{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2025-06-03T03:35:54.297683Z",
     "iopub.status.busy": "2025-06-03T03:35:54.297546Z",
     "iopub.status.idle": "2025-06-03T03:35:54.310109Z",
     "shell.execute_reply": "2025-06-03T03:35:54.309696Z",
     "shell.execute_reply.started": "2025-06-03T03:35:54.297665Z"
    }
   },
   "outputs": [],
   "source": [
    "# setup autoload\n",
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:58:21.040680Z",
     "start_time": "2024-05-16T13:58:19.777010Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T03:35:54.310690Z",
     "iopub.status.busy": "2025-06-03T03:35:54.310561Z",
     "iopub.status.idle": "2025-06-03T03:35:56.646367Z",
     "shell.execute_reply": "2025-06-03T03:35:56.645745Z",
     "shell.execute_reply.started": "2025-06-03T03:35:54.310677Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n"
     ]
    }
   ],
   "source": [
    "import ast\n",
    "import os\n",
    "import shutil\n",
    "import sys\n",
    "from collections import defaultdict\n",
    "\n",
    "import json\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "sys.path.append(\"/home/tony/Work/tony/Preference\")\n",
    "from preference_data_preparation_auk import *\n",
    "from preference_helper import *\n",
    "from sklearn.model_selection import train_test_split\n",
    "from suno_utils.utils.s3 import download_s3_files\n",
    "from suno_utils.utils.text import read_json, read_jsonl, write_json, write_jsonl\n",
    "from tqdm import tqdm\n",
    "\n",
    "pd.set_option(\"display.max_rows\", 500)\n",
    "pd.set_option(\"display.max_columns\", 500)\n",
    "pd.set_option(\"display.width\", 1000)\n",
    "\n",
    "# setup autoload\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "\n",
    "def custom_parse(x):\n",
    "    try:\n",
    "        return json.loads(x)\n",
    "    except:\n",
    "        return {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:58:21.082172Z",
     "start_time": "2024-05-16T13:58:21.041926Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T03:35:56.647153Z",
     "iopub.status.busy": "2025-06-03T03:35:56.646937Z",
     "iopub.status.idle": "2025-06-03T03:35:56.719671Z",
     "shell.execute_reply": "2025-06-03T03:35:56.719194Z",
     "shell.execute_reply.started": "2025-06-03T03:35:56.647137Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "N_TOKENS_AUDIO 12000\n"
     ]
    }
   ],
   "source": [
    "OUT_DATA_DIR = \"/app2/suno/data/dpo/auk_t1_v19\"\n",
    "os.makedirs(OUT_DATA_DIR, exist_ok=True)\n",
    "shutil.copyfile(\n",
    "    \"/app/suno/data/dpo/7v_v20_full/tokenizer_60k.json\",\n",
    "    os.path.join(OUT_DATA_DIR, \"tokenizer_60k.json\"),\n",
    ")\n",
    "NPZ_DIR = \"/app2/suno/data/dpo/auk_t1_npz\"\n",
    "N_TOKENS_AUDIO = 25 * 8 * 60\n",
    "print(\"N_TOKENS_AUDIO\", N_TOKENS_AUDIO)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:58:53.962528Z",
     "start_time": "2024-05-16T13:58:21.105919Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T03:35:56.720431Z",
     "iopub.status.busy": "2025-06-03T03:35:56.720286Z",
     "iopub.status.idle": "2025-06-03T03:38:37.898669Z",
     "shell.execute_reply": "2025-06-03T03:38:37.898092Z",
     "shell.execute_reply.started": "2025-06-03T03:35:56.720417Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Preference data shape (5106732, 90)\n"
     ]
    }
   ],
   "source": [
    "df = pd.read_pickle(\n",
    "    \"/home/tony/Data/Preference/auk_t1/interesting_clips_auk_t1_20250602.pkl\"\n",
    ")\n",
    "print(\"Preference data shape\", df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2025-06-03T03:38:37.900386Z",
     "iopub.status.busy": "2025-06-03T03:38:37.900125Z",
     "iopub.status.idle": "2025-06-03T03:38:38.730041Z",
     "shell.execute_reply": "2025-06-03T03:38:38.729581Z",
     "shell.execute_reply.started": "2025-06-03T03:38:37.900368Z"
    }
   },
   "outputs": [],
   "source": [
    "# df[\"created_at\"] = pd.to_datetime(df[\"created_at\"], utc=True)\n",
    "# cutoff_date = pd.to_datetime(\"2025-05-10\", utc=True)\n",
    "# # cutoff_date = pd.to_datetime(\"2025-04-17\", utc=True)\n",
    "# print(df.shape, df[df[\"created_at\"] > cutoff_date].shape)\n",
    "# df = df[(df[\"created_at\"] > cutoff_date)].copy()\n",
    "# print(\"after date cut\", df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2025-06-03T03:38:38.730778Z",
     "iopub.status.busy": "2025-06-03T03:38:38.730632Z",
     "iopub.status.idle": "2025-06-03T03:38:45.072118Z",
     "shell.execute_reply": "2025-06-03T03:38:45.071549Z",
     "shell.execute_reply.started": "2025-06-03T03:38:38.730764Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "after dropna (5106732, 85)\n"
     ]
    }
   ],
   "source": [
    "df = df.dropna(axis=1, how=\"all\")\n",
    "print(\"after dropna\", df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:58:56.199480Z",
     "start_time": "2024-05-16T13:58:53.963687Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T03:38:45.072860Z",
     "iopub.status.busy": "2025-06-03T03:38:45.072697Z",
     "iopub.status.idle": "2025-06-03T03:41:01.794943Z",
     "shell.execute_reply": "2025-06-03T03:41:01.794366Z",
     "shell.execute_reply.started": "2025-06-03T03:38:45.072844Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "12184904\n",
      "12184904\n",
      "pre-downloaded df (5106732, 85)\n",
      "downloaded df (5106729, 85)\n"
     ]
    }
   ],
   "source": [
    "converted_paths = os.listdir(NPZ_DIR)\n",
    "print(len(converted_paths))\n",
    "\n",
    "converted_paths = set([f.replace(\".npz\", \"\") for f in converted_paths])\n",
    "print(len(converted_paths))\n",
    "\n",
    "print(\"pre-downloaded df\", df.shape)\n",
    "df[df[\"s3_id\"].isin(converted_paths)].shape\n",
    "df = df[df[\"s3_id\"].isin(converted_paths)].copy()\n",
    "print(\"downloaded df\", df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:58:56.467253Z",
     "start_time": "2024-05-16T13:58:56.207647Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T03:41:01.799617Z",
     "iopub.status.busy": "2025-06-03T03:41:01.799337Z",
     "iopub.status.idle": "2025-06-03T03:41:02.808062Z",
     "shell.execute_reply": "2025-06-03T03:41:02.807612Z",
     "shell.execute_reply.started": "2025-06-03T03:41:01.799599Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "task\n",
       "                      3441265\n",
       "cover                  792860\n",
       "artist_consistency     449010\n",
       "artist_cover           161150\n",
       "upload_extend          127418\n",
       "extend                 111222\n",
