌   | 2646/4000 [35:41<17:50,  1.26it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.26it/s]
overall:  66%|██████▌   | 2647/4000 [35:41<17:46,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  66%|██████▌   | 2648/4000 [35:42<17:44,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  66%|██████▌   | 2649/4000 [35:43<17:43,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]train loss: 0.8799556493759155
train acc: 0.7450000047683716

overall:  66%|██████▋   | 2650/4000 [35:44<17:41,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  66%|██████▋   | 2651/4000 [35:45<17:45,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  66%|██████▋   | 2652/4000 [35:45<17:42,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  66%|██████▋   | 2653/4000 [35:46<17:41,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  66%|██████▋   | 2654/4000 [35:47<17:40,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:22<00:07,  1.27it/s]
overall:  66%|██████▋   | 2655/4000 [35:48<17:36,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  66%|██████▋   | 2656/4000 [35:48<17:35,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  66%|██████▋   | 2657/4000 [35:49<17:34,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  66%|██████▋   | 2658/4000 [35:50<17:36,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  66%|██████▋   | 2659/4000 [35:51<17:35,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]train loss: 0.8313538432121277
train acc: 0.7680000066757202

overall:  66%|██████▋   | 2660/4000 [35:52<17:35,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  67%|██████▋   | 2661/4000 [35:52<17:31,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  67%|██████▋   | 2662/4000 [35:53<17:32,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  67%|██████▋   | 2663/4000 [35:54<17:32,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  67%|██████▋   | 2664/4000 [35:55<16:54,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  67%|██████▋   | 2665/4000 [35:56<21:36,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.36s/it]
overall:  67%|██████▋   | 2666/4000 [35:57<20:29,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.04s/it]
overall:  67%|██████▋   | 2667/4000 [35:58<19:37,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  67%|██████▋   | 2668/4000 [35:59<18:56,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  67%|██████▋   | 2669/4000 [35:59<18:31,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]train loss: 0.8244982957839966
train acc: 0.7732793688774109

overall:  67%|██████▋   | 2670/4000 [36:00<18:11,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  67%|██████▋   | 2671/4000 [36:01<17:56,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  67%|██████▋   | 2672/4000 [36:02<17:48,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  67%|██████▋   | 2673/4000 [36:02<17:40,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  67%|██████▋   | 2674/4000 [36:03<17:38,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  67%|██████▋   | 2675/4000 [36:04<17:32,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  67%|██████▋   | 2676/4000 [36:05<17:29,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  67%|██████▋   | 2677/4000 [36:06<17:28,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  67%|██████▋   | 2678/4000 [36:06<17:28,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  67%|██████▋   | 2679/4000 [36:07<17:24,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]train loss: 0.8364564776420593
train acc: 0.7789999842643738

overall:  67%|██████▋   | 2680/4000 [36:08<17:22,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  67%|██████▋   | 2681/4000 [36:09<17:18,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  67%|██████▋   | 2682/4000 [36:10<17:17,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  67%|██████▋   | 2683/4000 [36:10<17:15,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  67%|██████▋   | 2684/4000 [36:11<17:14,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  67%|██████▋   | 2685/4000 [36:12<17:13,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  67%|██████▋   | 2686/4000 [36:13<17:15,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  67%|██████▋   | 2687/4000 [36:13<17:14,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  67%|██████▋   | 2688/4000 [36:14<17:11,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  67%|██████▋   | 2689/4000 [36:15<17:14,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]train loss: 0.8714934587478638
train acc: 0.7570000290870667

overall:  67%|██████▋   | 2690/4000 [36:16<17:16,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  67%|██████▋   | 2691/4000 [36:17<17:12,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  67%|██████▋   | 2692/4000 [36:17<17:07,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  67%|██████▋   | 2693/4000 [36:18<17:06,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  67%|██████▋   | 2694/4000 [36:19<17:05,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  67%|██████▋   | 2695/4000 [36:20<17:05,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  67%|██████▋   | 2696/4000 [36:21<17:04,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  67%|██████▋   | 2697/4000 [36:21<17:04,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  67%|██████▋   | 2698/4000 [36:22<17:06,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  67%|██████▋   | 2699/4000 [36:23<17:02,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]train loss: 0.8568153381347656
train acc: 0.7519999742507935

overall:  68%|██████▊   | 2700/4000 [36:24<17:01,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  68%|██████▊   | 2701/4000 [36:24<16:24,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  68%|██████▊   | 2702/4000 [36:26<21:20,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:50,  1.41s/it]
overall:  68%|██████▊   | 2703/4000 [36:27<20:07,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  68%|██████▊   | 2704/4000 [36:28<19:09,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  68%|██████▊   | 2705/4000 [36:28<18:30,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  68%|██████▊   | 2706/4000 [36:29<18:00,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  68%|██████▊   | 2707/4000 [36:30<17:39,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  68%|██████▊   | 2708/4000 [36:31<17:30,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  68%|██████▊   | 2709/4000 [36:31<17:22,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]train loss: 0.8358966112136841
train acc: 0.7793522477149963

overall:  68%|██████▊   | 2710/4000 [36:32<17:12,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  68%|██████▊   | 2711/4000 [36:33<17:10,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  68%|██████▊   | 2712/4000 [36:34<17:04,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  68%|██████▊   | 2713/4000 [36:35<17:00,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  68%|██████▊   | 2714/4000 [36:35<16:57,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.26it/s]
overall:  68%|██████▊   | 2715/4000 [36:36<16:54,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  68%|██████▊   | 2716/4000 [36:37<16:50,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  68%|██████▊   | 2717/4000 [36:38<16:52,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  68%|██████▊   | 2718/4000 [36:39<16:48,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  68%|██████▊   | 2719/4000 [36:39<16:44,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]train loss: 0.8326385617256165
train acc: 0.7749999761581421

overall:  68%|██████▊   | 2720/4000 [36:40<16:49,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  68%|██████▊   | 2721/4000 [36:41<16:46,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  68%|██████▊   | 2722/4000 [36:42<16:47,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  68%|██████▊   | 2723/4000 [36:42<16:45,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  68%|██████▊   | 2724/4000 [36:43<16:44,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  68%|██████▊   | 2725/4000 [36:44<16:44,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  68%|██████▊   | 2726/4000 [36:45<16:43,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  68%|██████▊   | 2727/4000 [36:46<16:43,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  68%|██████▊   | 2728/4000 [36:46<16:42,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  68%|██████▊   | 2729/4000 [36:47<16:42,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]train loss: 0.857380211353302
train acc: 0.7519999742507935

overall:  68%|██████▊   | 2730/4000 [36:48<16:42,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  68%|██████▊   | 2731/4000 [36:49<16:39,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  68%|██████▊   | 2732/4000 [36:50<16:41,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  68%|██████▊   | 2733/4000 [36:50<16:37,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  68%|██████▊   | 2734/4000 [36:51<16:34,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  68%|██████▊   | 2735/4000 [36:52<16:34,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  68%|██████▊   | 2736/4000 [36:53<16:33,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  68%|██████▊   | 2737/4000 [36:53<16:32,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  68%|██████▊   | 2738/4000 [36:54<15:59,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  68%|██████▊   | 2739/4000 [36:56<20:28,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]train loss: 0.8724212646484375
train acc: 0.7631579041481018

overall:  68%|██████▊   | 2740/4000 [36:56<19:16,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.02s/it]
overall:  69%|██████▊   | 2741/4000 [36:57<18:24,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  69%|██████▊   | 2742/4000 [36:58<17:51,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]
overall:  69%|██████▊   | 2743/4000 [36:59<17:22,  1.21it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.20it/s]
overall:  69%|██████▊   | 2744/4000 [37:00<17:05,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  69%|██████▊   | 2745/4000 [37:00<16:52,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.24it/s]
overall:  69%|██████▊   | 2746/4000 [37:01<16:44,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  69%|██████▊   | 2747/4000 [37:02<16:37,  1.26it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.26it/s]
overall:  69%|██████▊   | 2748/4000 [37:03<16:33,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  69%|██████▊   | 2749/4000 [37:04<16:31,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]train loss: 0.832262396812439
train acc: 0.7720000147819519

overall:  69%|██████▉   | 2750/4000 [37:04<16:28,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  69%|██████▉   | 2751/4000 [37:05<16:31,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  69%|██████▉   | 2752/4000 [37:06<16:27,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  69%|██████▉   | 2753/4000 [37:07<16:24,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  69%|██████▉   | 2754/4000 [37:07<16:19,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  69%|██████▉   | 2755/4000 [37:08<16:18,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  69%|██████▉   | 2756/4000 [37:09<16:20,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  69%|██████▉   | 2757/4000 [37:10<16:17,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  69%|██████▉   | 2758/4000 [37:11<16:15,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  69%|██████▉   | 2759/4000 [37:11<16:16,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]train loss: 0.8554965257644653
train acc: 0.7689999938011169

overall:  69%|██████▉   | 2760/4000 [37:12<16:15,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  69%|██████▉   | 2761/4000 [37:13<16:12,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  69%|██████▉   | 2762/4000 [37:14<16:11,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  69%|██████▉   | 2763/4000 [37:15<16:15,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  69%|██████▉   | 2764/4000 [37:15<16:13,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  69%|██████▉   | 2765/4000 [37:16<16:13,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  69%|██████▉   | 2766/4000 [37:17<16:09,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  69%|██████▉   | 2767/4000 [37:18<16:07,  1.28it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.28it/s]
overall:  69%|██████▉   | 2768/4000 [37:18<16:09,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  69%|██████▉   | 2769/4000 [37:19<16:08,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:24<00:04,  1.27it/s]train loss: 0.8421230316162109
train acc: 0.7670000195503235

overall:  69%|██████▉   | 2770/4000 [37:20<16:05,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  69%|██████▉   | 2771/4000 [37:21<16:04,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  69%|██████▉   | 2772/4000 [37:22<16:03,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  69%|██████▉   | 2773/4000 [37:22<16:02,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  69%|██████▉   | 2774/4000 [37:23<16:03,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  69%|██████▉   | 2775/4000 [37:24<15:28,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  69%|██████▉   | 2776/4000 [37:25<20:11,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  69%|██████▉   | 2777/4000 [37:26<18:56,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  69%|██████▉   | 2778/4000 [37:27<18:01,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  69%|██████▉   | 2779/4000 [37:28<17:23,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]train loss: 0.8595163226127625
train acc: 0.7661943435668945

overall:  70%|██████▉   | 2780/4000 [37:29<17:01,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  70%|██████▉   | 2781/4000 [37:29<16:40,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  70%|██████▉   | 2782/4000 [37:30<16:26,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  70%|██████▉   | 2783/4000 [37:31<16:20,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  70%|██████▉   | 2784/4000 [37:32<16:14,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  70%|██████▉   | 2785/4000 [37:32<16:08,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  70%|██████▉   | 2786/4000 [37:33<16:02,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  70%|██████▉   | 2787/4000 [37:34<16:00,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  70%|██████▉   | 2788/4000 [37:35<16:03,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  70%|██████▉   | 2789/4000 [37:36<15:59,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]train loss: 0.8426116108894348
train acc: 0.7580000162124634

overall:  70%|██████▉   | 2790/4000 [37:36<15:55,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  70%|██████▉   | 2791/4000 [37:37<15:54,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  70%|██████▉   | 2792/4000 [37:38<15:51,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  70%|██████▉   | 2793/4000 [37:39<15:49,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  70%|██████▉   | 2794/4000 [37:40<15:50,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  70%|██████▉   | 2795/4000 [37:40<15:51,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  70%|██████▉   | 2796/4000 [37:41<15:51,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  70%|██████▉   | 2797/4000 [37:42<15:50,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  70%|██████▉   | 2798/4000 [37:43<15:48,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  70%|██████▉   | 2799/4000 [37:44<15:46,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]train loss: 0.8294664621353149
train acc: 0.7789999842643738

overall:  70%|███████   | 2800/4000 [37:44<15:50,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  70%|███████   | 2801/4000 [37:45<15:47,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  70%|███████   | 2802/4000 [37:46<15:45,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  70%|███████   | 2803/4000 [37:47<15:42,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  70%|███████   | 2804/4000 [37:47<15:41,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  70%|███████   | 2805/4000 [37:48<15:41,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  70%|███████   | 2806/4000 [37:49<15:38,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  70%|███████   | 2807/4000 [37:50<15:41,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  70%|███████   | 2808/4000 [37:51<15:37,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  70%|███████   | 2809/4000 [37:51<15:38,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]train loss: 0.8382335901260376
train acc: 0.7689999938011169

overall:  70%|███████   | 2810/4000 [37:52<15:38,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  70%|███████   | 2811/4000 [37:53<15:40,  1.26it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.26it/s]
overall:  70%|███████   | 2812/4000 [37:54<15:05,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  70%|███████   | 2813/4000 [37:55<19:37,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.42s/it]
overall:  70%|███████   | 2814/4000 [37:56<18:26,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  70%|███████   | 2815/4000 [37:57<17:32,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  70%|███████   | 2816/4000 [37:58<16:54,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  70%|███████   | 2817/4000 [37:58<16:27,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  70%|███████   | 2818/4000 [37:59<16:09,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  70%|███████   | 2819/4000 [38:00<15:56,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]train loss: 0.8309394717216492
train acc: 0.7702429294586182

overall:  70%|███████   | 2820/4000 [38:01<15:47,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  71%|███████   | 2821/4000 [38:02<15:42,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  71%|███████   | 2822/4000 [38:02<15:36,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  71%|███████   | 2823/4000 [38:03<15:32,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  71%|███████   | 2824/4000 [38:04<15:30,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  71%|███████   | 2825/4000 [38:05<15:33,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  71%|███████   | 2826/4000 [38:05<15:32,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  71%|███████   | 2827/4000 [38:06<15:27,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  71%|███████   | 2828/4000 [38:07<15:26,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  71%|███████   | 2829/4000 [38:08<15:23,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]train loss: 0.8583779335021973
train acc: 0.7599999904632568