       "artist_extend           23804\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"task\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# LET's do the data prep"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:58:56.592883Z",
     "start_time": "2024-05-16T13:58:56.470781Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T03:41:02.808693Z",
     "iopub.status.busy": "2025-06-03T03:41:02.808550Z",
     "iopub.status.idle": "2025-06-03T03:41:05.715694Z",
     "shell.execute_reply": "2025-06-03T03:41:05.715134Z",
     "shell.execute_reply.started": "2025-06-03T03:41:02.808678Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "preference  model_name  \n",
      "False       chirp-auk-t1    2553363\n",
      "True        chirp-auk-t1    2553366\n",
      "Name: count, dtype: int64\n",
      "before filter on model name (5106729, 85)\n",
      "after filter on model name (5106729, 85)\n",
      "is_public\n",
      "False    4850823\n",
      "True      255906\n",
      "Name: count, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "## for 13b this is easy for now\n",
    "print(df.groupby([\"preference\"])[\"model_name\"].value_counts())\n",
    "print(\"before filter on model name\", df.shape)\n",
    "df = df[df[\"model_name\"].isin([\"chirp-auk-t1\"])]\n",
    "# df = df[df[\"model_name\"].isin([\"chirp-v3p5-engine-t-6\"])]\n",
    "print(\"after filter on model name\", df.shape)\n",
    "print(df[\"is_public\"].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:58:56.909539Z",
     "start_time": "2024-05-16T13:58:56.595736Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T03:41:05.716383Z",
     "iopub.status.busy": "2025-06-03T03:41:05.716235Z",
     "iopub.status.idle": "2025-06-03T03:41:13.702654Z",
     "shell.execute_reply": "2025-06-03T03:41:13.702050Z",
     "shell.execute_reply.started": "2025-06-03T03:41:05.716368Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "before filter on request id pairs (5106729, 85)\n",
      "after filter on request id pairs (5106726, 85)\n",
      "preference  model_name  \n",
      "False       chirp-auk-t1    2553363\n",
      "True        chirp-auk-t1    2553363\n",
      "Name: count, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(\"before filter on request id pairs\", df.shape)\n",
    "df = df[\n",
    "    df[\"request_id\"].isin(\n",
    "        df[\"request_id\"].value_counts().index[df[\"request_id\"].value_counts() == 2]\n",
    "    )\n",
    "]\n",
    "print(\"after filter on request id pairs\", df.shape)\n",
    "print(df.groupby([\"preference\"])[\"model_name\"].value_counts())\n",
    "assert df.shape[0] == df[\"request_id\"].nunique() * 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2025-06-03T03:41:13.703395Z",
     "iopub.status.busy": "2025-06-03T03:41:13.703241Z",
     "iopub.status.idle": "2025-06-03T03:54:45.790851Z",
     "shell.execute_reply": "2025-06-03T03:54:45.790258Z",
     "shell.execute_reply.started": "2025-06-03T03:41:13.703378Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "unique_requests 2553363\n",
      "before removing duplicates (5106726, 180)\n",
      "after removing duplicates (5106726, 173)\n"
     ]
    }
   ],
   "source": [
    "# Let's use the old selection for now -- for quality assurance\n",
    "# expand the metadata columns -- this takes forever...~ 6 mins\n",
    "# test_slice = df[\"metadata\"].apply(lambda x: ast.literal_eval(str(x)))\n",
    "# test_slice = df[\"metadata\"].apply(lambda x: custom_parse(x))\n",
    "test_slice = df[\"metadata\"]  # .apply(lambda x: custom_parse(x))\n",
    "test_slice_series = test_slice.apply(pd.Series)\n",
    "df = pd.concat([df, test_slice_series], axis=1, join=\"inner\")\n",
    "print(\"unique_requests\", df[\"request_id\"].nunique())\n",
    "# remove the duplicates\n",
    "print(\"before removing duplicates\", df.shape)\n",
    "df = df.loc[:, ~df.columns.duplicated()].copy()\n",
    "print(\"after removing duplicates\", df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2025-06-03T03:54:45.821154Z",
     "iopub.status.busy": "2025-06-03T03:54:45.820924Z",
     "iopub.status.idle": "2025-06-03T03:54:59.280204Z",
     "shell.execute_reply": "2025-06-03T03:54:59.279697Z",
     "shell.execute_reply.started": "2025-06-03T03:54:45.821136Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pos_diff_preference\n",
       "1.0    1759028\n",
       "2.0     794335\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.sort_values(by=[\"request_id\", \"preference\", \"diff_preference\"])\n",
    "df[\"pos_diff_preference\"] = df[\"diff_preference\"].diff()\n",
    "# df[\"cer_diff_preference\"] = df[\"cer\"].diff()\n",
    "df[df[\"preference\"]][\"pos_diff_preference\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2025-06-03T03:54:59.280934Z",
     "iopub.status.busy": "2025-06-03T03:54:59.280780Z",
     "iopub.status.idle": "2025-06-03T03:55:03.315223Z",
     "shell.execute_reply": "2025-06-03T03:55:03.314651Z",
     "shell.execute_reply.started": "2025-06-03T03:54:59.280918Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "positive param_experiment\n",
      "n_tag_2          65166\n",
      "cfg_steps_240    64837\n",
      "n_tag_1          64412\n",
      "temp_s_95        64185\n",
      "tag_cfg_1        63936\n",
      "cfg_steps_60     63879\n",
      "temp_s_85        63785\n",
      "temp_s_80        62884\n",
      "cfg_steps_10     61839\n",
      "tag_cfg_3        61561\n",
      "Name: count, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "try:\n",
    "    print(\"positive\", df[df[\"preference\"]][\"param_experiment\"].value_counts())\n",
    "except:\n",
    "    pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2025-06-03T03:55:03.315957Z",
     "iopub.status.busy": "2025-06-03T03:55:03.315810Z",
     "iopub.status.idle": "2025-06-03T03:56:47.013117Z",
     "shell.execute_reply": "2025-06-03T03:56:47.012541Z",
     "shell.execute_reply.started": "2025-06-03T03:55:03.315942Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 124565 duplicated prompts 62284 unique requests\n",
      "Found 29719 request_ids with duplicate prompts but not highest play counts in their group\n",
      "['1e32db78-9c98-4995-8492-a5e98aaf65ac', '730936a6-f6bb-4f43-a54d-d698b88b8cb6', 'fd2c071e-3d91-4149-aa31-28256a78ce9b', 'ff2db895-04ac-4bdf-a73d-ccd4fb68a53b', '1e5b8877-d64f-4982-abe9-83f3963a496d', '4950055f-b604-4197-bd52-92deaaa61f1b', '9f076cb4-12ab-4855-bc96-a33788f79198', 'b117ad54-f039-4253-9940-47bfd5cb4fb1', '21771d53-d589-4c7e-bd96-921d836c62ba', 'e7d72331-628d-4332-9b19-2f95745e0875']\n",
      "Before dedup user gen requests 5106726\n",
      "After dedup user gen requests 5106726\n"
     ]
    }
   ],
   "source": [
    "# Find duplicated prompts with count > 2\n",
    "duplicate_entries = df.groupby(\n",
    "    [\"user_id\", \"prompt_text\", \"tags\", \"task\", \"edited_clip_id\"]\n",
    ").filter(lambda x: len(x) > 2)\n",
    "print(\n",
    "    \"Found\",\n",
    "    len(duplicate_entries),\n",
    "    \"duplicated prompts\",\n",
    "    len(duplicate_entries[\"request_id\"].unique()),\n",
    "    \"unique requests\",\n",
    ")\n",
    "\n",
    "# Group by user_id, prompt_text, and tags to find duplicate prompt groups\n",
    "prompt_groups = duplicate_entries.groupby(\n",
    "    [\"user_id\", \"prompt_text\", \"tags\", \"task\", \"edited_clip_id\"]\n",
    ")\n",
    "\n",
    "# For each prompt group, find the request_id with the highest total reaction_play_count\n",
    "low_play_count_request_ids = []\n",
    "for prompt_key, prompt_group in prompt_groups:\n",
    "    # Get the sum of reaction_play_count for each request_id in this group\n",
    "    request_play_counts = prompt_group.groupby(\"request_id\")[\n",
    "        \"reaction_play_count\"\n",
    "    ].sum()\n",
    "\n",
    "    # Find the max play count in this group\n",
    "    max_play_count = request_play_counts.max()\n",
    "\n",
    "    # Add request_ids that don't have the max play count to our filter list\n",
    "    lower_play_count_request_ids = request_play_counts[\n",
    "        request_play_counts < max_play_count\n",
    "    ].index.tolist()\n",
    "    low_play_count_request_ids.extend(lower_play_count_request_ids)\n",
    "\n",
    "# Display the filtered request IDs\n",
    "print(\n",
    "    f\"Found {len(low_play_count_request_ids)} request_ids with duplicate prompts but not highest play counts in their group\"\n",
    ")\n",
    "print(\n",
    "    low_play_count_request_ids[:10]\n",
    "    if len(low_play_count_request_ids) > 10\n",
    "    else low_play_count_request_ids\n",
    ")\n",
    "print(\"Before dedup user gen requests\", df.shape[0])\n",
    "# df = df[~df[\"request_id\"].isin(low_play_count_request_ids)]\n",
    "print(\"After dedup user gen requests\", df.shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:40.799375Z",
     "start_time": "2024-05-16T13:59:36.394236Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T03:56:47.013849Z",
     "iopub.status.busy": "2025-06-03T03:56:47.013694Z",
     "iopub.status.idle": "2025-06-03T03:57:19.319234Z",
     "shell.execute_reply": "2025-06-03T03:57:19.318669Z",
     "shell.execute_reply.started": "2025-06-03T03:56:47.013832Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99343\n",
      "good_continue_at\n",
      "True     5093937\n",
      "False      12789\n",
      "Name: count, dtype: int64\n",
      "\n",
      " Check some basics... \n",
      " preference\n",
      "False    2553363\n",
      "True     2553363\n",
      "Name: count, dtype: int64 model_name\n",
      "chirp-auk-t1    5106726\n",
      "Name: count, dtype: int64 preference  model_name  \n",
      "False       chirp-auk-t1    2553363\n",
      "True        chirp-auk-t1    2553363\n",
      "Name: count, dtype: int64\n",
      "task\n",
      "                      3441262\n",
      "cover                  792860\n",
      "artist_consistency     449010\n",
      "artist_cover           161150\n",
      "upload_extend          127418\n",
      "extend                 111222\n",
      "artist_extend           23804\n",
      "Name: count, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "df[\"id\"] = df[\"str_id\"]\n",
    "# get the original duration of the clips, if they are concacted\n",
    "df[\"original_duration_s\"] = df[\"total_start_s\"] + df[\"duration\"]\n",
    "# classify the continue at behavoirs by the duration choice\n",
    "audio_prompt_id_to_continue_at = {}\n",
    "\n",
    "for _, row in df[~df[\"continued_parent\"].isna()].iterrows():\n",
    "    audio_prompt_id = row[\"continued_parent\"]\n",
    "    if audio_prompt_id not in audio_prompt_id_to_continue_at:\n",
    "        audio_prompt_id_to_continue_at[audio_prompt_id] = row[\"continue_at\"]\n",
    "    else:\n",
    "        # pick the max\n",
    "        audio_prompt_id = max(\n",
    "            audio_prompt_id_to_continue_at[audio_prompt_id], row[\"continue_at\"]\n",
    "        )\n",
    "print(len(audio_prompt_id_to_continue_at))\n",
    "df[\"has_continue_and_start_continue_at\"] = df[\"id\"].apply(\n",
    "    lambda x: audio_prompt_id_to_continue_at.get(x)\n",
    ")\n",
    "# we want continue at to be at most of the clip...\n",
    "df[\"good_continue_at\"] = (\n",
    "    (df[\"has_continue_and_start_continue_at\"] / df[\"duration\"]) > 0.9\n",
    ") | df[\"has_continue_and_start_continue_at\"].isna()\n",
    "print(df[\"good_continue_at\"].value_counts())\n",
    "\n",
    "\n",
    "print(\n",
    "    \"\\n Check some basics... \\n\",\n",
    "    df[\"preference\"].value_counts(),\n",
    "    df[\"model_name\"].value_counts(),\n",
    "    df.groupby([\"preference\"])[\"model_name\"].value_counts(),\n",
    ")\n",
    "\n",
    "df = df.sort_values(by=[\"request_id\", \"preference\"])\n",
    "df[\"duration_rel_diff\"] = df[\"duration\"].diff()\n",
    "df[\"play_rel_diff\"] = df[\"reaction_play_count\"].diff()\n",
    "print(df[\"task\"].value_counts())\n",
    "\n",
    "# df[\"post_infill_duration\"] = (\n",
    "#     df[\"duration\"]\n",
    "#     + df[\"infill_context_end_s\"]\n",
    "#     - df[\"infill_context_start_s\"]\n",
    "#     - df[\"include_future_s\"]\n",
    "#     - df[\"include_history_s\"]\n",
    "#     - df[\"infill_dur_s\"]\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.035167Z",
     "start_time": "2024-05-16T13:59:40.801098Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T04:11:59.472873Z",
     "iopub.status.busy": "2025-06-03T04:11:59.472537Z",
     "iopub.status.idle": "2025-06-03T04:12:18.495938Z",
     "shell.execute_reply": "2025-06-03T04:12:18.495337Z",
     "shell.execute_reply.started": "2025-06-03T04:11:59.472853Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "after duration 0.9995725245489967\n",
      "after infill duration 1.0\n",
      "neg_filter_reaction_play_count 1.0\n",
      "neg_filter_upvote_count 0.9894\n",
      "neg_filter_norm_play_frac 1.0\n",
      "neg_filter_continues 1.0\n",
      "----------------\n",
      "pos_filter_continues 0.995\n",
      "pos_filter_reaction_play_count 1.0\n",
      "pos_filter_relative_play_count 0.9783\n",
      "pos_filter_cer_diff_preference 1.0\n",