overall:  71%|███████   | 2830/4000 [38:09<15:22,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  71%|███████   | 2831/4000 [38:09<15:22,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  71%|███████   | 2832/4000 [38:10<15:20,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  71%|███████   | 2833/4000 [38:11<15:23,  1.26it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.26it/s]
overall:  71%|███████   | 2834/4000 [38:12<15:18,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  71%|███████   | 2835/4000 [38:13<15:15,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  71%|███████   | 2836/4000 [38:13<15:15,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  71%|███████   | 2837/4000 [38:14<15:19,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  71%|███████   | 2838/4000 [38:15<15:19,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  71%|███████   | 2839/4000 [38:16<15:15,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]train loss: 0.8242160677909851
train acc: 0.7789999842643738

overall:  71%|███████   | 2840/4000 [38:17<15:14,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  71%|███████   | 2841/4000 [38:17<15:13,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  71%|███████   | 2842/4000 [38:18<15:11,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  71%|███████   | 2843/4000 [38:19<15:12,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  71%|███████   | 2844/4000 [38:20<15:08,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  71%|███████   | 2845/4000 [38:20<15:10,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  71%|███████   | 2846/4000 [38:21<15:09,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  71%|███████   | 2847/4000 [38:22<15:06,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  71%|███████   | 2848/4000 [38:23<15:09,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  71%|███████   | 2849/4000 [38:24<14:35,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]train loss: 0.8250277638435364
train acc: 0.7692307829856873

overall:  71%|███████▏  | 2850/4000 [38:25<19:01,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  71%|███████▏  | 2851/4000 [38:26<17:50,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  71%|███████▏  | 2852/4000 [38:27<16:58,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  71%|███████▏  | 2853/4000 [38:27<16:22,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  71%|███████▏  | 2854/4000 [38:28<16:01,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  71%|███████▏  | 2855/4000 [38:29<15:45,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  71%|███████▏  | 2856/4000 [38:30<15:29,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  71%|███████▏  | 2857/4000 [38:31<15:23,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  71%|███████▏  | 2858/4000 [38:31<15:13,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  71%|███████▏  | 2859/4000 [38:32<15:08,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]train loss: 0.8593258857727051
train acc: 0.7720000147819519

overall:  72%|███████▏  | 2860/4000 [38:33<15:06,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  72%|███████▏  | 2861/4000 [38:34<15:03,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  72%|███████▏  | 2862/4000 [38:35<15:07,  1.25it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.25it/s]
overall:  72%|███████▏  | 2863/4000 [38:35<15:03,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  72%|███████▏  | 2864/4000 [38:36<15:01,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  72%|███████▏  | 2865/4000 [38:37<14:58,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  72%|███████▏  | 2866/4000 [38:38<14:58,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  72%|███████▏  | 2867/4000 [38:38<14:53,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  72%|███████▏  | 2868/4000 [38:39<14:54,  1.26it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  72%|███████▏  | 2869/4000 [38:40<14:54,  1.26it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.26it/s]train loss: 0.8758707046508789
train acc: 0.7599999904632568

overall:  72%|███████▏  | 2870/4000 [38:41<14:54,  1.26it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.26it/s]
overall:  72%|███████▏  | 2871/4000 [38:42<14:49,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  72%|███████▏  | 2872/4000 [38:42<14:47,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  72%|███████▏  | 2873/4000 [38:43<14:48,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  72%|███████▏  | 2874/4000 [38:44<14:51,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  72%|███████▏  | 2875/4000 [38:45<14:50,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  72%|███████▏  | 2876/4000 [38:46<14:48,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  72%|███████▏  | 2877/4000 [38:46<14:44,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  72%|███████▏  | 2878/4000 [38:47<14:44,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  72%|███████▏  | 2879/4000 [38:48<14:43,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]train loss: 0.8253615498542786
train acc: 0.7749999761581421

overall:  72%|███████▏  | 2880/4000 [38:49<14:41,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  72%|███████▏  | 2881/4000 [38:50<14:42,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  72%|███████▏  | 2882/4000 [38:50<14:40,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  72%|███████▏  | 2883/4000 [38:51<14:37,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  72%|███████▏  | 2884/4000 [38:52<14:36,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  72%|███████▏  | 2885/4000 [38:53<14:36,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  72%|███████▏  | 2886/4000 [38:53<14:05,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  72%|███████▏  | 2887/4000 [38:55<18:28,  1.00it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  72%|███████▏  | 2888/4000 [38:56<17:17,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.06s/it]
overall:  72%|███████▏  | 2889/4000 [38:56<16:29,  1.12it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]train loss: 0.7983323335647583
train acc: 0.7803643941879272

overall:  72%|███████▏  | 2890/4000 [38:57<15:52,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  72%|███████▏  | 2891/4000 [38:58<15:27,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  72%|███████▏  | 2892/4000 [38:59<15:10,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  72%|███████▏  | 2893/4000 [39:00<14:57,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  72%|███████▏  | 2894/4000 [39:00<14:51,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  72%|███████▏  | 2895/4000 [39:01<14:43,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  72%|███████▏  | 2896/4000 [39:02<14:37,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  72%|███████▏  | 2897/4000 [39:03<14:34,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  72%|███████▏  | 2898/4000 [39:04<14:31,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  72%|███████▏  | 2899/4000 [39:04<14:33,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]train loss: 0.888358473777771
train acc: 0.7540000081062317

overall:  72%|███████▎  | 2900/4000 [39:05<14:29,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  73%|███████▎  | 2901/4000 [39:06<14:29,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  73%|███████▎  | 2902/4000 [39:07<14:27,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  73%|███████▎  | 2903/4000 [39:08<14:25,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  73%|███████▎  | 2904/4000 [39:08<14:23,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  73%|███████▎  | 2905/4000 [39:09<14:22,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  73%|███████▎  | 2906/4000 [39:10<14:21,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  73%|███████▎  | 2907/4000 [39:11<14:20,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  73%|███████▎  | 2908/4000 [39:11<14:18,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  73%|███████▎  | 2909/4000 [39:12<14:16,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]train loss: 0.8194149136543274
train acc: 0.7789999842643738

overall:  73%|███████▎  | 2910/4000 [39:13<14:17,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  73%|███████▎  | 2911/4000 [39:14<14:18,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  73%|███████▎  | 2912/4000 [39:15<14:17,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  73%|███████▎  | 2913/4000 [39:15<14:17,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  73%|███████▎  | 2914/4000 [39:16<14:17,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  73%|███████▎  | 2915/4000 [39:17<14:16,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  73%|███████▎  | 2916/4000 [39:18<14:15,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  73%|███████▎  | 2917/4000 [39:19<14:13,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  73%|███████▎  | 2918/4000 [39:19<14:10,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  73%|███████▎  | 2919/4000 [39:20<14:08,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]train loss: 0.8052447438240051
train acc: 0.7870000004768372

overall:  73%|███████▎  | 2920/4000 [39:21<14:07,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  73%|███████▎  | 2921/4000 [39:22<14:08,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  73%|███████▎  | 2922/4000 [39:22<14:05,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  73%|███████▎  | 2923/4000 [39:23<13:35,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  73%|███████▎  | 2924/4000 [39:25<17:29,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.38s/it]
overall:  73%|███████▎  | 2925/4000 [39:25<16:28,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  73%|███████▎  | 2926/4000 [39:26<15:45,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  73%|███████▎  | 2927/4000 [39:27<15:16,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  73%|███████▎  | 2928/4000 [39:28<14:54,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  73%|███████▎  | 2929/4000 [39:29<14:37,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]train loss: 0.8039442896842957
train acc: 0.7813765406608582

overall:  73%|███████▎  | 2930/4000 [39:29<14:27,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  73%|███████▎  | 2931/4000 [39:30<14:20,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  73%|███████▎  | 2932/4000 [39:31<14:14,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  73%|███████▎  | 2933/4000 [39:32<14:10,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  73%|███████▎  | 2934/4000 [39:33<14:07,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  73%|███████▎  | 2935/4000 [39:33<14:03,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  73%|███████▎  | 2936/4000 [39:34<14:07,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  73%|███████▎  | 2937/4000 [39:35<14:03,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  73%|███████▎  | 2938/4000 [39:36<14:00,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  73%|███████▎  | 2939/4000 [39:36<13:56,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]train loss: 0.8091007471084595
train acc: 0.7839999794960022

overall:  74%|███████▎  | 2940/4000 [39:37<13:55,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  74%|███████▎  | 2941/4000 [39:38<13:54,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  74%|███████▎  | 2942/4000 [39:39<13:51,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  74%|███████▎  | 2943/4000 [39:40<13:51,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  74%|███████▎  | 2944/4000 [39:40<13:51,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  74%|███████▎  | 2945/4000 [39:41<13:50,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  74%|███████▎  | 2946/4000 [39:42<13:50,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  74%|███████▎  | 2947/4000 [39:43<13:49,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  74%|███████▎  | 2948/4000 [39:44<13:54,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  74%|███████▎  | 2949/4000 [39:44<13:52,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]train loss: 0.8526383638381958
train acc: 0.7609999775886536

overall:  74%|███████▍  | 2950/4000 [39:45<13:49,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  74%|███████▍  | 2951/4000 [39:46<13:45,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  74%|███████▍  | 2952/4000 [39:47<13:45,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  74%|███████▍  | 2953/4000 [39:48<13:43,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  74%|███████▍  | 2954/4000 [39:48<13:41,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  74%|███████▍  | 2955/4000 [39:49<13:41,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  74%|███████▍  | 2956/4000 [39:50<13:41,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  74%|███████▍  | 2957/4000 [39:51<13:41,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  74%|███████▍  | 2958/4000 [39:51<13:40,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  74%|███████▍  | 2959/4000 [39:52<13:38,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]train loss: 0.8521707653999329
train acc: 0.7560728788375854

overall:  74%|███████▍  | 2960/4000 [39:53<13:09,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  74%|███████▍  | 2961/4000 [39:54<17:04,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.42s/it]
overall:  74%|███████▍  | 2962/4000 [39:55<16:01,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  74%|███████▍  | 2963/4000 [39:56<15:15,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  74%|███████▍  | 2964/4000 [39:57<14:50,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  74%|███████▍  | 2965/4000 [39:58<15:06,  1.14it/s]

train:  14%|█▎        | 5/37 [00:04<00:28,  1.12it/s]
overall:  74%|███████▍  | 2966/4000 [39:59<14:37,  1.18it/s]

train:  16%|█▌        | 6/37 [00:05<00:26,  1.17it/s]
overall:  74%|███████▍  | 2967/4000 [39:59<14:18,  1.20it/s]

train:  19%|█▉        | 7/37 [00:06<00:25,  1.20it/s]
overall:  74%|███████▍  | 2968/4000 [40:00<14:07,  1.22it/s]

train:  22%|██▏       | 8/37 [00:07<00:23,  1.22it/s]
overall:  74%|███████▍  | 2969/4000 [40:01<13:55,  1.23it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.23it/s]train loss: 0.8199052810668945
train acc: 0.7710000276565552

overall:  74%|███████▍  | 2970/4000 [40:02<13:48,  1.24it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.24it/s]
overall:  74%|███████▍  | 2971/4000 [40:02<13:43,  1.25it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.25it/s]
overall:  74%|███████▍  | 2972/4000 [40:03<13:38,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  74%|███████▍  | 2973/4000 [40:04<13:36,  1.26it/s]

train:  35%|███▌      | 13/37 [00:11<00:19,  1.26it/s]
overall:  74%|███████▍  | 2974/4000 [40:05<13:31,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  74%|███████▍  | 2975/4000 [40:06<13:31,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  74%|███████▍  | 2976/4000 [40:06<13:32,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  74%|███████▍  | 2977/4000 [40:07<13:29,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  74%|███████▍  | 2978/4000 [40:08<13:25,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  74%|███████▍  | 2979/4000 [40:09<13:23,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]train loss: 0.8481431007385254
train acc: 0.765999972820282

overall:  74%|███████▍  | 2980/4000 [40:10<13:24,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  75%|███████▍  | 2981/4000 [40:10<13:22,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  75%|███████▍  | 2982/4000 [40:11<13:20,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  75%|███████▍  | 2983/4000 [40:12<13:21,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  75%|███████▍  | 2984/4000 [40:13<13:18,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  75%|███████▍  | 2985/4000 [40:13<13:18,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  75%|███████▍  | 2986/4000 [40:14<13:19,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  75%|███████▍  | 2987/4000 [40:15<13:17,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:22<00:07,  1.27it/s]
overall:  75%|███████▍  | 2988/4000 [40:16<13:19,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  75%|███████▍  | 2989/4000 [40:17<13:18,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]train loss: 0.796707034111023
train acc: 0.7879999876022339

overall:  75%|███████▍  | 2990/4000 [40:17<13:17,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  75%|███████▍  | 2991/4000 [40:18<13:14,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  75%|███████▍  | 2992/4000 [40:19<13:13,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  75%|███████▍  | 2993/4000 [40:20<13:13,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  75%|███████▍  | 2994/4000 [40:21<13:10,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  75%|███████▍  | 2995/4000 [40:21<13:11,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  75%|███████▍  | 2996/4000 [40:22<13:11,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  75%|███████▍  | 2997/4000 [40:23<12:41,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  75%|███████▍  | 2998/4000 [40:24<16:12,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  75%|███████▍  | 2999/4000 [40:25<15:18,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]train loss: 0.8652181625366211
train acc: 0.7601214647293091

overall:  75%|███████▌  | 3000/4000 [40:26<14:38,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  75%|███████▌  | 3001/4000 [40:27<14:11,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  75%|███████▌  | 3002/4000 [40:27<13:51,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  75%|███████▌  | 3003/4000 [40:28<13:37,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  75%|███████▌  | 3004/4000 [40:29<13:26,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  75%|███████▌  | 3005/4000 [40:30<13:18,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  75%|███████▌  | 3006/4000 [40:31<13:15,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  75%|███████▌  | 3007/4000 [40:31<13:09,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  75%|███████▌  | 3008/4000 [40:32<13:08,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  75%|███████▌  | 3009/4000 [40:33<13:08,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]train loss: 0.7915628552436829
train acc: 0.7860000133514404