      "pos_filter_bad_flags 0.9998\n",
      "after filter on play counts 0.9861\n",
      "after filter on higher quality 0.3166\n",
      "----------------\n",
      "negative 2524932 positive 752995\n",
      "----------------\n",
      "total pair requests 2553363  --> selected pair requests 745142 frac 0.292  --> total intitial users 270208\n"
     ]
    }
   ],
   "source": [
    "normal_pos_play_count = 3\n",
    "# this is lower, cause a concat is probably already ensuring that it is good\n",
    "concat_pos_play_count = 1\n",
    "# this is a filter on the concated clip\n",
    "concat_total_play_count = 3\n",
    "\n",
    "all_fitlers = (df[\"duration\"] >= 10) & (df[\"duration\"] <= 480)\n",
    "print(\"after duration\", all_fitlers.sum() / df.shape[0])\n",
    "infill_duration_filter = ~df[\"task\"].isin(\n",
    "    [\n",
    "        \"infill\",\n",
    "        \"infill_intro\",\n",
    "        \"infill_outro\",\n",
    "    ]\n",
    ")  # | (df[\"post_infill_duration\"] <= 239)\n",
    "print(\"after infill duration\", infill_duration_filter.sum() / df.shape[0])\n",
    "# negative fitlers\n",
    "total_negative = df[~df[\"preference\"]].shape[0]\n",
    "neg_filter_reaction_play_count = (~df[\"preference\"]) & (df[\"reaction_play_count\"] >= 1)\n",
    "print(\n",
    "    \"neg_filter_reaction_play_count\",\n",
    "    round(neg_filter_reaction_play_count.sum() / total_negative, 4),\n",
    ")\n",
    "neg_filter_upvote_count = (~df[\"preference\"]) & (df[\"upvote_count\"] == 0)\n",
    "print(\n",
    "    \"neg_filter_upvote_count\",\n",
    "    round(neg_filter_upvote_count.sum() / total_negative, 4),\n",
    ")\n",
    "neg_filter_norm_play_frac = (~df[\"preference\"]) & (df[\"norm_play_frac\"] <= 3.1)\n",
    "print(\n",
    "    \"neg_filter_norm_play_frac\",\n",
    "    round(neg_filter_norm_play_frac.sum() / total_negative, 4),\n",
    ")\n",
    "neg_filter_continues = (~df[\"preference\"]) & (\n",
    "    df[\"has_continue_and_start_continue_at\"].isna()\n",
    ")\n",
    "print(\n",
    "    \"neg_filter_continues\",\n",
    "    round(neg_filter_continues.sum() / total_negative, 4),\n",
    ")\n",
    "\n",
    "neg_filter_selection_mask = (\n",
    "    all_fitlers\n",
    "    & infill_duration_filter\n",
    "    & neg_filter_reaction_play_count\n",
    "    & neg_filter_upvote_count\n",
    "    & neg_filter_norm_play_frac\n",
    "    & neg_filter_continues\n",
    ")\n",
    "\n",
    "print(\"----------------\")\n",
    "total_positive = df[df[\"preference\"]].shape[0]\n",
    "assert total_positive == total_negative\n",
    "pos_filter_continues = (df[\"preference\"]) & (df[\"good_continue_at\"])\n",
    "print(\"pos_filter_continues\", round(pos_filter_continues.sum() / total_positive, 4))\n",
    "pos_filter_reaction_play_count = (df[\"preference\"]) & (df[\"reaction_play_count\"] >= 1)\n",
    "print(\n",
    "    \"pos_filter_reaction_play_count\",\n",
    "    round(pos_filter_reaction_play_count.sum() / total_positive, 4),\n",
    ")\n",
    "pos_filter_relative_play_count = (df[\"preference\"]) & (df[\"play_rel_diff\"] >= 0)\n",
    "print(\n",
    "    \"pos_filter_relative_play_count\",\n",
    "    round(pos_filter_relative_play_count.sum() / total_positive, 4),\n",
    ")\n",
    "pos_filter_cer_diff_preference = (\n",
    "    df[\n",
    "        \"preference\"\n",
    "    ]  # & (df[\"pos_diff_preference\"] == 2) # & (df[\"cer_diff_preference\"] < 0.5) & (df[\"cer\"] < 0.99)\n",
    ")\n",
    "print(\n",
    "    \"pos_filter_cer_diff_preference\",\n",
    "    round(pos_filter_cer_diff_preference.sum() / total_positive, 4),\n",
    ")\n",
    "pos_filter_bad_flags = (\n",
    "    (df[\"preference\"]) & (df[\"flag_count\"] == 0) & (df[\"dislike_count\"] == 0)\n",
    ")\n",
    "print(\n",
    "    \"pos_filter_bad_flags\",\n",
    "    round(pos_filter_bad_flags.sum() / total_positive, 4),\n",
    ")\n",
    "pos_filter_play_counts = (df[\"preference\"]) & (\n",
    "    (\n",
    "        (df[\"part_of_concat\"])\n",
    "        & (df[\"reaction_play_count\"] >= concat_pos_play_count)\n",
    "        & (df[\"concat_play_counts\"] >= concat_total_play_count)\n",
    "    )\n",
    "    | (\n",
    "        (~df[\"part_of_concat\"]) & (df[\"reaction_play_count\"] >= normal_pos_play_count)\n",
    "        # & (df[\"norm_play_frac\"] >= 2.1)  # this is a bit of a luxury cut...\n",
    "    )\n",
    "    | (df[\"task\"].isin([\"infill\", \"infill_intro\", \"infill_outro\"]))\n",
    ")\n",
    "print(\n",
    "    \"after filter on play counts\",\n",
    "    round(pos_filter_play_counts.sum() / total_positive, 4),\n",
    ")\n",
    "high_quality_tasks_filter = (\n",
    "    (\n",
    "        df[\"task\"].isin(\n",
    "            [\n",
    "                \"cover\",\n",
    "                \"upload_extend\",\n",
    "                \"cover_extend\",\n",
    "                \"artist_cover\",\n",
    "                \"extend\",\n",
    "                \"artist_consistency\",\n",
    "                \"artist_extend\",\n",
    "                \"\",\n",
    "            ]\n",
    "        )\n",
    "    )\n",
    "    & (\n",
    "        (df[\"upvote_count\"] >= 1)  # (df[\"upvote_count\"] >= 1)\n",
    "        | (df[\"reaction_play_count\"] >= 5)\n",
    "        | (df[\"concat_play_counts\"] >= 5)\n",
    "    )\n",
    "    & (\n",
    "        (df[\"part_of_concat\"])\n",
    "        | (\n",
    "            (~df[\"part_of_concat\"])\n",
    "            & (df[\"norm_play_frac\"] >= 5.1)  # this is a bit of a luxury cut...\n",
    "            & (\n",
    "                df[\"norm_play_frac\"] >= df[\"reaction_play_count\"] / 3\n",
    "            )  # play duration is not low on average\n",
    "        )\n",
    "    )\n",
    ")\n",
    "medium_quality_tasks_filter = (\n",
    "    df[\"task\"].isin(\n",
    "        [\n",
    "            \"infill\",\n",
    "            \"infill_intro\",\n",
    "            \"infill_outro\",\n",
    "        ]\n",
    "    )\n",
    ") & (  # let more infill through only in this case...\n",
    "    (\n",
    "        df[\"upvote_count\"] >= 0\n",
    "    )  # (df[\"upvote_count\"] >= 1)  (df[\"pos_diff_preference\"] == 2)\n",
    "    | (df[\"reaction_play_count\"] >= 1)\n",
    "    | (df[\"concat_play_counts\"] >= 1)\n",
    ")\n",
    "pos_filter_higher_quality = (df[\"preference\"]) & (\n",
    "    high_quality_tasks_filter | medium_quality_tasks_filter\n",
    ")\n",
    "print(\n",
    "    \"after filter on higher quality\",\n",
    "    round(pos_filter_higher_quality.sum() / total_positive, 4),\n",
    ")\n",
    "\n",
    "user_gen_filter = (\n",
    "    df[\"user_n_clips\"] >= 20\n",
    ")  # user needs to have genereated at least 100 over the time period\n",
    "\n",
    "pos_duration_filter = (df[\"task\"] != \"\") | (\n",
    "    (df[\"duration\"] < 8 * 60 - 10)\n",
    "    & ((df[\"duration_rel_diff\"] / df[\"duration\"]) < 0.50)\n",
    "    & (df[\"duration_rel_diff\"] < 60 * 2)\n",
    ")\n",
    "\n",
    "print(\"----------------\")\n",
    "pos_filter_selectin_mask = (\n",
    "    (df[\"preference\"])  # get basics aligned\n",
    "    & all_fitlers\n",
    "    & infill_duration_filter\n",
    "    & pos_filter_continues\n",
    "    & pos_filter_reaction_play_count\n",
    "    & pos_filter_relative_play_count\n",
    "    & pos_filter_cer_diff_preference\n",
    "    & pos_filter_bad_flags\n",
    "    & pos_filter_play_counts\n",
    "    & pos_filter_higher_quality\n",
    "    & user_gen_filter\n",
    "    & pos_duration_filter\n",
    ")\n",
    "print(\n",
    "    \"negative\",\n",
    "    sum(neg_filter_selection_mask),\n",
    "    \"positive\",\n",
    "    sum(pos_filter_selectin_mask),\n",
    ")\n",
    "\n",
    "neg_filter_requests = df[neg_filter_selection_mask][\"request_id\"].unique()\n",
    "pos_filter_requests = df[pos_filter_selectin_mask][\"request_id\"].unique()\n",
    "# looking for very strong signal here:\n",
    "# listen to the positive/negative more than once\n",
    "# disliked one of the clips\n",
    "unique_requests = set(pos_filter_requests).intersection(neg_filter_requests)\n",
    "print(\"----------------\")\n",
    "print(\n",
    "    \"total pair requests\",\n",
    "    df[\"request_id\"].nunique(),\n",
    "    \" --> selected pair requests\",\n",
    "    len(unique_requests),\n",
    "    f\"frac {len(unique_requests) / df['request_id'].nunique():.3f}\",\n",
    "    \" --> total intitial users\",\n",
    "    df[\"user_id\"].nunique(),\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.250737Z",
     "start_time": "2024-05-16T13:59:41.036434Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T04:12:18.496912Z",
     "iopub.status.busy": "2025-06-03T04:12:18.496743Z",
     "iopub.status.idle": "2025-06-03T04:12:26.265339Z",
     "shell.execute_reply": "2025-06-03T04:12:26.264751Z",
     "shell.execute_reply.started": "2025-06-03T04:12:18.496896Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "auk_t1_v19 requests 745142 clips 1490284 total khrs 76.555; N gpus for 1000 iters 93.143; 4 gpus for x iters 23285.688; n unique users 155310 n pro users 147994\n"
     ]
    }
   ],
   "source": [
    "df_slice = df[df[\"request_id\"].isin(set(unique_requests))].copy()\n",
    "print(\n",
    "    f\"{os.path.basename(OUT_DATA_DIR)} requests\",\n",
    "    df_slice[\"request_id\"].nunique(),\n",
    "    \"clips\",\n",
    "    df_slice.shape[0],\n",
    "    f\"total khrs {sum(df_slice['duration'] / 3600 / 1000):.3f};\",\n",
    "    f\"N gpus for 1000 iters {df_slice.shape[0] / 8 / 2 / 1000:.3f};\",\n",
    "    f\"4 gpus for x iters {df_slice.shape[0] / 8 / 2 / 4:.3f};\",\n",
    "    f\"n unique users {df_slice['user_id'].nunique()}\",\n",
    "    f\"n pro users {df_slice[df_slice['is_pro_user']]['user_id'].nunique()}\",\n",
    ")\n",
    "# auk_mix_t1_v2 requests 102002 clips 204004 total khrs 9.191; N gpus for 1000 iters 12.750; 4 gpus for x iters 3187.562; n unique users 36408 n pro users 34038\n",
    "# auk_t1_v1 requests 9179 clips 18358 total khrs 0.854; N gpus for 1000 iters 1.147; 4 gpus for x iters 286.844; n unique users 6288 n pro users 6275\n",
    "# auk_t1_v2 requests 40903 clips 81806 total khrs 3.864; N gpus for 1000 iters 5.113; 4 gpus for x iters 1278.219; n unique users 21079 n pro users 20966\n",
    "# auk_t1_v3 requests 102015 clips 204030 total khrs 9.703; N gpus for 1000 iters 12.752; 4 gpus for x iters 3187.969; n unique users 42837 n pro users 42462\n",
    "# auk_t1_v4 requests 211452 clips 422904 total khrs 20.216; N gpus for 1000 iters 26.431; 4 gpus for x iters 6607.875; n unique users 71182 n pro users 70059\n",
    "# auk_t1_v9 requests 547743 clips 1095486 total khrs 56.567; N gpus for 1000 iters 68.468; 4 gpus for x iters 17116.969; n unique users 129354 n pro users 125126\n",
    "# auk_t1_v13 requests 627646 clips 1255292 total khrs 64.530; N gpus for 1000 iters 78.456; 4 gpus for x iters 19613.938; n unique users 139873 n pro users 134738\n",
    "# auk_t1_v17 requests 494902 clips 989804 total khrs 51.445; N gpus for 1000 iters 61.863; 4 gpus for x iters 15465.688; n unique users 114240 n pro users 109708\n",
    "# auk_t1_v19 requests 740881 clips 1481762 total khrs 76.133; N gpus for 1000 iters 92.610; 4 gpus for x iters 23152.531; n unique users 154570 n pro users 147261"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.277006Z",
     "start_time": "2024-05-16T13:59:41.252105Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T03:57:41.888685Z",
     "iopub.status.busy": "2025-06-03T03:57:41.888531Z",
     "iopub.status.idle": "2025-06-03T03:57:42.258120Z",
     "shell.execute_reply": "2025-06-03T03:57:42.257552Z",
     "shell.execute_reply.started": "2025-06-03T03:57:41.888669Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "positive in playlist (226778, 179)\n",
      "task\n",
      "                      1030548\n",
      "cover                  185636\n",
      "artist_consistency     135816\n",
      "extend                  54530\n",
      "artist_cover            42648\n",
      "upload_extend           28786\n",
      "artist_extend           12320\n",
      "Name: count, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "test_mask = (df_slice[\"preference\"]) & (\n",
    "    (df_slice[\"is_in_playlist\"]) | (df_slice[\"concat_in_playlist\"])\n",
    ")\n",
    "print(\"positive in playlist\", df_slice[test_mask].shape)\n",
    "print(df_slice[\"task\"].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2025-06-03T04:12:52.087825Z",
     "iopub.status.busy": "2025-06-03T04:12:52.087459Z",
     "iopub.status.idle": "2025-06-03T04:12:52.109089Z",
     "shell.execute_reply": "2025-06-03T04:12:52.108551Z",
     "shell.execute_reply.started": "2025-06-03T04:12:52.087806Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "is_public\n",
      "False    1364496\n",
      "True      125788\n",
      "Name: count, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(df_slice[\"is_public\"].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2025-06-03T04:12:53.143206Z",
     "iopub.status.busy": "2025-06-03T04:12:53.143048Z",