overall:  75%|███████▌  | 3010/4000 [40:34<13:11,  1.25it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.25it/s]
overall:  75%|███████▌  | 3011/4000 [40:35<13:07,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  75%|███████▌  | 3012/4000 [40:35<13:05,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  75%|███████▌  | 3013/4000 [40:36<13:00,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  75%|███████▌  | 3014/4000 [40:37<12:59,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  75%|███████▌  | 3015/4000 [40:38<12:56,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  75%|███████▌  | 3016/4000 [40:39<12:54,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  75%|███████▌  | 3017/4000 [40:39<12:52,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  75%|███████▌  | 3018/4000 [40:40<12:54,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  75%|███████▌  | 3019/4000 [40:41<12:51,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]train loss: 0.8234642744064331
train acc: 0.7720000147819519

overall:  76%|███████▌  | 3020/4000 [40:42<12:51,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  76%|███████▌  | 3021/4000 [40:42<12:50,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  76%|███████▌  | 3022/4000 [40:43<12:56,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  76%|███████▌  | 3023/4000 [40:44<12:53,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  76%|███████▌  | 3024/4000 [40:45<12:50,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  76%|███████▌  | 3025/4000 [40:46<12:49,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  76%|███████▌  | 3026/4000 [40:46<12:47,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  76%|███████▌  | 3027/4000 [40:47<12:45,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  76%|███████▌  | 3028/4000 [40:48<12:43,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  76%|███████▌  | 3029/4000 [40:49<12:42,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]train loss: 0.8521655201911926
train acc: 0.7670000195503235

overall:  76%|███████▌  | 3030/4000 [40:50<12:43,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  76%|███████▌  | 3031/4000 [40:50<12:41,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  76%|███████▌  | 3032/4000 [40:51<12:40,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  76%|███████▌  | 3033/4000 [40:52<12:40,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  76%|███████▌  | 3034/4000 [40:53<12:12,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  76%|███████▌  | 3035/4000 [40:54<15:55,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  76%|███████▌  | 3036/4000 [40:55<14:55,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  76%|███████▌  | 3037/4000 [40:56<14:14,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  76%|███████▌  | 3038/4000 [40:57<13:43,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  76%|███████▌  | 3039/4000 [40:57<13:23,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]train loss: 0.8359734416007996
train acc: 0.7591093182563782

overall:  76%|███████▌  | 3040/4000 [40:58<13:09,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  76%|███████▌  | 3041/4000 [40:59<13:00,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  76%|███████▌  | 3042/4000 [41:00<12:53,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  76%|███████▌  | 3043/4000 [41:00<12:47,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  76%|███████▌  | 3044/4000 [41:01<12:42,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  76%|███████▌  | 3045/4000 [41:02<12:39,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  76%|███████▌  | 3046/4000 [41:03<12:35,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  76%|███████▌  | 3047/4000 [41:04<12:37,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  76%|███████▌  | 3048/4000 [41:04<12:32,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  76%|███████▌  | 3049/4000 [41:05<12:29,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]train loss: 0.8415440320968628
train acc: 0.7670000195503235

overall:  76%|███████▋  | 3050/4000 [41:06<12:30,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  76%|███████▋  | 3051/4000 [41:07<12:31,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  76%|███████▋  | 3052/4000 [41:08<12:31,  1.26it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.26it/s]
overall:  76%|███████▋  | 3053/4000 [41:08<12:30,  1.26it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.26it/s]
overall:  76%|███████▋  | 3054/4000 [41:09<12:30,  1.26it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.26it/s]
overall:  76%|███████▋  | 3055/4000 [41:10<12:26,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  76%|███████▋  | 3056/4000 [41:11<12:25,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  76%|███████▋  | 3057/4000 [41:12<12:25,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  76%|███████▋  | 3058/4000 [41:12<12:20,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  76%|███████▋  | 3059/4000 [41:13<12:23,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]train loss: 0.8174281120300293
train acc: 0.7879999876022339

overall:  76%|███████▋  | 3060/4000 [41:14<12:21,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  77%|███████▋  | 3061/4000 [41:15<12:18,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  77%|███████▋  | 3062/4000 [41:15<12:18,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  77%|███████▋  | 3063/4000 [41:16<12:17,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  77%|███████▋  | 3064/4000 [41:17<12:17,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  77%|███████▋  | 3065/4000 [41:18<12:16,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  77%|███████▋  | 3066/4000 [41:19<12:15,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  77%|███████▋  | 3067/4000 [41:19<12:15,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  77%|███████▋  | 3068/4000 [41:20<12:13,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  77%|███████▋  | 3069/4000 [41:21<12:12,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]train loss: 0.8107002377510071
train acc: 0.7730000019073486

overall:  77%|███████▋  | 3070/4000 [41:22<12:11,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  77%|███████▋  | 3071/4000 [41:22<11:44,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  77%|███████▋  | 3072/4000 [41:24<15:19,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  77%|███████▋  | 3073/4000 [41:25<14:25,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:  77%|███████▋  | 3074/4000 [41:26<13:44,  1.12it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  77%|███████▋  | 3075/4000 [41:26<13:15,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  77%|███████▋  | 3076/4000 [41:27<12:53,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  77%|███████▋  | 3077/4000 [41:28<12:37,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  77%|███████▋  | 3078/4000 [41:29<12:28,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  77%|███████▋  | 3079/4000 [41:29<12:19,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]train loss: 0.8494757413864136
train acc: 0.7661943435668945

overall:  77%|███████▋  | 3080/4000 [41:30<12:16,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  77%|███████▋  | 3081/4000 [41:31<12:11,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  77%|███████▋  | 3082/4000 [41:32<12:07,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  77%|███████▋  | 3083/4000 [41:33<12:05,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  77%|███████▋  | 3084/4000 [41:33<12:07,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  77%|███████▋  | 3085/4000 [41:34<12:06,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  77%|███████▋  | 3086/4000 [41:35<12:04,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  77%|███████▋  | 3087/4000 [41:36<12:01,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  77%|███████▋  | 3088/4000 [41:37<11:59,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  77%|███████▋  | 3089/4000 [41:37<11:58,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]train loss: 0.8260378837585449
train acc: 0.7730000019073486

overall:  77%|███████▋  | 3090/4000 [41:38<11:56,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  77%|███████▋  | 3091/4000 [41:39<11:53,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  77%|███████▋  | 3092/4000 [41:40<11:54,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  77%|███████▋  | 3093/4000 [41:41<11:52,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  77%|███████▋  | 3094/4000 [41:41<11:51,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  77%|███████▋  | 3095/4000 [41:42<11:50,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  77%|███████▋  | 3096/4000 [41:43<11:51,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  77%|███████▋  | 3097/4000 [41:44<11:53,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  77%|███████▋  | 3098/4000 [41:44<11:50,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  77%|███████▋  | 3099/4000 [41:45<11:48,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]train loss: 0.8179314732551575
train acc: 0.7630000114440918

overall:  78%|███████▊  | 3100/4000 [41:46<11:47,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  78%|███████▊  | 3101/4000 [41:47<11:46,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  78%|███████▊  | 3102/4000 [41:48<11:46,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  78%|███████▊  | 3103/4000 [41:48<11:44,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  78%|███████▊  | 3104/4000 [41:49<11:45,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  78%|███████▊  | 3105/4000 [41:50<11:43,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  78%|███████▊  | 3106/4000 [41:51<11:41,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  78%|███████▊  | 3107/4000 [41:52<11:40,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  78%|███████▊  | 3108/4000 [41:52<11:17,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  78%|███████▊  | 3109/4000 [41:54<14:21,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:48,  1.35s/it]train loss: 0.8415406942367554
train acc: 0.7783401012420654

overall:  78%|███████▊  | 3110/4000 [41:54<13:34,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.02s/it]
overall:  78%|███████▊  | 3111/4000 [41:55<12:58,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  78%|███████▊  | 3112/4000 [41:56<12:37,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  78%|███████▊  | 3113/4000 [41:57<12:17,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  78%|███████▊  | 3114/4000 [41:58<12:04,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  78%|███████▊  | 3115/4000 [41:58<11:56,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  78%|███████▊  | 3116/4000 [41:59<11:49,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  78%|███████▊  | 3117/4000 [42:00<11:46,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  78%|███████▊  | 3118/4000 [42:01<11:43,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  78%|███████▊  | 3119/4000 [42:02<11:40,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]train loss: 0.8027439117431641
train acc: 0.781000018119812

overall:  78%|███████▊  | 3120/4000 [42:02<11:36,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  78%|███████▊  | 3121/4000 [42:03<11:37,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  78%|███████▊  | 3122/4000 [42:04<11:36,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  78%|███████▊  | 3123/4000 [42:05<11:33,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  78%|███████▊  | 3124/4000 [42:06<11:33,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  78%|███████▊  | 3125/4000 [42:06<11:33,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  78%|███████▊  | 3126/4000 [42:07<11:31,  1.26it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.26it/s]
overall:  78%|███████▊  | 3127/4000 [42:08<11:29,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  78%|███████▊  | 3128/4000 [42:09<11:30,  1.26it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.26it/s]
overall:  78%|███████▊  | 3129/4000 [42:09<11:27,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]train loss: 0.7947247624397278
train acc: 0.7910000085830688

overall:  78%|███████▊  | 3130/4000 [42:10<11:28,  1.26it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.26it/s]
overall:  78%|███████▊  | 3131/4000 [42:11<11:26,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  78%|███████▊  | 3132/4000 [42:12<11:23,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  78%|███████▊  | 3133/4000 [42:13<11:23,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  78%|███████▊  | 3134/4000 [42:13<11:23,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  78%|███████▊  | 3135/4000 [42:14<11:22,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  78%|███████▊  | 3136/4000 [42:15<11:21,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  78%|███████▊  | 3137/4000 [42:16<11:20,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  78%|███████▊  | 3138/4000 [42:17<11:18,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  78%|███████▊  | 3139/4000 [42:17<11:18,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]train loss: 0.8705493211746216
train acc: 0.7599999904632568

overall:  78%|███████▊  | 3140/4000 [42:18<11:18,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  79%|███████▊  | 3141/4000 [42:19<11:19,  1.26it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.26it/s]
overall:  79%|███████▊  | 3142/4000 [42:20<11:20,  1.26it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.26it/s]
overall:  79%|███████▊  | 3143/4000 [42:21<11:17,  1.26it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.26it/s]
overall:  79%|███████▊  | 3144/4000 [42:21<11:17,  1.26it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.26it/s]
overall:  79%|███████▊  | 3145/4000 [42:22<10:52,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  79%|███████▊  | 3146/4000 [42:23<13:48,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:48,  1.36s/it]
overall:  79%|███████▊  | 3147/4000 [42:24<13:01,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.02s/it]
overall:  79%|███████▊  | 3148/4000 [42:25<12:26,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  79%|███████▊  | 3149/4000 [42:26<12:03,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]train loss: 0.8038098216056824
train acc: 0.7844129800796509

overall:  79%|███████▉  | 3150/4000 [42:27<11:47,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  79%|███████▉  | 3151/4000 [42:27<11:36,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  79%|███████▉  | 3152/4000 [42:28<11:30,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  79%|███████▉  | 3153/4000 [42:29<11:22,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  79%|███████▉  | 3154/4000 [42:30<11:17,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  79%|███████▉  | 3155/4000 [42:31<11:14,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  79%|███████▉  | 3156/4000 [42:31<11:09,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  79%|███████▉  | 3157/4000 [42:32<11:07,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  79%|███████▉  | 3158/4000 [42:33<11:07,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  79%|███████▉  | 3159/4000 [42:34<11:06,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]train loss: 0.8067226409912109
train acc: 0.7870000004768372

overall:  79%|███████▉  | 3160/4000 [42:35<11:04,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  79%|███████▉  | 3161/4000 [42:35<11:02,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  79%|███████▉  | 3162/4000 [42:36<11:01,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  79%|███████▉  | 3163/4000 [42:37<11:01,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.27it/s]
overall:  79%|███████▉  | 3164/4000 [42:38<11:00,  1.26it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.26it/s]
overall:  79%|███████▉  | 3165/4000 [42:38<10:59,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  79%|███████▉  | 3166/4000 [42:39<10:56,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  79%|███████▉  | 3167/4000 [42:40<10:54,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  79%|███████▉  | 3168/4000 [42:41<10:53,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  79%|███████▉  | 3169/4000 [42:42<10:51,  1.28it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.28it/s]train loss: 0.8005163073539734
train acc: 0.7919999957084656

overall:  79%|███████▉  | 3170/4000 [42:42<10:53,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  79%|███████▉  | 3171/4000 [42:43<10:52,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  79%|███████▉  | 3172/4000 [42:44<10:53,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  79%|███████▉  | 3173/4000 [42:45<10:49,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  79%|███████▉  | 3174/4000 [42:46<10:49,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  79%|███████▉  | 3175/4000 [42:46<10:50,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  79%|███████▉  | 3176/4000 [42:47<10:51,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  79%|███████▉  | 3177/4000 [42:48<10:49,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  79%|███████▉  | 3178/4000 [42:49<10:46,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  79%|███████▉  | 3179/4000 [42:49<10:44,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]train loss: 0.8665248155593872
train acc: 0.7440000176429749

overall:  80%|███████▉  | 3180/4000 [42:50<10:43,  1.28it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  80%|███████▉  | 3181/4000 [42:51<10:42,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  80%|███████▉  | 3182/4000 [42:52<10:20,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.25it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  80%|███████▉  | 3183/4000 [42:53<13:08,  1.04it/s]

train:   3%|▎         | 1/37 [00:01<00:48,  1.36s/it]
overall:  80%|███████▉  | 3184/4000 [42:54<12:25,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.03s/it]
overall:  80%|███████▉  | 3185/4000 [42:55<11:54,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  80%|███████▉  | 3186/4000 [42:56<11:32,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  80%|███████▉  | 3187/4000 [42:56<11:15,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  80%|███████▉  | 3188/4000 [42:57<11:03,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  80%|███████▉  | 3189/4000 [42:58<10:57,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]train loss: 0.8369467854499817
train acc: 0.7651821970939636