     "iopub.status.idle": "2025-06-03T04:12:53.441243Z",
     "shell.execute_reply": "2025-06-03T04:12:53.440744Z",
     "shell.execute_reply.started": "2025-06-03T04:12:53.143191Z"
    }
   },
   "outputs": [],
   "source": [
    "df_slice[\"npz_path\"] = df_slice[\"s3_id\"].map(lambda x: f\"{NPZ_DIR}/{x}.npz\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2025-06-03T04:12:54.547877Z",
     "iopub.status.busy": "2025-06-03T04:12:54.547478Z",
     "iopub.status.idle": "2025-06-03T04:13:11.071443Z",
     "shell.execute_reply": "2025-06-03T04:13:11.070835Z",
     "shell.execute_reply.started": "2025-06-03T04:12:54.547861Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "846462\n",
      "(1490284, 180)\n",
      "(676032, 180)\n"
     ]
    }
   ],
   "source": [
    "tr_metas_t1_v7 = read_jsonl(\n",
    "    os.path.join(\"/app2/suno/data/dpo/auk_t1_v6\", f\"meta_tr.jsonl\")\n",
    ")\n",
    "tr_metas_t1_v6 = read_jsonl(\n",
    "    os.path.join(\"/app2/suno/data/dpo/auk_t1_v7\", f\"meta_tr.jsonl\")\n",
    ")\n",
    "known_train_ids = set()\n",
    "for prev_tr_meta in tr_metas_t1_v7:\n",
    "    known_train_ids.add(prev_tr_meta[\"id\"])\n",
    "for prev_tr_meta in tr_metas_t1_v6:\n",
    "    known_train_ids.add(prev_tr_meta[\"id\"])\n",
    "print(len(known_train_ids))\n",
    "print(df_slice.shape)\n",
    "df_slice = df_slice[~df_slice[\"id\"].isin(known_train_ids)].copy()\n",
    "print(df_slice.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2025-06-03T04:13:11.072415Z",
     "iopub.status.busy": "2025-06-03T04:13:11.072249Z",
     "iopub.status.idle": "2025-06-03T04:13:11.880569Z",
     "shell.execute_reply": "2025-06-03T04:13:11.880083Z",
     "shell.execute_reply.started": "2025-06-03T04:13:11.072399Z"
    }
   },
   "outputs": [],
   "source": [
    "# df_slice[\"created_at\"] = pd.to_datetime(df_slice[\"created_at\"], utc=True)\n",
    "# cutoff_date = pd.to_datetime(\"2025-05-12\", utc=True)\n",
    "# print(df_slice.shape, df_slice[df_slice[\"created_at\"] >= cutoff_date].shape)\n",
    "# df_slice = df_slice[(df_slice[\"created_at\"] >= cutoff_date)].copy()\n",
    "# print(\"after date cut\", df_slice.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T14:00:20.866354Z",
     "start_time": "2024-05-16T14:00:12.443344Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T04:13:12.827480Z",
     "iopub.status.busy": "2025-06-03T04:13:12.827327Z",
     "iopub.status.idle": "2025-06-03T04:13:12.919019Z",
     "shell.execute_reply": "2025-06-03T04:13:12.918432Z",
     "shell.execute_reply.started": "2025-06-03T04:13:12.827464Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(676032, 180)\n",
      "task\n",
      "                      471072\n",
      "cover                  88364\n",
      "artist_consistency     58252\n",
      "extend                 22628\n",
      "artist_cover           19654\n",
      "upload_extend          11420\n",
      "artist_extend           4642\n",
      "Name: count, dtype: int64\n"
     ]
    },
    {
     "ename": "NameError",
     "evalue": "name 'BREAK' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[24], line 6\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[38;5;28mprint\u001b[39m(df_slice\u001b[38;5;241m.\u001b[39mshape)\n\u001b[1;32m      5\u001b[0m \u001b[38;5;28mprint\u001b[39m(df_slice[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtask\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mvalue_counts())\n\u001b[0;32m----> 6\u001b[0m \u001b[43mBREAK\u001b[49m\n",
      "\u001b[0;31mNameError\u001b[0m: name 'BREAK' is not defined"
     ]
    }
   ],
   "source": [
    "# df_slice.to_pickle(\n",
    "#     \"/home/tony/Data/Preference/auk_t1/interesting_clips_auk_t1_20250602_v19_slice.pkl\"\n",
    "# )\n",
    "print(df_slice.shape)\n",
    "print(df_slice[\"task\"].value_counts())\n",
    "BREAK"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Need to kick out the ones has gpt prompt -- these are pairs with different text inputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.932296Z",
     "start_time": "2024-05-16T13:59:41.932287Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T04:13:20.321278Z",
     "iopub.status.busy": "2025-06-03T04:13:20.320928Z",
     "iopub.status.idle": "2025-06-03T04:13:20.360053Z",
     "shell.execute_reply": "2025-06-03T04:13:20.359514Z",
     "shell.execute_reply.started": "2025-06-03T04:13:20.321259Z"
    }
   },
   "outputs": [],
   "source": [
    "# don't have continue at\n",
    "df_slice[\"request_id\"] = df_slice[\"request_id\"].astype(str)\n",
    "# df_slice[\"npz_path\"] = df_slice[\"npz_path\"].apply(lambda x: str(x).replace(\"_npz\", \"_npz/\"))\n",
    "# df_slice[df_slice[\"continue_at\"].isna()][\"request_id\"].nunique(), df_slice[\"request_id\"].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2025-06-03T04:13:20.361121Z",
     "iopub.status.busy": "2025-06-03T04:13:20.360955Z",
     "iopub.status.idle": "2025-06-03T04:13:22.682141Z",
     "shell.execute_reply": "2025-06-03T04:13:22.681553Z",
     "shell.execute_reply.started": "2025-06-03T04:13:20.361105Z"
    }
   },
   "outputs": [],
   "source": [
    "print(df_slice.shape)\n",
    "df_slice = df_slice[df_slice[\"request_id\"].apply(lambda x: len(x) > 3)]\n",
    "print(df_slice.shape)\n",
    "# df_slice = df_slice[df_slice[\"is_pro_user\"]].copy()\n",
    "print(df_slice.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.932966Z",
     "start_time": "2024-05-16T13:59:41.932957Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T04:13:22.682851Z",
     "iopub.status.busy": "2025-06-03T04:13:22.682700Z",
     "iopub.status.idle": "2025-06-03T04:13:22.807060Z",
     "shell.execute_reply": "2025-06-03T04:13:22.806487Z",
     "shell.execute_reply.started": "2025-06-03T04:13:22.682835Z"
    }
   },
   "outputs": [],
   "source": [
    "final_filtered_requests = df_slice[\"request_id\"].astype(str).unique()\n",
    "# final_filtered_requests = df_slice[df_slice[\"is_pro_user\"]][\"request_id\"].astype(str).unique()\n",
    "print(len(final_filtered_requests))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.934277Z",
     "start_time": "2024-05-16T13:59:41.934268Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T04:13:22.807864Z",
     "iopub.status.busy": "2025-06-03T04:13:22.807716Z",
     "iopub.status.idle": "2025-06-03T04:13:26.533531Z",
     "shell.execute_reply": "2025-06-03T04:13:26.532946Z",
     "shell.execute_reply.started": "2025-06-03T04:13:22.807848Z"
    }
   },
   "outputs": [],
   "source": [
    "train_requests, val_requests = train_test_split(\n",