overall:  80%|███████▉  | 3190/4000 [42:59<10:49,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  80%|███████▉  | 3191/4000 [43:00<10:47,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  80%|███████▉  | 3192/4000 [43:00<10:42,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  80%|███████▉  | 3193/4000 [43:01<10:40,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  80%|███████▉  | 3194/4000 [43:02<10:38,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  80%|███████▉  | 3195/4000 [43:03<10:39,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  80%|███████▉  | 3196/4000 [43:03<10:38,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  80%|███████▉  | 3197/4000 [43:04<10:36,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  80%|███████▉  | 3198/4000 [43:05<10:35,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  80%|███████▉  | 3199/4000 [43:06<10:33,  1.26it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.26it/s]train loss: 0.8301399350166321
train acc: 0.7599999904632568

overall:  80%|████████  | 3200/4000 [43:07<10:31,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.27it/s]
overall:  80%|████████  | 3201/4000 [43:07<10:30,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  80%|████████  | 3202/4000 [43:08<10:28,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  80%|████████  | 3203/4000 [43:09<10:27,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  80%|████████  | 3204/4000 [43:10<10:27,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  80%|████████  | 3205/4000 [43:11<10:27,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  80%|████████  | 3206/4000 [43:11<10:26,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  80%|████████  | 3207/4000 [43:12<10:26,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  80%|████████  | 3208/4000 [43:13<10:24,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  80%|████████  | 3209/4000 [43:14<10:23,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]train loss: 0.8311951756477356
train acc: 0.7799999713897705

overall:  80%|████████  | 3210/4000 [43:15<10:23,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  80%|████████  | 3211/4000 [43:15<10:21,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  80%|████████  | 3212/4000 [43:16<10:19,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  80%|████████  | 3213/4000 [43:17<10:20,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  80%|████████  | 3214/4000 [43:18<10:17,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  80%|████████  | 3215/4000 [43:18<10:19,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  80%|████████  | 3216/4000 [43:19<10:16,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  80%|████████  | 3217/4000 [43:20<10:17,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  80%|████████  | 3218/4000 [43:21<10:16,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  80%|████████  | 3219/4000 [43:22<09:52,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]train loss: 0.7893862128257751
train acc: 0.7904858589172363

overall:  80%|████████  | 3220/4000 [43:23<12:38,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  81%|████████  | 3221/4000 [43:24<11:55,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  81%|████████  | 3222/4000 [43:25<11:23,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  81%|████████  | 3223/4000 [43:25<11:03,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  81%|████████  | 3224/4000 [43:26<10:45,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  81%|████████  | 3225/4000 [43:27<10:35,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  81%|████████  | 3226/4000 [43:28<10:29,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  81%|████████  | 3227/4000 [43:29<10:23,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  81%|████████  | 3228/4000 [43:29<10:21,  1.24it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.24it/s]
overall:  81%|████████  | 3229/4000 [43:30<10:17,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]train loss: 0.8059126138687134
train acc: 0.7820000052452087

overall:  81%|████████  | 3230/4000 [43:31<10:13,  1.25it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.25it/s]
overall:  81%|████████  | 3231/4000 [43:32<10:12,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  81%|████████  | 3232/4000 [43:32<10:09,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  81%|████████  | 3233/4000 [43:33<10:09,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  81%|████████  | 3234/4000 [43:34<10:07,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  81%|████████  | 3235/4000 [43:35<10:06,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  81%|████████  | 3236/4000 [43:36<10:03,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  81%|████████  | 3237/4000 [43:36<10:02,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  81%|████████  | 3238/4000 [43:37<10:00,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  81%|████████  | 3239/4000 [43:38<10:00,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]train loss: 0.8061524629592896
train acc: 0.7829999923706055

overall:  81%|████████  | 3240/4000 [43:39<10:00,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  81%|████████  | 3241/4000 [43:40<09:57,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  81%|████████  | 3242/4000 [43:40<09:57,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  81%|████████  | 3243/4000 [43:41<09:55,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  81%|████████  | 3244/4000 [43:42<09:54,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  81%|████████  | 3245/4000 [43:43<09:55,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  81%|████████  | 3246/4000 [43:44<09:55,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  81%|████████  | 3247/4000 [43:44<09:54,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  81%|████████  | 3248/4000 [43:45<09:54,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  81%|████████  | 3249/4000 [43:46<09:52,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]train loss: 0.8259333372116089
train acc: 0.7770000100135803

overall:  81%|████████▏ | 3250/4000 [43:47<09:51,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  81%|████████▏ | 3251/4000 [43:47<09:50,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  81%|████████▏ | 3252/4000 [43:48<09:50,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  81%|████████▏ | 3253/4000 [43:49<09:49,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  81%|████████▏ | 3254/4000 [43:50<09:47,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  81%|████████▏ | 3255/4000 [43:51<09:46,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  81%|████████▏ | 3256/4000 [43:51<09:25,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  81%|████████▏ | 3257/4000 [43:53<11:59,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.36s/it]
overall:  81%|████████▏ | 3258/4000 [43:54<11:17,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.02s/it]
overall:  81%|████████▏ | 3259/4000 [43:54<10:49,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]train loss: 0.8477370142936707
train acc: 0.7540485858917236

overall:  82%|████████▏ | 3260/4000 [43:55<10:29,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  82%|████████▏ | 3261/4000 [43:56<10:15,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  82%|████████▏ | 3262/4000 [43:57<10:03,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  82%|████████▏ | 3263/4000 [43:58<09:57,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  82%|████████▏ | 3264/4000 [43:58<09:53,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  82%|████████▏ | 3265/4000 [43:59<09:48,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  82%|████████▏ | 3266/4000 [44:00<09:44,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  82%|████████▏ | 3267/4000 [44:01<09:42,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  82%|████████▏ | 3268/4000 [44:01<09:38,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  82%|████████▏ | 3269/4000 [44:02<09:39,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]train loss: 0.8274914622306824
train acc: 0.7590000033378601

overall:  82%|████████▏ | 3270/4000 [44:03<09:36,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  82%|████████▏ | 3271/4000 [44:04<09:35,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  82%|████████▏ | 3272/4000 [44:05<09:34,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  82%|████████▏ | 3273/4000 [44:05<09:32,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  82%|████████▏ | 3274/4000 [44:06<09:30,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  82%|████████▏ | 3275/4000 [44:07<09:30,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  82%|████████▏ | 3276/4000 [44:08<09:33,  1.26it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.26it/s]
overall:  82%|████████▏ | 3277/4000 [44:09<09:30,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  82%|████████▏ | 3278/4000 [44:09<09:28,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  82%|████████▏ | 3279/4000 [44:10<09:28,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]train loss: 0.8200761675834656
train acc: 0.7799999713897705

overall:  82%|████████▏ | 3280/4000 [44:11<09:27,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  82%|████████▏ | 3281/4000 [44:12<09:27,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  82%|████████▏ | 3282/4000 [44:12<09:26,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  82%|████████▏ | 3283/4000 [44:13<09:26,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  82%|████████▏ | 3284/4000 [44:14<09:24,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  82%|████████▏ | 3285/4000 [44:15<09:21,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  82%|████████▏ | 3286/4000 [44:16<09:20,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  82%|████████▏ | 3287/4000 [44:16<09:21,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  82%|████████▏ | 3288/4000 [44:17<09:23,  1.26it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.26it/s]
overall:  82%|████████▏ | 3289/4000 [44:18<09:21,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]train loss: 0.7886567711830139
train acc: 0.7960000038146973

overall:  82%|████████▏ | 3290/4000 [44:19<09:20,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  82%|████████▏ | 3291/4000 [44:20<09:19,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  82%|████████▏ | 3292/4000 [44:20<09:16,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  82%|████████▏ | 3293/4000 [44:21<08:57,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  82%|████████▏ | 3294/4000 [44:23<11:40,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  82%|████████▏ | 3295/4000 [44:23<10:58,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:  82%|████████▏ | 3296/4000 [44:24<10:26,  1.12it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  82%|████████▏ | 3297/4000 [44:25<10:02,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  82%|████████▏ | 3298/4000 [44:26<09:48,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  82%|████████▏ | 3299/4000 [44:27<09:37,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]train loss: 0.8134851455688477
train acc: 0.7732793688774109

overall:  82%|████████▎ | 3300/4000 [44:27<09:28,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  83%|████████▎ | 3301/4000 [44:28<09:22,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  83%|████████▎ | 3302/4000 [44:29<09:20,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  83%|████████▎ | 3303/4000 [44:30<09:17,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  83%|████████▎ | 3304/4000 [44:30<09:13,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  83%|████████▎ | 3305/4000 [44:31<09:11,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  83%|████████▎ | 3306/4000 [44:32<09:14,  1.25it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.25it/s]
overall:  83%|████████▎ | 3307/4000 [44:33<09:10,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  83%|████████▎ | 3308/4000 [44:34<09:07,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  83%|████████▎ | 3309/4000 [44:34<09:05,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]train loss: 0.827468991279602
train acc: 0.7639999985694885

overall:  83%|████████▎ | 3310/4000 [44:35<09:04,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  83%|████████▎ | 3311/4000 [44:36<09:02,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  83%|████████▎ | 3312/4000 [44:37<09:00,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  83%|████████▎ | 3313/4000 [44:38<09:01,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  83%|████████▎ | 3314/4000 [44:38<09:01,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  83%|████████▎ | 3315/4000 [44:39<09:00,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  83%|████████▎ | 3316/4000 [44:40<08:58,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  83%|████████▎ | 3317/4000 [44:41<08:56,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  83%|████████▎ | 3318/4000 [44:42<08:58,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  83%|████████▎ | 3319/4000 [44:42<08:56,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]train loss: 0.8073908686637878
train acc: 0.7670000195503235

overall:  83%|████████▎ | 3320/4000 [44:43<08:55,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  83%|████████▎ | 3321/4000 [44:44<08:54,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  83%|████████▎ | 3322/4000 [44:45<08:53,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  83%|████████▎ | 3323/4000 [44:45<08:51,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  83%|████████▎ | 3324/4000 [44:46<08:51,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  83%|████████▎ | 3325/4000 [44:47<08:49,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  83%|████████▎ | 3326/4000 [44:48<08:51,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  83%|████████▎ | 3327/4000 [44:49<08:50,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  83%|████████▎ | 3328/4000 [44:49<08:49,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  83%|████████▎ | 3329/4000 [44:50<08:48,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]train loss: 0.829553484916687
train acc: 0.796558678150177

overall:  83%|████████▎ | 3330/4000 [44:51<08:28,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  83%|████████▎ | 3331/4000 [44:52<10:49,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.38s/it]
overall:  83%|████████▎ | 3332/4000 [44:53<10:11,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  83%|████████▎ | 3333/4000 [44:54<09:43,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  83%|████████▎ | 3334/4000 [44:55<09:25,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  83%|████████▎ | 3335/4000 [44:55<09:12,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  83%|████████▎ | 3336/4000 [44:56<09:03,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  83%|████████▎ | 3337/4000 [44:57<08:56,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  83%|████████▎ | 3338/4000 [44:58<08:51,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  83%|████████▎ | 3339/4000 [44:59<08:48,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]train loss: 0.8266547322273254
train acc: 0.7710000276565552

overall:  84%|████████▎ | 3340/4000 [44:59<08:45,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  84%|████████▎ | 3341/4000 [45:00<08:43,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  84%|████████▎ | 3342/4000 [45:01<08:41,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  84%|████████▎ | 3343/4000 [45:02<08:41,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  84%|████████▎ | 3344/4000 [45:03<08:40,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  84%|████████▎ | 3345/4000 [45:03<08:38,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  84%|████████▎ | 3346/4000 [45:04<08:35,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  84%|████████▎ | 3347/4000 [45:05<08:35,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  84%|████████▎ | 3348/4000 [45:06<08:33,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  84%|████████▎ | 3349/4000 [45:07<08:31,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]train loss: 0.8198239207267761
train acc: 0.7739999890327454

overall:  84%|████████▍ | 3350/4000 [45:07<08:31,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  84%|████████▍ | 3351/4000 [45:08<08:30,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  84%|████████▍ | 3352/4000 [45:09<08:29,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  84%|████████▍ | 3353/4000 [45:10<08:28,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:10,  1.27it/s]
overall:  84%|████████▍ | 3354/4000 [45:10<08:28,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  84%|████████▍ | 3355/4000 [45:11<08:27,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  84%|████████▍ | 3356/4000 [45:12<08:27,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  84%|████████▍ | 3357/4000 [45:13<08:26,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  84%|████████▍ | 3358/4000 [45:14<08:25,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  84%|████████▍ | 3359/4000 [45:14<08:25,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]train loss: 0.8079169988632202
train acc: 0.7730000019073486

overall:  84%|████████▍ | 3360/4000 [45:15<08:24,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  84%|████████▍ | 3361/4000 [45:16<08:24,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  84%|████████▍ | 3362/4000 [45:17<08:23,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  84%|████████▍ | 3363/4000 [45:18<08:23,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  84%|████████▍ | 3364/4000 [45:18<08:21,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  84%|████████▍ | 3365/4000 [45:19<08:20,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  84%|████████▍ | 3366/4000 [45:20<08:19,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  84%|████████▍ | 3367/4000 [45:21<08:00,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  84%|████████▍ | 3368/4000 [45:22<10:22,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.42s/it]
overall:  84%|████████▍ | 3369/4000 [45:23<09:45,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]train loss: 0.7983142733573914
train acc: 0.7864372730255127

overall:  84%|████████▍ | 3370/4000 [45:24<09:17,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  84%|████████▍ | 3371/4000 [45:25<08:59,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  84%|████████▍ | 3372/4000 [45:25<08:45,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  84%|████████▍ | 3373/4000 [45:26<08:38,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.20it/s]
overall:  84%|████████▍ | 3374/4000 [45:27<08:30,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.22it/s]
overall:  84%|████████▍ | 3375/4000 [45:28<08:24,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  84%|████████▍ | 3376/4000 [45:28<08:20,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  84%|████████▍ | 3377/4000 [45:29<08:17,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  84%|████████▍ | 3378/4000 [45:30<08:14,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  84%|████████▍ | 3379/4000 [45:31<08:12,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]train loss: 0.8207916021347046
train acc: 0.7630000114440918

overall:  84%|████████▍ | 3380/4000 [45:32<08:11,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  85%|████████▍ | 3381/4000 [45:32<08:08,  1.27it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.27it/s]
overall:  85%|████████▍ | 3382/4000 [45:33<08:08,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  85%|████████▍ | 3383/4000 [45:34<08:07,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  85%|████████▍ | 3384/4000 [45:35<08:05,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  85%|████████▍ | 3385/4000 [45:36<08:06,  1.26it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.26it/s]
overall:  85%|████████▍ | 3386/4000 [45:36<08:04,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  85%|████████▍ | 3387/4000 [45:37<08:03,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  85%|████████▍ | 3388/4000 [45:38<08:01,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  85%|████████▍ | 3389/4000 [45:39<08:00,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]train loss: 0.8318028450012207
train acc: 0.7799999713897705

overall:  85%|████████▍ | 3390/4000 [45:39<07:59,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  85%|████████▍ | 3391/4000 [45:40<07:59,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  85%|████████▍ | 3392/4000 [45:41<07:59,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  85%|████████▍ | 3393/4000 [45:42<07:57,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  85%|████████▍ | 3394/4000 [45:43<07:56,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  85%|████████▍ | 3395/4000 [45:43<07:56,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  85%|████████▍ | 3396/4000 [45:44<07:55,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  85%|████████▍ | 3397/4000 [45:45<08:19,  1.21it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.21it/s]
overall:  85%|████████▍ | 3398/4000 [45:46<08:11,  1.23it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.23it/s]
overall:  85%|████████▍ | 3399/4000 [45:47<08:04,  1.24it/s]

train:  86%|████████▋ | 32/37 [00:26<00:04,  1.24it/s]train loss: 0.7976760864257812
train acc: 0.777999997138977

overall:  85%|████████▌ | 3400/4000 [45:48<08:00,  1.25it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.25it/s]
overall:  85%|████████▌ | 3401/4000 [45:48<07:55,  1.26it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.26it/s]
overall:  85%|████████▌ | 3402/4000 [45:49<07:53,  1.26it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.26it/s]
overall:  85%|████████▌ | 3403/4000 [45:50<07:52,  1.26it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.26it/s]
overall:  85%|████████▌ | 3404/4000 [45:51<07:34,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  85%|████████▌ | 3405/4000 [45:52<09:42,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:50,  1.39s/it]
overall:  85%|████████▌ | 3406/4000 [45:53<09:09,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.04s/it]
overall:  85%|████████▌ | 3407/4000 [45:54<08:42,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  85%|████████▌ | 3408/4000 [45:54<08:24,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  85%|████████▌ | 3409/4000 [45:55<08:12,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]train loss: 0.8052770495414734
train acc: 0.7864372730255127

overall:  85%|████████▌ | 3410/4000 [45:56<08:03,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  85%|████████▌ | 3411/4000 [45:57<07:56,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.24it/s]
overall:  85%|████████▌ | 3412/4000 [45:58<07:52,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  85%|████████▌ | 3413/4000 [45:58<07:50,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  85%|████████▌ | 3414/4000 [45:59<07:47,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  85%|████████▌ | 3415/4000 [46:00<07:43,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  85%|████████▌ | 3416/4000 [46:01<07:41,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  85%|████████▌ | 3417/4000 [46:01<07:41,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  85%|████████▌ | 3418/4000 [46:02<07:41,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  85%|████████▌ | 3419/4000 [46:03<07:38,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]train loss: 0.7692280411720276
train acc: 0.7940000295639038

overall:  86%|████████▌ | 3420/4000 [46:04<07:36,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  86%|████████▌ | 3421/4000 [46:05<07:34,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  86%|████████▌ | 3422/4000 [46:05<07:33,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  86%|████████▌ | 3423/4000 [46:06<07:34,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  86%|████████▌ | 3424/4000 [46:07<07:32,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  86%|████████▌ | 3425/4000 [46:08<07:33,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  86%|████████▌ | 3426/4000 [46:09<07:31,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  86%|████████▌ | 3427/4000 [46:09<07:30,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  86%|████████▌ | 3428/4000 [46:10<07:30,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  86%|████████▌ | 3429/4000 [46:11<07:30,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]train loss: 0.8507823944091797
train acc: 0.7639999985694885

overall:  86%|████████▌ | 3430/4000 [46:12<07:30,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  86%|████████▌ | 3431/4000 [46:13<07:28,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  86%|████████▌ | 3432/4000 [46:13<07:27,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  86%|████████▌ | 3433/4000 [46:14<07:26,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  86%|████████▌ | 3434/4000 [46:15<07:24,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  86%|████████▌ | 3435/4000 [46:16<07:23,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  86%|████████▌ | 3436/4000 [46:16<07:23,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  86%|████████▌ | 3437/4000 [46:17<07:23,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  86%|████████▌ | 3438/4000 [46:18<07:23,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  86%|████████▌ | 3439/4000 [46:19<07:21,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]train loss: 0.827044665813446
train acc: 0.7770000100135803

overall:  86%|████████▌ | 3440/4000 [46:20<07:21,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  86%|████████▌ | 3441/4000 [46:20<07:05,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  86%|████████▌ | 3442/4000 [46:22<09:01,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:48,  1.36s/it]
overall:  86%|████████▌ | 3443/4000 [46:23<08:31,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  86%|████████▌ | 3444/4000 [46:23<08:09,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  86%|████████▌ | 3445/4000 [46:24<07:51,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]
overall:  86%|████████▌ | 3446/4000 [46:25<07:39,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.20it/s]
overall:  86%|████████▌ | 3447/4000 [46:26<07:32,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  86%|████████▌ | 3448/4000 [46:26<07:25,  1.24it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.24it/s]
overall:  86%|████████▌ | 3449/4000 [46:27<07:21,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]train loss: 0.8485985994338989
train acc: 0.7712550759315491

overall:  86%|████████▋ | 3450/4000 [46:28<07:18,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  86%|████████▋ | 3451/4000 [46:29<07:15,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  86%|████████▋ | 3452/4000 [46:30<07:14,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  86%|████████▋ | 3453/4000 [46:30<07:13,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  86%|████████▋ | 3454/4000 [46:31<07:12,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  86%|████████▋ | 3455/4000 [46:32<07:12,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  86%|████████▋ | 3456/4000 [46:33<07:10,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  86%|████████▋ | 3457/4000 [46:34<07:08,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  86%|████████▋ | 3458/4000 [46:34<07:06,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  86%|████████▋ | 3459/4000 [46:35<07:06,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]train loss: 0.8058129549026489
train acc: 0.781000018119812

overall:  86%|████████▋ | 3460/4000 [46:36<07:05,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  87%|████████▋ | 3461/4000 [46:37<07:04,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  87%|████████▋ | 3462/4000 [46:38<07:03,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  87%|████████▋ | 3463/4000 [46:38<07:02,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  87%|████████▋ | 3464/4000 [46:39<07:02,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  87%|████████▋ | 3465/4000 [46:40<07:01,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  87%|████████▋ | 3466/4000 [46:41<07:01,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  87%|████████▋ | 3467/4000 [46:41<07:01,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  87%|████████▋ | 3468/4000 [46:42<07:00,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  87%|████████▋ | 3469/4000 [46:43<06:58,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]train loss: 0.8289099931716919
train acc: 0.7599999904632568

overall:  87%|████████▋ | 3470/4000 [46:44<06:57,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  87%|████████▋ | 3471/4000 [46:45<06:56,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  87%|████████▋ | 3472/4000 [46:45<06:55,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  87%|████████▋ | 3473/4000 [46:46<06:54,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  87%|████████▋ | 3474/4000 [46:47<06:54,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  87%|████████▋ | 3475/4000 [46:48<06:53,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  87%|████████▋ | 3476/4000 [46:49<06:53,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  87%|████████▋ | 3477/4000 [46:49<06:52,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall:  87%|████████▋ | 3478/4000 [46:50<06:37,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  87%|████████▋ | 3479/4000 [46:52<08:38,  1.00it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]train loss: 0.7815964221954346
train acc: 0.7884615659713745

overall:  87%|████████▋ | 3480/4000 [46:52<08:06,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:  87%|████████▋ | 3481/4000 [46:53<07:44,  1.12it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.06it/s]
overall:  87%|████████▋ | 3482/4000 [46:54<07:25,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  87%|████████▋ | 3483/4000 [46:55<07:13,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  87%|████████▋ | 3484/4000 [46:56<07:04,  1.21it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  87%|████████▋ | 3485/4000 [46:56<06:58,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  87%|████████▋ | 3486/4000 [46:57<06:54,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  87%|████████▋ | 3487/4000 [46:58<06:49,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  87%|████████▋ | 3488/4000 [46:59<06:48,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  87%|████████▋ | 3489/4000 [46:59<06:45,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]train loss: 0.8428307771682739
train acc: 0.7760000228881836

overall:  87%|████████▋ | 3490/4000 [47:00<06:44,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  87%|████████▋ | 3491/4000 [47:01<06:43,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  87%|████████▋ | 3492/4000 [47:02<06:43,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  87%|████████▋ | 3493/4000 [47:03<06:43,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  87%|████████▋ | 3494/4000 [47:03<06:40,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  87%|████████▋ | 3495/4000 [47:04<06:38,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  87%|████████▋ | 3496/4000 [47:05<06:37,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  87%|████████▋ | 3497/4000 [47:06<06:36,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  87%|████████▋ | 3498/4000 [47:07<06:34,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  87%|████████▋ | 3499/4000 [47:07<06:33,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]train loss: 0.7673825025558472
train acc: 0.8050000071525574

overall:  88%|████████▊ | 3500/4000 [47:08<06:33,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  88%|████████▊ | 3501/4000 [47:09<06:32,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  88%|████████▊ | 3502/4000 [47:10<06:31,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  88%|████████▊ | 3503/4000 [47:11<06:31,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  88%|████████▊ | 3504/4000 [47:11<06:31,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  88%|████████▊ | 3505/4000 [47:12<06:31,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  88%|████████▊ | 3506/4000 [47:13<06:30,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  88%|████████▊ | 3507/4000 [47:14<06:28,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  88%|████████▊ | 3508/4000 [47:14<06:26,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  88%|████████▊ | 3509/4000 [47:15<06:25,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]train loss: 0.8361417055130005
train acc: 0.7549999952316284

overall:  88%|████████▊ | 3510/4000 [47:16<06:25,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  88%|████████▊ | 3511/4000 [47:17<06:24,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  88%|████████▊ | 3512/4000 [47:18<06:25,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  88%|████████▊ | 3513/4000 [47:18<06:23,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  88%|████████▊ | 3514/4000 [47:19<06:22,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  88%|████████▊ | 3515/4000 [47:20<06:08,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  88%|████████▊ | 3516/4000 [47:21<07:56,  1.02it/s]

train:   3%|▎         | 1/37 [00:01<00:50,  1.41s/it]
overall:  88%|████████▊ | 3517/4000 [47:22<07:28,  1.08it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  88%|████████▊ | 3518/4000 [47:23<07:06,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  88%|████████▊ | 3519/4000 [47:24<06:50,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]train loss: 0.7767949104309082
train acc: 0.7783401012420654

overall:  88%|████████▊ | 3520/4000 [47:25<06:41,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  88%|████████▊ | 3521/4000 [47:25<06:33,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  88%|████████▊ | 3522/4000 [47:26<06:27,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  88%|████████▊ | 3523/4000 [47:27<06:23,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  88%|████████▊ | 3524/4000 [47:28<06:20,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  88%|████████▊ | 3525/4000 [47:28<06:18,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  88%|████████▊ | 3526/4000 [47:29<06:17,  1.25it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.25it/s]
overall:  88%|████████▊ | 3527/4000 [47:30<06:15,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  88%|████████▊ | 3528/4000 [47:31<06:14,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  88%|████████▊ | 3529/4000 [47:32<06:15,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]train loss: 0.7847487330436707
train acc: 0.7919999957084656

overall:  88%|████████▊ | 3530/4000 [47:32<06:12,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  88%|████████▊ | 3531/4000 [47:33<06:10,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  88%|████████▊ | 3532/4000 [47:34<06:09,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  88%|████████▊ | 3533/4000 [47:35<06:07,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  88%|████████▊ | 3534/4000 [47:36<06:06,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  88%|████████▊ | 3535/4000 [47:36<06:06,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  88%|████████▊ | 3536/4000 [47:37<06:05,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  88%|████████▊ | 3537/4000 [47:38<06:03,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  88%|████████▊ | 3538/4000 [47:39<06:03,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  88%|████████▊ | 3539/4000 [47:40<06:02,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]train loss: 0.8238064050674438
train acc: 0.7649999856948853

overall:  88%|████████▊ | 3540/4000 [47:40<06:03,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  89%|████████▊ | 3541/4000 [47:41<06:01,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  89%|████████▊ | 3542/4000 [47:42<06:00,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  89%|████████▊ | 3543/4000 [47:43<06:00,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  89%|████████▊ | 3544/4000 [47:43<05:59,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  89%|████████▊ | 3545/4000 [47:44<05:57,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  89%|████████▊ | 3546/4000 [47:45<05:57,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  89%|████████▊ | 3547/4000 [47:46<05:57,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  89%|████████▊ | 3548/4000 [47:47<05:56,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  89%|████████▊ | 3549/4000 [47:47<05:55,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]train loss: 0.8441452980041504
train acc: 0.7699999809265137

overall:  89%|████████▉ | 3550/4000 [47:48<05:55,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  89%|████████▉ | 3551/4000 [47:49<05:53,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  89%|████████▉ | 3552/4000 [47:50<05:41,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  89%|████████▉ | 3553/4000 [47:51<07:26,  1.00it/s]

train:   3%|▎         | 1/37 [00:01<00:52,  1.45s/it]
overall:  89%|████████▉ | 3554/4000 [47:52<06:57,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:  89%|████████▉ | 3555/4000 [47:53<06:36,  1.12it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  89%|████████▉ | 3556/4000 [47:54<06:21,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  89%|████████▉ | 3557/4000 [47:54<06:11,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  89%|████████▉ | 3558/4000 [47:55<06:03,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  89%|████████▉ | 3559/4000 [47:56<05:58,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]train loss: 0.7913724184036255
train acc: 0.78340083360672