    "    sorted(list(final_filtered_requests)), test_size=0.01, random_state=42\n",
    ")\n",
    "print(len(train_requests), len(val_requests))\n",
    "\n",
    "train_df = df_slice[df_slice[\"request_id\"].astype(str).isin(set(train_requests))].copy()\n",
    "val_df = df_slice[df_slice[\"request_id\"].astype(str).isin(set(val_requests))].copy()\n",
    "train_df = train_df.sort_values(by=[\"request_id\", \"preference\"])\n",
    "train_df = train_df  # .reset_index()\n",
    "val_df = val_df.sort_values(by=[\"request_id\", \"preference\"])\n",
    "val_df = val_df  # .reset_index()\n",
    "train_df = train_df.reset_index(drop=True)\n",
    "val_df = val_df.reset_index(drop=True)\n",
    "print(train_df.shape, val_df.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Actually make"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.935620Z",
     "start_time": "2024-05-16T13:59:41.935613Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T04:13:26.535093Z",
     "iopub.status.busy": "2025-06-03T04:13:26.534753Z",
     "iopub.status.idle": "2025-06-03T04:14:03.432317Z",
     "shell.execute_reply": "2025-06-03T04:14:03.431740Z",
     "shell.execute_reply.started": "2025-06-03T04:13:26.535076Z"
    }
   },
   "outputs": [],
   "source": [
    "total_duration = 0\n",
    "for i, row in tqdm(train_df.iterrows(), total=len(train_df)):\n",
    "    # we need to alternate between preference: neg, pos\n",
    "    # print(i, row)\n",
    "    try:\n",
    "        assert row[\"preference\"] == (i % 2 == 1)\n",
    "        total_duration += row[\"duration\"]\n",
    "    except Exception as E:\n",
    "        print(i, row)\n",
    "        print(E)\n",
    "        raise ValueError()\n",
    "\n",
    "print(\n",
    "    f\"{round(total_duration / 60 / 60):,} hours of {train_df.shape[0]} clips, {train_df.shape[0] / 8 / 2 / 1000} nodes, {train_df.shape[0] / 8 / 8 / 16} iters\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.936268Z",
     "start_time": "2024-05-16T13:59:41.936260Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T04:14:03.433051Z",
     "iopub.status.busy": "2025-06-03T04:14:03.432894Z",
     "iopub.status.idle": "2025-06-03T04:15:10.046637Z",
     "shell.execute_reply": "2025-06-03T04:15:10.046090Z",
     "shell.execute_reply.started": "2025-06-03T04:14:03.433035Z"
    }
   },
   "outputs": [],
   "source": [
    "make_dataset(\n",
    "    val_df, OUT_DATA_DIR, is_val=True, npz_dir=NPZ_DIR, t_data_memmap=N_TOKENS_AUDIO\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2025-06-03T04:15:10.047314Z",
     "iopub.status.busy": "2025-06-03T04:15:10.047167Z",
     "iopub.status.idle": "2025-06-03T04:15:10.863917Z",
     "shell.execute_reply": "2025-06-03T04:15:10.863457Z",
     "shell.execute_reply.started": "2025-06-03T04:15:10.047298Z"
    }
   },
   "outputs": [],
   "source": [
    "# test_npz = np.load(\"/app/suno/data/dpo/30b_npz/26d19085-18da-4701-af43-122684543891.npz\")\n",
    "# for k in test_npz.keys():\n",
    "#     print(k)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.936964Z",
     "start_time": "2024-05-16T13:59:41.936957Z"
    },
    "execution": {
     "iopub.execute_input": "2025-06-03T04:15:10.864717Z",
     "iopub.status.busy": "2025-06-03T04:15:10.864568Z"
    }
   },
   "outputs": [],
   "source": [
    "make_dataset(\n",
    "    train_df, OUT_DATA_DIR, is_val=False, npz_dir=NPZ_DIR, t_data_memmap=N_TOKENS_AUDIO\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-01-29T19:46:47.549860Z",
     "start_time": "2024-01-29T19:46:47.548015Z"
    }
   },
   "source": [
    "# Validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.937879Z",
     "start_time": "2024-05-16T13:59:41.937870Z"
    }
   },
   "outputs": [],
   "source": [
    "# verify\n",
    "mm = np.memmap(os.path.join(OUT_DATA_DIR, f\"data_val.bin\"), dtype=np.uint16, mode=\"r\")\n",
    "test_metas = read_jsonl(os.path.join(OUT_DATA_DIR, f\"meta_val.jsonl\"))\n",
    "test_info = read_json(os.path.join(OUT_DATA_DIR, f\"info_val.json\"))\n",
    "mm = mm.reshape(-1, N_TOKENS_AUDIO, 1)\n",
    "assert len(mm) == len(test_metas)\n",
    "assert mm[:100, :, 0].min() >= 0\n",
    "assert mm[:100, :, 0].max() <= 4000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "task_counts = Counter()\n",
    "for test_meta in test_metas:\n",
    "    task_counts[test_meta.get(\"task\")] += 1\n",
    "print(task_counts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.938629Z",
     "start_time": "2024-05-16T13:59:41.938621Z"
    }
   },
   "outputs": [],
   "source": [
    "# # randomly listen to some stuff\n",
    "# from suno_utils.tasks.dac_2c_12cb import preload_models as preload_codec_models\n",
    "# from suno_utils.tasks.dac_2c_12cb import (\n",
    "#     encode as codec_encode,\n",
    "#     decode_stream_to_full_audio as codec_decode,\n",
    "#     EMBEDDING_RATE as CODEC_EMBEDDING_RATE,\n",
    "#     decode as decode\n",
    "# )\n",
    "# os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"0\"\n",
    "# _ = preload_codec_models(\"/app/suno/data/dpo/models/dac_2c_25x12.pt\", device=\"cuda\")\n",
    "# assert len(test_metas) == len(mm)\n",
    "# idx_list = list(range(len(test_metas)))\n",
    "# # random.shuffle(idx_list)\n",
    "# # idx_list = [idx for idx in idx_list if \"text\" in test_metas[idx]]\n",
    "# print(len(mm))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.939205Z",
     "start_time": "2024-05-16T13:59:41.939198Z"
    }
   },
   "outputs": [],
   "source": [
    "# import random\n",
    "# idx = random.choice(test_info[\"perference_0\"][\"idx_list\"])\n",
    "# assert \"original_duration_s\" in test_metas[idx]\n",
    "# # positive index should be shifted by 1\n",
    "# pos_idx = idx + 1\n",
    "# print(\n",
    "#     \"tags:\",\n",
    "#     test_metas[idx].get(\"tags\") == test_metas[pos_idx].get(\"tags\"),\n",
    "#     test_metas[idx].get(\"tags\"),\n",
    "# )\n",
    "# arr = mm[idx, 1:].copy().astype(np.int16)[:, 1:]\n",
    "# pos_arr = mm[pos_idx, 1:].copy().astype(np.int16)[:, 1:]\n",
    "# pad_idx_arr = np.where(arr == COARSE_PAD_TOKEN)[0]\n",
    "# if len(pad_idx_arr) > 0:\n",
    "#     arr = arr[: pad_idx_arr[0], :]\n",
    "# pos_pad_idx_arr = np.where(pos_arr == COARSE_PAD_TOKEN)[0]\n",
    "# if len(pos_pad_idx_arr) > 0:\n",
    "#     pos_arr = pos_arr[: pos_pad_idx_arr[0], :]\n",
    "# a = decode(arr)\n",
    "# print(\"\\n negative example \\n\", test_metas[idx])\n",
    "# a.play(compress=False)\n",
    "# pos_a = decode(pos_arr)\n",
    "# print(\"\\n positive example \\n\", test_metas[pos_idx])\n",
    "# pos_a.play(compress=False)\n",