overall:  89%|████████▉ | 3560/4000 [47:57<05:54,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  89%|████████▉ | 3561/4000 [47:58<05:52,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  89%|████████▉ | 3562/4000 [47:58<05:49,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  89%|████████▉ | 3563/4000 [47:59<05:47,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  89%|████████▉ | 3564/4000 [48:00<05:45,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  89%|████████▉ | 3565/4000 [48:01<05:46,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  89%|████████▉ | 3566/4000 [48:02<05:46,  1.25it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.25it/s]
overall:  89%|████████▉ | 3567/4000 [48:02<05:44,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  89%|████████▉ | 3568/4000 [48:03<05:41,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  89%|████████▉ | 3569/4000 [48:04<05:40,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]train loss: 0.7715685963630676
train acc: 0.7960000038146973

overall:  89%|████████▉ | 3570/4000 [48:05<05:39,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.27it/s]
overall:  89%|████████▉ | 3571/4000 [48:05<05:38,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  89%|████████▉ | 3572/4000 [48:06<05:36,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  89%|████████▉ | 3573/4000 [48:07<05:36,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  89%|████████▉ | 3574/4000 [48:08<05:35,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  89%|████████▉ | 3575/4000 [48:09<05:34,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  89%|████████▉ | 3576/4000 [48:09<05:34,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  89%|████████▉ | 3577/4000 [48:10<05:34,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  89%|████████▉ | 3578/4000 [48:11<05:33,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  89%|████████▉ | 3579/4000 [48:12<05:31,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]train loss: 0.8358579874038696
train acc: 0.7680000066757202

overall:  90%|████████▉ | 3580/4000 [48:13<05:30,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  90%|████████▉ | 3581/4000 [48:13<05:29,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  90%|████████▉ | 3582/4000 [48:14<05:28,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  90%|████████▉ | 3583/4000 [48:15<05:27,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  90%|████████▉ | 3584/4000 [48:16<05:26,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  90%|████████▉ | 3585/4000 [48:16<05:26,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  90%|████████▉ | 3586/4000 [48:17<05:26,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  90%|████████▉ | 3587/4000 [48:18<05:25,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  90%|████████▉ | 3588/4000 [48:19<05:23,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  90%|████████▉ | 3589/4000 [48:20<05:12,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]train loss: 0.8365710377693176
train acc: 0.7661943435668945

overall:  90%|████████▉ | 3590/4000 [48:21<06:47,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  90%|████████▉ | 3591/4000 [48:22<06:21,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:  90%|████████▉ | 3592/4000 [48:23<06:02,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  90%|████████▉ | 3593/4000 [48:23<05:49,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  90%|████████▉ | 3594/4000 [48:24<05:40,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  90%|████████▉ | 3595/4000 [48:25<05:32,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  90%|████████▉ | 3596/4000 [48:26<05:27,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  90%|████████▉ | 3597/4000 [48:27<05:25,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  90%|████████▉ | 3598/4000 [48:27<05:22,  1.24it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.24it/s]
overall:  90%|████████▉ | 3599/4000 [48:28<05:20,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]train loss: 0.8166624307632446
train acc: 0.7889999747276306

overall:  90%|█████████ | 3600/4000 [48:29<05:18,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  90%|█████████ | 3601/4000 [48:30<05:16,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  90%|█████████ | 3602/4000 [48:31<05:17,  1.25it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.25it/s]
overall:  90%|█████████ | 3603/4000 [48:31<05:17,  1.25it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.25it/s]
overall:  90%|█████████ | 3604/4000 [48:32<05:14,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  90%|█████████ | 3605/4000 [48:33<05:12,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  90%|█████████ | 3606/4000 [48:34<05:10,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  90%|█████████ | 3607/4000 [48:35<05:10,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  90%|█████████ | 3608/4000 [48:35<05:08,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  90%|█████████ | 3609/4000 [48:36<05:08,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]train loss: 0.7753710746765137
train acc: 0.7900000214576721

overall:  90%|█████████ | 3610/4000 [48:37<05:12,  1.25it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.25it/s]
overall:  90%|█████████ | 3611/4000 [48:38<05:09,  1.26it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.26it/s]
overall:  90%|█████████ | 3612/4000 [48:38<05:07,  1.26it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.26it/s]
overall:  90%|█████████ | 3613/4000 [48:39<05:06,  1.26it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.26it/s]
overall:  90%|█████████ | 3614/4000 [48:40<05:07,  1.25it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.25it/s]
overall:  90%|█████████ | 3615/4000 [48:41<05:05,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  90%|█████████ | 3616/4000 [48:42<05:03,  1.26it/s]

train:  73%|███████▎  | 27/37 [00:22<00:07,  1.26it/s]
overall:  90%|█████████ | 3617/4000 [48:42<05:02,  1.26it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.26it/s]
overall:  90%|█████████ | 3618/4000 [48:43<05:01,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  90%|█████████ | 3619/4000 [48:44<05:00,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]train loss: 0.8296058773994446
train acc: 0.7609999775886536

overall:  90%|█████████ | 3620/4000 [48:45<05:00,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  91%|█████████ | 3621/4000 [48:46<04:58,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  91%|█████████ | 3622/4000 [48:46<04:59,  1.26it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.26it/s]
overall:  91%|█████████ | 3623/4000 [48:47<04:57,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  91%|█████████ | 3624/4000 [48:48<04:56,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  91%|█████████ | 3625/4000 [48:49<04:55,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  91%|█████████ | 3626/4000 [48:49<04:44,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  91%|█████████ | 3627/4000 [48:51<06:11,  1.00it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  91%|█████████ | 3628/4000 [48:52<05:47,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.06s/it]
overall:  91%|█████████ | 3629/4000 [48:53<05:30,  1.12it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]train loss: 0.8015843629837036
train acc: 0.7742915153503418

overall:  91%|█████████ | 3630/4000 [48:53<05:17,  1.16it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  91%|█████████ | 3631/4000 [48:54<05:09,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  91%|█████████ | 3632/4000 [48:55<05:02,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  91%|█████████ | 3633/4000 [48:56<04:57,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  91%|█████████ | 3634/4000 [48:57<04:55,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  91%|█████████ | 3635/4000 [48:57<04:52,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  91%|█████████ | 3636/4000 [48:58<04:50,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  91%|█████████ | 3637/4000 [48:59<04:48,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  91%|█████████ | 3638/4000 [49:00<04:46,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  91%|█████████ | 3639/4000 [49:00<04:47,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]train loss: 0.8055349588394165
train acc: 0.777999997138977

overall:  91%|█████████ | 3640/4000 [49:01<04:46,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  91%|█████████ | 3641/4000 [49:02<04:44,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  91%|█████████ | 3642/4000 [49:03<04:43,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  91%|█████████ | 3643/4000 [49:04<04:42,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  91%|█████████ | 3644/4000 [49:04<04:41,  1.26it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.26it/s]
overall:  91%|█████████ | 3645/4000 [49:05<04:40,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  91%|█████████ | 3646/4000 [49:06<04:39,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  91%|█████████ | 3647/4000 [49:07<04:39,  1.26it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.26it/s]
overall:  91%|█████████ | 3648/4000 [49:08<04:37,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  91%|█████████ | 3649/4000 [49:08<04:37,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]train loss: 0.8043460845947266
train acc: 0.7839999794960022

overall:  91%|█████████▏| 3650/4000 [49:09<04:36,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  91%|█████████▏| 3651/4000 [49:10<04:36,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  91%|█████████▏| 3652/4000 [49:11<04:35,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  91%|█████████▏| 3653/4000 [49:12<04:33,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  91%|█████████▏| 3654/4000 [49:12<04:33,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  91%|█████████▏| 3655/4000 [49:13<04:32,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  91%|█████████▏| 3656/4000 [49:14<04:31,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  91%|█████████▏| 3657/4000 [49:15<04:30,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  91%|█████████▏| 3658/4000 [49:15<04:29,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  91%|█████████▏| 3659/4000 [49:16<04:29,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]train loss: 0.8181894421577454
train acc: 0.7720000147819519

overall:  92%|█████████▏| 3660/4000 [49:17<04:27,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  92%|█████████▏| 3661/4000 [49:18<04:27,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  92%|█████████▏| 3662/4000 [49:19<04:26,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  92%|█████████▏| 3663/4000 [49:19<04:15,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  92%|█████████▏| 3664/4000 [49:21<05:25,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  92%|█████████▏| 3665/4000 [49:22<05:06,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.03s/it]
overall:  92%|█████████▏| 3666/4000 [49:22<04:53,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  92%|█████████▏| 3667/4000 [49:23<04:43,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]
overall:  92%|█████████▏| 3668/4000 [49:24<04:36,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  92%|█████████▏| 3669/4000 [49:25<04:32,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]train loss: 0.8020566701889038
train acc: 0.7793522477149963

overall:  92%|█████████▏| 3670/4000 [49:26<04:28,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  92%|█████████▏| 3671/4000 [49:26<04:24,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  92%|█████████▏| 3672/4000 [49:27<04:22,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  92%|█████████▏| 3673/4000 [49:28<04:19,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  92%|█████████▏| 3674/4000 [49:29<04:18,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  92%|█████████▏| 3675/4000 [49:29<04:18,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  92%|█████████▏| 3676/4000 [49:30<04:17,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  92%|█████████▏| 3677/4000 [49:31<04:16,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  92%|█████████▏| 3678/4000 [49:32<04:14,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  92%|█████████▏| 3679/4000 [49:33<04:13,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]train loss: 0.8272916674613953
train acc: 0.7670000195503235

overall:  92%|█████████▏| 3680/4000 [49:33<04:13,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  92%|█████████▏| 3681/4000 [49:34<04:13,  1.26it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.26it/s]
overall:  92%|█████████▏| 3682/4000 [49:35<04:11,  1.26it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.26it/s]
overall:  92%|█████████▏| 3683/4000 [49:36<04:10,  1.26it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.26it/s]
overall:  92%|█████████▏| 3684/4000 [49:37<04:09,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  92%|█████████▏| 3685/4000 [49:37<04:08,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  92%|█████████▏| 3686/4000 [49:38<04:07,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  92%|█████████▏| 3687/4000 [49:39<04:07,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  92%|█████████▏| 3688/4000 [49:40<04:06,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  92%|█████████▏| 3689/4000 [49:41<04:06,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]train loss: 0.7692688703536987
train acc: 0.7910000085830688

overall:  92%|█████████▏| 3690/4000 [49:41<04:04,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  92%|█████████▏| 3691/4000 [49:42<04:04,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  92%|█████████▏| 3692/4000 [49:43<04:02,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  92%|█████████▏| 3693/4000 [49:44<04:03,  1.26it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.26it/s]
overall:  92%|█████████▏| 3694/4000 [49:44<04:01,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  92%|█████████▏| 3695/4000 [49:45<04:00,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  92%|█████████▏| 3696/4000 [49:46<03:59,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  92%|█████████▏| 3697/4000 [49:47<03:58,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  92%|█████████▏| 3698/4000 [49:48<03:57,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  92%|█████████▏| 3699/4000 [49:48<03:57,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]train loss: 0.8351153135299683
train acc: 0.7702429294586182

overall:  92%|█████████▎| 3700/4000 [49:49<03:47,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  93%|█████████▎| 3701/4000 [49:51<04:58,  1.00it/s]

train:   3%|▎         | 1/37 [00:01<00:52,  1.45s/it]
overall:  93%|█████████▎| 3702/4000 [49:51<04:38,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:  93%|█████████▎| 3703/4000 [49:52<04:23,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  93%|█████████▎| 3704/4000 [49:53<04:13,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  93%|█████████▎| 3705/4000 [49:54<04:07,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  93%|█████████▎| 3706/4000 [49:55<04:01,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  93%|█████████▎| 3707/4000 [49:55<03:57,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  93%|█████████▎| 3708/4000 [49:56<03:54,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  93%|█████████▎| 3709/4000 [49:57<03:53,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]train loss: 0.8036273121833801
train acc: 0.7829999923706055

overall:  93%|█████████▎| 3710/4000 [49:58<03:51,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  93%|█████████▎| 3711/4000 [49:59<03:50,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  93%|█████████▎| 3712/4000 [49:59<03:48,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  93%|█████████▎| 3713/4000 [50:00<03:47,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  93%|█████████▎| 3714/4000 [50:01<03:46,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  93%|█████████▎| 3715/4000 [50:02<03:45,  1.27it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.27it/s]
overall:  93%|█████████▎| 3716/4000 [50:02<03:44,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  93%|█████████▎| 3717/4000 [50:03<03:42,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  93%|█████████▎| 3718/4000 [50:04<03:42,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  93%|█████████▎| 3719/4000 [50:05<03:41,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]train loss: 0.7990056276321411
train acc: 0.7839999794960022

overall:  93%|█████████▎| 3720/4000 [50:06<03:40,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  93%|█████████▎| 3721/4000 [50:06<03:40,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  93%|█████████▎| 3722/4000 [50:07<03:39,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]
overall:  93%|█████████▎| 3723/4000 [50:08<03:38,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  93%|█████████▎| 3724/4000 [50:09<03:37,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  93%|█████████▎| 3725/4000 [50:10<03:37,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  93%|█████████▎| 3726/4000 [50:10<03:36,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  93%|█████████▎| 3727/4000 [50:11<03:35,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  93%|█████████▎| 3728/4000 [50:12<03:34,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  93%|█████████▎| 3729/4000 [50:13<03:34,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]train loss: 0.8096538782119751
train acc: 0.7749999761581421

overall:  93%|█████████▎| 3730/4000 [50:14<03:33,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  93%|█████████▎| 3731/4000 [50:14<03:32,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  93%|█████████▎| 3732/4000 [50:15<03:31,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  93%|█████████▎| 3733/4000 [50:16<03:30,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  93%|█████████▎| 3734/4000 [50:17<03:29,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  93%|█████████▎| 3735/4000 [50:17<03:28,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  93%|█████████▎| 3736/4000 [50:18<03:28,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  93%|█████████▎| 3737/4000 [50:19<03:19,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  93%|█████████▎| 3738/4000 [50:20<04:20,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  93%|█████████▎| 3739/4000 [50:21<04:03,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.06s/it]train loss: 0.8086700439453125
train acc: 0.7753036618232727

overall:  94%|█████████▎| 3740/4000 [50:22<03:50,  1.13it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  94%|█████████▎| 3741/4000 [50:23<03:41,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  94%|█████████▎| 3742/4000 [50:24<03:35,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  94%|█████████▎| 3743/4000 [50:24<03:30,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  94%|█████████▎| 3744/4000 [50:25<03:27,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  94%|█████████▎| 3745/4000 [50:26<03:26,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  94%|█████████▎| 3746/4000 [50:27<03:23,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  94%|█████████▎| 3747/4000 [50:28<03:21,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  94%|█████████▎| 3748/4000 [50:28<03:20,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  94%|█████████▎| 3749/4000 [50:29<03:19,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]train loss: 0.8020012974739075
train acc: 0.777999997138977

overall:  94%|█████████▍| 3750/4000 [50:30<03:19,  1.25it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.25it/s]
overall:  94%|█████████▍| 3751/4000 [50:31<03:18,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  94%|█████████▍| 3752/4000 [50:32<03:16,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  94%|█████████▍| 3753/4000 [50:32<03:15,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  94%|█████████▍| 3754/4000 [50:33<03:14,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  94%|█████████▍| 3755/4000 [50:34<03:13,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  94%|█████████▍| 3756/4000 [50:35<03:11,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  94%|█████████▍| 3757/4000 [50:35<03:11,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  94%|█████████▍| 3758/4000 [50:36<03:10,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  94%|█████████▍| 3759/4000 [50:37<03:10,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:18<00:11,  1.27it/s]train loss: 0.8060708045959473
train acc: 0.7770000100135803

overall:  94%|█████████▍| 3760/4000 [50:38<03:09,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  94%|█████████▍| 3761/4000 [50:39<03:08,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  94%|█████████▍| 3762/4000 [50:39<03:08,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  94%|█████████▍| 3763/4000 [50:40<03:07,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  94%|█████████▍| 3764/4000 [50:41<03:05,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  94%|█████████▍| 3765/4000 [50:42<03:05,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  94%|█████████▍| 3766/4000 [50:43<03:04,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  94%|█████████▍| 3767/4000 [50:43<03:03,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  94%|█████████▍| 3768/4000 [50:44<03:02,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  94%|█████████▍| 3769/4000 [50:45<03:01,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]train loss: 0.8232555389404297
train acc: 0.7710000276565552

overall:  94%|█████████▍| 3770/4000 [50:46<03:01,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  94%|█████████▍| 3771/4000 [50:47<03:00,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  94%|█████████▍| 3772/4000 [50:47<02:59,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  94%|█████████▍| 3773/4000 [50:48<02:59,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  94%|█████████▍| 3774/4000 [50:49<02:51,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  94%|█████████▍| 3775/4000 [50:50<03:43,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  94%|█████████▍| 3776/4000 [50:51<03:28,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  94%|█████████▍| 3777/4000 [50:52<03:17,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  94%|█████████▍| 3778/4000 [50:53<03:10,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  94%|█████████▍| 3779/4000 [50:53<03:04,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]train loss: 0.791384220123291
train acc: 0.78340083360672

overall:  94%|█████████▍| 3780/4000 [50:54<03:01,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  95%|█████████▍| 3781/4000 [50:55<02:58,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  95%|█████████▍| 3782/4000 [50:56<02:56,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  95%|█████████▍| 3783/4000 [50:57<02:53,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  95%|█████████▍| 3784/4000 [50:57<02:52,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  95%|█████████▍| 3785/4000 [50:58<02:50,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  95%|█████████▍| 3786/4000 [50:59<02:49,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  95%|█████████▍| 3787/4000 [51:00<02:49,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  95%|█████████▍| 3788/4000 [51:01<02:48,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  95%|█████████▍| 3789/4000 [51:01<02:47,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]train loss: 0.8160454630851746
train acc: 0.7730000019073486

overall:  95%|█████████▍| 3790/4000 [51:02<02:45,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  95%|█████████▍| 3791/4000 [51:03<02:44,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  95%|█████████▍| 3792/4000 [51:04<02:43,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  95%|█████████▍| 3793/4000 [51:05<02:43,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  95%|█████████▍| 3794/4000 [51:05<02:42,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  95%|█████████▍| 3795/4000 [51:06<02:41,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  95%|█████████▍| 3796/4000 [51:07<02:40,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  95%|█████████▍| 3797/4000 [51:08<02:39,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  95%|█████████▍| 3798/4000 [51:08<02:38,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  95%|█████████▍| 3799/4000 [51:09<02:38,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]train loss: 0.7938432097434998
train acc: 0.7919999957084656

overall:  95%|█████████▌| 3800/4000 [51:10<02:37,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  95%|█████████▌| 3801/4000 [51:11<02:36,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  95%|█████████▌| 3802/4000 [51:12<02:35,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  95%|█████████▌| 3803/4000 [51:12<02:34,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  95%|█████████▌| 3804/4000 [51:13<02:34,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  95%|█████████▌| 3805/4000 [51:14<02:33,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall:  95%|█████████▌| 3806/4000 [51:15<02:33,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  95%|█████████▌| 3807/4000 [51:16<02:31,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  95%|█████████▌| 3808/4000 [51:16<02:31,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  95%|█████████▌| 3809/4000 [51:17<02:30,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]train loss: 0.8238089680671692
train acc: 0.7590000033378601

overall:  95%|█████████▌| 3810/4000 [51:18<02:29,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  95%|█████████▌| 3811/4000 [51:19<02:23,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  95%|█████████▌| 3812/4000 [51:20<03:06,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.43s/it]
overall:  95%|█████████▌| 3813/4000 [51:21<02:54,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:37,  1.06s/it]
overall:  95%|█████████▌| 3814/4000 [51:22<02:45,  1.12it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  95%|█████████▌| 3815/4000 [51:23<02:38,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]
overall:  95%|█████████▌| 3816/4000 [51:23<02:34,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  95%|█████████▌| 3817/4000 [51:24<02:30,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  95%|█████████▌| 3818/4000 [51:25<02:27,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  95%|█████████▌| 3819/4000 [51:26<02:25,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]train loss: 0.7895665168762207
train acc: 0.7844129800796509

overall:  96%|█████████▌| 3820/4000 [51:26<02:24,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  96%|█████████▌| 3821/4000 [51:27<02:22,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  96%|█████████▌| 3822/4000 [51:28<02:21,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  96%|█████████▌| 3823/4000 [51:29<02:19,  1.27it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.27it/s]
overall:  96%|█████████▌| 3824/4000 [51:30<02:19,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:18,  1.26it/s]
overall:  96%|█████████▌| 3825/4000 [51:30<02:18,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  96%|█████████▌| 3826/4000 [51:31<02:17,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  96%|█████████▌| 3827/4000 [51:32<02:16,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  96%|█████████▌| 3828/4000 [51:33<02:16,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]
overall:  96%|█████████▌| 3829/4000 [51:34<02:14,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]train loss: 0.7971316576004028
train acc: 0.7829999923706055

overall:  96%|█████████▌| 3830/4000 [51:34<02:14,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  96%|█████████▌| 3831/4000 [51:35<02:13,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  96%|█████████▌| 3832/4000 [51:36<02:12,  1.26it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  96%|█████████▌| 3833/4000 [51:37<02:11,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  96%|█████████▌| 3834/4000 [51:38<02:11,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  96%|█████████▌| 3835/4000 [51:38<02:10,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  96%|█████████▌| 3836/4000 [51:39<02:09,  1.27it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.27it/s]
overall:  96%|█████████▌| 3837/4000 [51:40<02:15,  1.20it/s]

train:  70%|███████   | 26/37 [00:21<00:09,  1.20it/s]
overall:  96%|█████████▌| 3838/4000 [51:41<02:12,  1.22it/s]

train:  73%|███████▎  | 27/37 [00:22<00:08,  1.22it/s]
overall:  96%|█████████▌| 3839/4000 [51:42<02:10,  1.24it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.24it/s]train loss: 0.8372176885604858
train acc: 0.7649999856948853

overall:  96%|█████████▌| 3840/4000 [51:42<02:08,  1.24it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.24it/s]
overall:  96%|█████████▌| 3841/4000 [51:43<02:07,  1.25it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.25it/s]
overall:  96%|█████████▌| 3842/4000 [51:44<02:05,  1.26it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.26it/s]
overall:  96%|█████████▌| 3843/4000 [51:45<02:04,  1.26it/s]

train:  86%|████████▋ | 32/37 [00:26<00:03,  1.26it/s]
overall:  96%|█████████▌| 3844/4000 [51:46<02:03,  1.26it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.26it/s]
overall:  96%|█████████▌| 3845/4000 [51:46<02:02,  1.26it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.26it/s]
overall:  96%|█████████▌| 3846/4000 [51:47<02:01,  1.26it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.26it/s]
overall:  96%|█████████▌| 3847/4000 [51:48<02:00,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  96%|█████████▌| 3848/4000 [51:49<01:55,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.23it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  96%|█████████▌| 3849/4000 [51:50<02:29,  1.01it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.42s/it]train loss: 0.7948693633079529
train acc: 0.7864372730255127

overall:  96%|█████████▋| 3850/4000 [51:51<02:19,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  96%|█████████▋| 3851/4000 [51:52<02:12,  1.13it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]
overall:  96%|█████████▋| 3852/4000 [51:52<02:06,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  96%|█████████▋| 3853/4000 [51:53<02:03,  1.19it/s]

train:  14%|█▎        | 5/37 [00:04<00:27,  1.18it/s]
overall:  96%|█████████▋| 3854/4000 [51:54<01:59,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  96%|█████████▋| 3855/4000 [51:55<01:57,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  96%|█████████▋| 3856/4000 [51:56<01:55,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  96%|█████████▋| 3857/4000 [51:56<01:54,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  96%|█████████▋| 3858/4000 [51:57<01:53,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  96%|█████████▋| 3859/4000 [51:58<01:52,  1.25it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.25it/s]train loss: 0.7915487289428711
train acc: 0.777999997138977

overall:  96%|█████████▋| 3860/4000 [51:59<01:51,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  97%|█████████▋| 3861/4000 [52:00<01:50,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  97%|█████████▋| 3862/4000 [52:00<01:49,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  97%|█████████▋| 3863/4000 [52:01<01:48,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  97%|█████████▋| 3864/4000 [52:02<01:47,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  97%|█████████▋| 3865/4000 [52:03<01:46,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  97%|█████████▋| 3866/4000 [52:04<01:45,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  97%|█████████▋| 3867/4000 [52:04<01:44,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  97%|█████████▋| 3868/4000 [52:05<01:44,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  97%|█████████▋| 3869/4000 [52:06<01:43,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]train loss: 0.7947141528129578
train acc: 0.7860000133514404

overall:  97%|█████████▋| 3870/4000 [52:07<01:42,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  97%|█████████▋| 3871/4000 [52:07<01:42,  1.26it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.26it/s]
overall:  97%|█████████▋| 3872/4000 [52:08<01:41,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  97%|█████████▋| 3873/4000 [52:09<01:40,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  97%|█████████▋| 3874/4000 [52:10<01:39,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  97%|█████████▋| 3875/4000 [52:11<01:38,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  97%|█████████▋| 3876/4000 [52:11<01:37,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  97%|█████████▋| 3877/4000 [52:12<01:37,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  97%|█████████▋| 3878/4000 [52:13<01:36,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  97%|█████████▋| 3879/4000 [52:14<01:35,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]train loss: 0.8378192186355591
train acc: 0.7689999938011169

overall:  97%|█████████▋| 3880/4000 [52:15<01:34,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  97%|█████████▋| 3881/4000 [52:15<01:33,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  97%|█████████▋| 3882/4000 [52:16<01:32,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  97%|█████████▋| 3883/4000 [52:17<01:32,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  97%|█████████▋| 3884/4000 [52:18<01:31,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  97%|█████████▋| 3885/4000 [52:18<01:27,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  97%|█████████▋| 3886/4000 [52:20<01:53,  1.00it/s]

train:   3%|▎         | 1/37 [00:01<00:51,  1.44s/it]
overall:  97%|█████████▋| 3887/4000 [52:21<01:45,  1.07it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.05s/it]
overall:  97%|█████████▋| 3888/4000 [52:22<01:39,  1.12it/s]

train:   8%|▊         | 3/37 [00:03<00:31,  1.07it/s]
overall:  97%|█████████▋| 3889/4000 [52:22<01:35,  1.17it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.14it/s]train loss: 0.793069064617157
train acc: 0.7753036618232727

overall:  97%|█████████▋| 3890/4000 [52:23<01:31,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  97%|█████████▋| 3891/4000 [52:24<01:29,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.21it/s]
overall:  97%|█████████▋| 3892/4000 [52:25<01:28,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.22it/s]
overall:  97%|█████████▋| 3893/4000 [52:25<01:26,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  97%|█████████▋| 3894/4000 [52:26<01:24,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  97%|█████████▋| 3895/4000 [52:27<01:23,  1.25it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.25it/s]
overall:  97%|█████████▋| 3896/4000 [52:28<01:22,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  97%|█████████▋| 3897/4000 [52:29<01:21,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  97%|█████████▋| 3898/4000 [52:29<01:20,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  97%|█████████▋| 3899/4000 [52:30<01:20,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]train loss: 0.8115634918212891
train acc: 0.7770000100135803

overall:  98%|█████████▊| 3900/4000 [52:31<01:19,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  98%|█████████▊| 3901/4000 [52:32<01:18,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  98%|█████████▊| 3902/4000 [52:33<01:17,  1.27it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.27it/s]
overall:  98%|█████████▊| 3903/4000 [52:33<01:16,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  98%|█████████▊| 3904/4000 [52:34<01:15,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  98%|█████████▊| 3905/4000 [52:35<01:14,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]
overall:  98%|█████████▊| 3906/4000 [52:36<01:14,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  98%|█████████▊| 3907/4000 [52:37<01:13,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  98%|█████████▊| 3908/4000 [52:37<01:12,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  98%|█████████▊| 3909/4000 [52:38<01:11,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]train loss: 0.7948781847953796
train acc: 0.7829999923706055