    "# print(\n",
    "#     \"text:\",\n",
    "#     test_metas[idx].get(\"text\") == test_metas[pos_idx].get(\"text\"),\n",
    "#     test_metas[idx].get(\"text\"),\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.939977Z",
     "start_time": "2024-05-16T13:59:41.939969Z"
    }
   },
   "outputs": [],
   "source": [
    "# val_df[val_df[\"tags\"] == 'a vibrant blend of experimental jazz fusion, drum-and-bass and swagger fuzzed-out guitars']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.940610Z",
     "start_time": "2024-05-16T13:59:41.940603Z"
    }
   },
   "outputs": [],
   "source": [
    "# from collections import Counter\n",
    "# c = Counter()\n",
    "# for _, row in df_slice.iterrows():\n",
    "#     # print(row[\"metadata\"])\n",
    "#     for k in ast.literal_eval(row[\"metadata\"]).keys():\n",
    "#         c[k] += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.941167Z",
     "start_time": "2024-05-16T13:59:41.941159Z"
    }
   },
   "outputs": [],
   "source": [
    "# original_npz_path = f\"/app/suno/data/dpo/7b_npz/{test_metas[idx]['id']}.npz\"\n",
    "# original_npz_path = \"/app/suno/data/dpo/7b_npz/729c3011-f672-4ccd-8d82-1cbf2b52ff69.npz\"\n",
    "# original_arr = np.load(original_npz_path)[\"v2_raw\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.941801Z",
     "start_time": "2024-05-16T13:59:41.941793Z"
    }
   },
   "outputs": [],
   "source": [
    "def validation_on_metas(input_metas):\n",
    "    total_bad = 0\n",
    "    total_good = 0\n",
    "    for idx in range(len(input_metas)):\n",
    "        if idx % 2 == 0:\n",
    "            pos_idx = idx + 1\n",
    "            if input_metas[idx].get(\"tags\") != input_metas[pos_idx].get(\"tags\"):\n",
    "                # print(test_metas[idx].get(\"text\") == test_metas[pos_idx].get(\"text\"), test_metas[idx].get(\"tags\"), test_metas[pos_idx].get(\"tags\"))\n",
    "                total_bad += 1\n",
    "            else:\n",
    "                total_good += 1\n",
    "    print(total_good, total_bad)\n",
    "    return\n",
    "\n",
    "\n",
    "validation_on_metas(test_metas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.942520Z",
     "start_time": "2024-05-16T13:59:41.942511Z"
    }
   },
   "outputs": [],
   "source": [
    "train_info = read_json(os.path.join(OUT_DATA_DIR, f\"info_tr.json\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.943072Z",
     "start_time": "2024-05-16T13:59:41.943065Z"
    }
   },
   "outputs": [],
   "source": [
    "n_neg_tr = train_info[\"perference_0\"][\"idx_list\"]\n",
    "n_pos_tr = train_info[\"perference_1\"][\"idx_list\"]\n",
    "assert len(n_pos_tr) == len(n_neg_tr)\n",
    "# make sure they are offset by 1 and exactly 1\n",
    "for i, j in zip(n_neg_tr, n_pos_tr):\n",
    "    assert i == j - 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.944246Z",
     "start_time": "2024-05-16T13:59:41.944237Z"
    }
   },
   "outputs": [],
   "source": [
    "total_iters = len(n_neg_tr) + len(n_pos_tr)\n",
    "print(\"total samples\", total_iters, train_df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.945249Z",
     "start_time": "2024-05-16T13:59:41.945241Z"
    }
   },
   "outputs": [],
   "source": [
    "print(\"1 epoch per batch 4, total\", total_iters / 8 / 2 / 4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import time\n",
    "\n",
    "time.sleep(60 * 60 * 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.945972Z",
     "start_time": "2024-05-16T13:59:41.945964Z"
    }
   },
   "outputs": [],
   "source": [
    "!cd /home/tony/Work/tony/slurm/bluejay && sbatch sbatch_ipo_bluejay"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import shutil\n",
    "\n",
    "# Basic file copy\n",
    "shutil.copy(\n",
    "    \"/home/tony/Work/tony/Preference/make_dataset_auk_t1.ipynb\",\n",
    "    os.path.join(OUT_DATA_DIR, \"make_dataset.ipynb\"),\n",
    ")\n",
    "print(\"Cache kept!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# some gymathtics loading prev data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-16T13:59:41.946562Z",
     "start_time": "2024-05-16T13:59:41.946555Z"
    }
   },
   "outputs": [],
   "source": [
    "# prev_v3_data = \"/app/suno/data/dpo/7v_v20_full/\"\n",
    "\n",
    "# test_val_metas = read_jsonl(os.path.join(prev_v3_data, f\"meta_val.jsonl\"))\n",
    "# test_tr_metas = read_jsonl(os.path.join(prev_v3_data, f\"meta_tr.jsonl\"))\n",
    "\n",
    "# all_ids = set()\n",
    "# for meta in test_val_metas:\n",
    "#     all_ids.add(meta[\"id\"])\n",
    "# for meta in test_tr_metas:\n",
    "#     all_ids.add(meta[\"id\"])\n",
    "# print(len(all_ids), len(test_val_metas) + len(test_tr_metas))\n",
    "\n",
    "# all_ids = list(all_ids)\n",
    "# with open(\"/home/tony/Data/Preference/7b_v2/7v_v20_full_recut_id.json\", \"w\") as fp:\n",
    "#     json.dump(all_ids, fp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# x_data = train_df[train_df[\"preference\"]][\"similarity\"]\n",
    "# y_data = train_df[~train_df[\"preference\"]][\"similarity\"]\n",
    "# from matplotlib.colors import LogNorm\n",
    "\n",
    "# fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(24, 10))\n",
    "\n",
    "# # 2D Histogram\n",
    "# h = ax1.hist2d(\n",
    "#     x_data,\n",
    "#     y_data,\n",
    "#     bins=(50, 50),\n",
    "#     cmap=\"coolwarm\",\n",
    "#     range=[[0, 1], [0, 1]],\n",
    "#     norm=LogNorm(),\n",
    "# )\n",
    "\n",
    "# ax1.set_xlabel(\"Semantic Distance (Preferred)\")\n",
    "# ax1.set_ylabel(\"Semantic Distance (Non-Preferred)\")\n",
    "# ax1.set_title(\n",
    "#     \"2D Histogram of Semantic Distances: Preferred vs Non-Preferred (Log Scale)\"\n",
    "# )\n",
    "\n",
    "# cbar1 = plt.colorbar(h[3], ax=ax1)\n",
    "# cbar1.set_label(\"Number of Request IDs (Log Scale)\")\n",
    "\n",
    "# # Scatter plot\n",
    "# ax2.scatter(x_data, y_data, alpha=0.1, s=1)\n",
    "# ax2.set_xlabel(\"Semantic Distance (Preferred)\")\n",
    "# ax2.set_ylabel(\"Semantic Distance (Non-Preferred)\")\n",
    "# ax2.set_title(\"Scatter Plot of Semantic Distances: Preferred vs Non-Preferred\")\n",
    "# ax2.set_xlim(0, 1)\n",
    "# ax2.set_ylim(0, 1)\n",
    "\n",
    "# plt.tight_layout()\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# train_metas = read_jsonl(os.path.join(OUT_DATA_DIR, f\"meta_tr.jsonl\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_info.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "suno_env_dev",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.15"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}