overall:  98%|█████████▊| 3910/4000 [52:39<01:11,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  98%|█████████▊| 3911/4000 [52:40<01:10,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall:  98%|█████████▊| 3912/4000 [52:40<01:09,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall:  98%|█████████▊| 3913/4000 [52:41<01:08,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  98%|█████████▊| 3914/4000 [52:42<01:07,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall:  98%|█████████▊| 3915/4000 [52:43<01:07,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]
overall:  98%|█████████▊| 3916/4000 [52:44<01:06,  1.26it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.26it/s]
overall:  98%|█████████▊| 3917/4000 [52:44<01:05,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  98%|█████████▊| 3918/4000 [52:45<01:04,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall:  98%|█████████▊| 3919/4000 [52:46<01:03,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]train loss: 0.8181213140487671
train acc: 0.7739999890327454

overall:  98%|█████████▊| 3920/4000 [52:47<01:03,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  98%|█████████▊| 3921/4000 [52:48<01:02,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  98%|█████████▊| 3922/4000 [52:48<00:59,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall:  98%|█████████▊| 3923/4000 [52:50<01:14,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall:  98%|█████████▊| 3924/4000 [52:51<01:09,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:35,  1.03s/it]
overall:  98%|█████████▊| 3925/4000 [52:51<01:05,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  98%|█████████▊| 3926/4000 [52:52<01:02,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.16it/s]
overall:  98%|█████████▊| 3927/4000 [52:53<01:00,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  98%|█████████▊| 3928/4000 [52:54<00:59,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  98%|█████████▊| 3929/4000 [52:54<00:57,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]train loss: 0.7818125486373901
train acc: 0.78340083360672

overall:  98%|█████████▊| 3930/4000 [52:55<00:56,  1.24it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.24it/s]
overall:  98%|█████████▊| 3931/4000 [52:56<00:55,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  98%|█████████▊| 3932/4000 [52:57<00:54,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]
overall:  98%|█████████▊| 3933/4000 [52:58<00:53,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  98%|█████████▊| 3934/4000 [52:58<00:52,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  98%|█████████▊| 3935/4000 [52:59<00:51,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  98%|█████████▊| 3936/4000 [53:00<00:50,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  98%|█████████▊| 3937/4000 [53:01<00:49,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  98%|█████████▊| 3938/4000 [53:02<00:49,  1.26it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.26it/s]
overall:  98%|█████████▊| 3939/4000 [53:02<00:48,  1.26it/s]

train:  46%|████▌     | 17/37 [00:14<00:15,  1.26it/s]train loss: 0.8209506869316101
train acc: 0.7789999842643738

overall:  98%|█████████▊| 3940/4000 [53:03<00:47,  1.26it/s]

train:  49%|████▊     | 18/37 [00:14<00:15,  1.26it/s]
overall:  99%|█████████▊| 3941/4000 [53:04<00:46,  1.26it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.26it/s]
overall:  99%|█████████▊| 3942/4000 [53:05<00:45,  1.26it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.26it/s]
overall:  99%|█████████▊| 3943/4000 [53:06<00:45,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall:  99%|█████████▊| 3944/4000 [53:06<00:44,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall:  99%|█████████▊| 3945/4000 [53:07<00:43,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall:  99%|█████████▊| 3946/4000 [53:08<00:42,  1.27it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.27it/s]
overall:  99%|█████████▊| 3947/4000 [53:09<00:41,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall:  99%|█████████▊| 3948/4000 [53:09<00:41,  1.27it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.27it/s]
overall:  99%|█████████▊| 3949/4000 [53:10<00:40,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]train loss: 0.8329669237136841
train acc: 0.7670000195503235

overall:  99%|█████████▉| 3950/4000 [53:11<00:39,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall:  99%|█████████▉| 3951/4000 [53:12<00:38,  1.26it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.26it/s]
overall:  99%|█████████▉| 3952/4000 [53:13<00:37,  1.26it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.26it/s]
overall:  99%|█████████▉| 3953/4000 [53:13<00:37,  1.26it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.26it/s]
overall:  99%|█████████▉| 3954/4000 [53:14<00:36,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall:  99%|█████████▉| 3955/4000 [53:15<00:35,  1.26it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.26it/s]
overall:  99%|█████████▉| 3956/4000 [53:16<00:34,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall:  99%|█████████▉| 3957/4000 [53:17<00:33,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall:  99%|█████████▉| 3958/4000 [53:17<00:33,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:29<00:00,  1.27it/s]
overall:  99%|█████████▉| 3959/4000 [53:18<00:31,  1.31it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.31it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]train loss: 0.7686865925788879
train acc: 0.7874494194984436

overall:  99%|█████████▉| 3960/4000 [53:20<00:38,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.36s/it]
overall:  99%|█████████▉| 3961/4000 [53:20<00:35,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall:  99%|█████████▉| 3962/4000 [53:21<00:33,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.09it/s]
overall:  99%|█████████▉| 3963/4000 [53:22<00:31,  1.18it/s]

train:  11%|█         | 4/37 [00:03<00:28,  1.15it/s]
overall:  99%|█████████▉| 3964/4000 [53:23<00:30,  1.20it/s]

train:  14%|█▎        | 5/37 [00:04<00:26,  1.19it/s]
overall:  99%|█████████▉| 3965/4000 [53:23<00:28,  1.22it/s]

train:  16%|█▌        | 6/37 [00:05<00:25,  1.22it/s]
overall:  99%|█████████▉| 3966/4000 [53:24<00:27,  1.23it/s]

train:  19%|█▉        | 7/37 [00:06<00:24,  1.23it/s]
overall:  99%|█████████▉| 3967/4000 [53:25<00:26,  1.25it/s]

train:  22%|██▏       | 8/37 [00:06<00:23,  1.25it/s]
overall:  99%|█████████▉| 3968/4000 [53:26<00:25,  1.25it/s]

train:  24%|██▍       | 9/37 [00:07<00:22,  1.25it/s]
overall:  99%|█████████▉| 3969/4000 [53:27<00:24,  1.26it/s]

train:  27%|██▋       | 10/37 [00:08<00:21,  1.26it/s]train loss: 0.7993084788322449
train acc: 0.7979999780654907

overall:  99%|█████████▉| 3970/4000 [53:27<00:23,  1.26it/s]

train:  30%|██▉       | 11/37 [00:09<00:20,  1.26it/s]
overall:  99%|█████████▉| 3971/4000 [53:28<00:23,  1.26it/s]

train:  32%|███▏      | 12/37 [00:10<00:19,  1.26it/s]
overall:  99%|█████████▉| 3972/4000 [53:29<00:22,  1.26it/s]

train:  35%|███▌      | 13/37 [00:10<00:19,  1.26it/s]
overall:  99%|█████████▉| 3973/4000 [53:30<00:21,  1.26it/s]

train:  38%|███▊      | 14/37 [00:11<00:18,  1.26it/s]
overall:  99%|█████████▉| 3974/4000 [53:31<00:20,  1.26it/s]

train:  41%|████      | 15/37 [00:12<00:17,  1.26it/s]
overall:  99%|█████████▉| 3975/4000 [53:31<00:19,  1.27it/s]

train:  43%|████▎     | 16/37 [00:13<00:16,  1.27it/s]
overall:  99%|█████████▉| 3976/4000 [53:32<00:18,  1.27it/s]

train:  46%|████▌     | 17/37 [00:13<00:15,  1.27it/s]
overall:  99%|█████████▉| 3977/4000 [53:33<00:18,  1.27it/s]

train:  49%|████▊     | 18/37 [00:14<00:14,  1.27it/s]
overall:  99%|█████████▉| 3978/4000 [53:34<00:17,  1.27it/s]

train:  51%|█████▏    | 19/37 [00:15<00:14,  1.27it/s]
overall:  99%|█████████▉| 3979/4000 [53:35<00:16,  1.27it/s]

train:  54%|█████▍    | 20/37 [00:16<00:13,  1.27it/s]train loss: 0.818572998046875
train acc: 0.781000018119812

overall: 100%|█████████▉| 3980/4000 [53:35<00:15,  1.27it/s]

train:  57%|█████▋    | 21/37 [00:17<00:12,  1.27it/s]
overall: 100%|█████████▉| 3981/4000 [53:36<00:15,  1.27it/s]

train:  59%|█████▉    | 22/37 [00:17<00:11,  1.27it/s]
overall: 100%|█████████▉| 3982/4000 [53:37<00:14,  1.27it/s]

train:  62%|██████▏   | 23/37 [00:18<00:11,  1.27it/s]
overall: 100%|█████████▉| 3983/4000 [53:38<00:13,  1.26it/s]

train:  65%|██████▍   | 24/37 [00:19<00:10,  1.26it/s]
overall: 100%|█████████▉| 3984/4000 [53:38<00:12,  1.26it/s]

train:  68%|██████▊   | 25/37 [00:20<00:09,  1.26it/s]
overall: 100%|█████████▉| 3985/4000 [53:39<00:11,  1.26it/s]

train:  70%|███████   | 26/37 [00:21<00:08,  1.26it/s]
overall: 100%|█████████▉| 3986/4000 [53:40<00:11,  1.27it/s]

train:  73%|███████▎  | 27/37 [00:21<00:07,  1.27it/s]
overall: 100%|█████████▉| 3987/4000 [53:41<00:10,  1.27it/s]

train:  76%|███████▌  | 28/37 [00:22<00:07,  1.27it/s]
overall: 100%|█████████▉| 3988/4000 [53:42<00:09,  1.27it/s]

train:  78%|███████▊  | 29/37 [00:23<00:06,  1.27it/s]
overall: 100%|█████████▉| 3989/4000 [53:42<00:08,  1.27it/s]

train:  81%|████████  | 30/37 [00:24<00:05,  1.27it/s]train loss: 0.8006998896598816
train acc: 0.7720000147819519

overall: 100%|█████████▉| 3990/4000 [53:43<00:07,  1.27it/s]

train:  84%|████████▍ | 31/37 [00:25<00:04,  1.27it/s]
overall: 100%|█████████▉| 3991/4000 [53:44<00:07,  1.27it/s]

train:  86%|████████▋ | 32/37 [00:25<00:03,  1.27it/s]
overall: 100%|█████████▉| 3992/4000 [53:45<00:06,  1.27it/s]

train:  89%|████████▉ | 33/37 [00:26<00:03,  1.27it/s]
overall: 100%|█████████▉| 3993/4000 [53:46<00:05,  1.27it/s]

train:  92%|█████████▏| 34/37 [00:27<00:02,  1.27it/s]
overall: 100%|█████████▉| 3994/4000 [53:46<00:04,  1.27it/s]

train:  95%|█████████▍| 35/37 [00:28<00:01,  1.27it/s]
overall: 100%|█████████▉| 3995/4000 [53:47<00:03,  1.27it/s]

train:  97%|█████████▋| 36/37 [00:28<00:00,  1.27it/s]
overall: 100%|█████████▉| 3996/4000 [53:48<00:03,  1.32it/s]

train: 100%|██████████| 37/37 [00:29<00:00,  1.32it/s]
train: 100%|██████████| 37/37 [00:29<00:00,  1.24it/s]


train:   0%|          | 0/37 [00:00<?, ?it/s]
overall: 100%|█████████▉| 3997/4000 [53:49<00:02,  1.03it/s]

train:   3%|▎         | 1/37 [00:01<00:49,  1.37s/it]
overall: 100%|█████████▉| 3998/4000 [53:50<00:01,  1.09it/s]

train:   5%|▌         | 2/37 [00:02<00:36,  1.03s/it]
overall: 100%|█████████▉| 3999/4000 [53:51<00:00,  1.14it/s]

train:   8%|▊         | 3/37 [00:02<00:31,  1.08it/s]train loss: 0.8135684132575989
train acc: 0.7570850253105164



test:   0%|          | 0/15 [00:00<?, ?it/s]


test:   7%|▋         | 1/15 [00:01<00:20,  1.45s/it]


test:  13%|█▎        | 2/15 [00:02<00:13,  1.04s/it]


test:  20%|██        | 3/15 [00:02<00:10,  1.10it/s]


test:  27%|██▋       | 4/15 [00:03<00:09,  1.17it/s]


test:  33%|███▎      | 5/15 [00:04<00:08,  1.22it/s]


test:  40%|████      | 6/15 [00:05<00:07,  1.25it/s]


test:  47%|████▋     | 7/15 [00:05<00:06,  1.28it/s]


test:  53%|█████▎    | 8/15 [00:06<00:05,  1.29it/s]


test:  60%|██████    | 9/15 [00:07<00:04,  1.29it/s]


test:  67%|██████▋   | 10/15 [00:08<00:03,  1.31it/s]


test:  73%|███████▎  | 11/15 [00:09<00:03,  1.30it/s]


test:  80%|████████  | 12/15 [00:09<00:02,  1.31it/s]


test:  87%|████████▋ | 13/15 [00:10<00:01,  1.31it/s]


test:  93%|█████████▎| 14/15 [00:11<00:00,  1.31it/s]
test: 100%|██████████| 15/15 [00:11<00:00,  1.31it/s]
test loss: 1.1945980787277222
test acc: 0.6740213632583618

                                                            

                                                     
[Runner] - Save the checkpoint to:

overall: 100%|█████████▉| 3999/4000 [54:03<00:00,  1.14it/s]

train:   8%|▊         | 3/37 [00:15<00:31,  1.08it/s]
                                                            

                                                     
1. result/downstream/grid/states-4000.ckpt

overall: 100%|█████████▉| 3999/4000 [54:03<00:00,  1.14it/s]

train:   8%|▊         | 3/37 [00:15<00:31,  1.08it/s]
overall: 100%|██████████| 4000/4000 [54:03<00:00,  4.35s/it]

train:  11%|█         | 4/37 [00:15<03:00,  5.48s/it]
train:  11%|█         | 4/37 [00:15<02:09,  3.92s/it]

overall: 100%|██████████| 4000/4000 [54:04<00:00,  1.23it/s]