{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:58:21.040680Z", "start_time": "2024-05-16T13:58:19.777010Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:36.163942Z", "iopub.status.busy": "2024-09-03T01:03:36.163801Z", "iopub.status.idle": "2024-09-03T01:03:37.744395Z", "shell.execute_reply": "2024-09-03T01:03:37.743847Z", "shell.execute_reply.started": "2024-09-03T01:03:36.163923Z" } }, "outputs": [], "source": [ "import ast\n", "import os\n", "import shutil\n", "import sys\n", "from collections import defaultdict\n", "\n", "import numpy as np\n", "import pandas as pd\n", "from preference_data_preparation_4min_30b 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)" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:58:21.082172Z", "start_time": "2024-05-16T13:58:21.041926Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:37.745231Z", "iopub.status.busy": "2024-09-03T01:03:37.745040Z", "iopub.status.idle": "2024-09-03T01:03:37.770345Z", "shell.execute_reply": "2024-09-03T01:03:37.769896Z", "shell.execute_reply.started": "2024-09-03T01:03:37.745215Z" } }, "outputs": [], "source": [ "OUT_DATA_DIR = \"/app/suno/data/dpo/30b_t3_v7_cycle\"\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 = \"/app/suno/data/dpo/30b_cycle_npz\"" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:58:53.962528Z", "start_time": "2024-05-16T13:58:21.105919Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:37.771071Z", "iopub.status.busy": "2024-09-03T01:03:37.770927Z", "iopub.status.idle": "2024-09-03T01:03:39.280267Z", "shell.execute_reply": "2024-09-03T01:03:39.279683Z", "shell.execute_reply.started": "2024-09-03T01:03:37.771056Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Preference data shape (80478, 68)\n" ] } ], "source": [ "df = pd.read_pickle(\n", " \"/home/tony/Data/Preference/30b_v2/interesting_clips_v4_t_3_20240902_full_with_cer.pkl\"\n", ") # , engine='python')\n", "# df = pd.read_csv(\n", "# \"/home/tony/Data/Preference/30b_v0/interesting_clips_v4_t_1_20240808.csv\"\n", "# ) # , engine='python')\n", "print(\"Preference data shape\", df.shape)" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "execution": { "iopub.execute_input": "2024-09-03T01:03:39.281091Z", "iopub.status.busy": "2024-09-03T01:03:39.280939Z", "iopub.status.idle": "2024-09-03T01:03:39.370933Z", "shell.execute_reply": "2024-09-03T01:03:39.370367Z", "shell.execute_reply.started": "2024-09-03T01:03:39.281075Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(80478, 68)\n", "(80478, 64)\n" ] } ], "source": [ "print(df.shape)\n", "df = df.dropna(axis=1, how=\"all\")\n", "print(df.shape)" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:58:56.199480Z", "start_time": "2024-05-16T13:58:53.963687Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:39.371801Z", "iopub.status.busy": "2024-09-03T01:03:39.371641Z", "iopub.status.idle": "2024-09-03T01:03:39.530245Z", "shell.execute_reply": "2024-09-03T01:03:39.529667Z", "shell.execute_reply.started": "2024-09-03T01:03:39.371786Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "39466\n", "39466\n", "pre-downloaded df (80478, 64)\n", "downloaded df (39466, 64)\n" ] } ], "source": [ "converted_paths = os.listdir(NPZ_DIR)\n", "print(len(converted_paths))\n", "\n", "converted_paths = set([f.replace(\"_gen_cycle.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": 6, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:58:56.467253Z", "start_time": "2024-05-16T13:58:56.207647Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:39.532726Z", "iopub.status.busy": "2024-09-03T01:03:39.532389Z", "iopub.status.idle": "2024-09-03T01:03:39.556760Z", "shell.execute_reply": "2024-09-03T01:03:39.556314Z", "shell.execute_reply.started": "2024-09-03T01:03:39.532709Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "is_30b\n", "True 39466\n", "Name: count, dtype: int64\n" ] }, { "data": { "text/plain": [ "task\n", " 34076\n", "extend 5390\n", "Name: count, dtype: int64" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[\"is_30b\"] = df[\"model_name\"].str.contains(\"-t\")\n", "print(df[\"is_30b\"].value_counts())\n", "df[\"task\"].value_counts()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# LET's do the data prep" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:58:56.592883Z", "start_time": "2024-05-16T13:58:56.470781Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:39.557654Z", "iopub.status.busy": "2024-09-03T01:03:39.557515Z", "iopub.status.idle": "2024-09-03T01:03:39.598355Z", "shell.execute_reply": "2024-09-03T01:03:39.597893Z", "shell.execute_reply.started": "2024-09-03T01:03:39.557640Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "preference model_name \n", "False chirp-v3p5-engine-t-3 19733\n", "True chirp-v3p5-engine-t-3 19733\n", "Name: count, dtype: int64\n", "(39466, 65)\n", "(39466, 65)\n" ] } ], "source": [ "## for 13b this is easy for now\n", "print(df.groupby([\"preference\"])[\"model_name\"].value_counts())\n", "print(df.shape)\n", "df = df[df[\"model_name\"].isin([\"chirp-v3p5-engine-t-3\"])]\n", "print(df.shape)" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:58:56.909539Z", "start_time": "2024-05-16T13:58:56.595736Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:39.599088Z", "iopub.status.busy": "2024-09-03T01:03:39.598946Z", "iopub.status.idle": "2024-09-03T01:03:39.713430Z", "shell.execute_reply": "2024-09-03T01:03:39.712895Z", "shell.execute_reply.started": "2024-09-03T01:03:39.599074Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(39466, 65)\n", "(39466, 65)\n", "preference model_name \n", "False chirp-v3p5-engine-t-3 19733\n", "True chirp-v3p5-engine-t-3 19733\n", "Name: count, dtype: int64\n" ] } ], "source": [ "print(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(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": 9, "metadata": { "execution": { "iopub.execute_input": "2024-09-03T01:03:39.714228Z", "iopub.status.busy": "2024-09-03T01:03:39.714078Z", "iopub.status.idle": "2024-09-03T01:03:49.480209Z", "shell.execute_reply": "2024-09-03T01:03:49.479628Z", "shell.execute_reply.started": "2024-09-03T01:03:39.714213Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "unique_requests 19733\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_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())" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "execution": { "iopub.execute_input": "2024-09-03T01:03:49.481032Z", "iopub.status.busy": "2024-09-03T01:03:49.480886Z", "iopub.status.idle": "2024-09-03T01:03:49.573240Z", "shell.execute_reply": "2024-09-03T01:03:49.572657Z", "shell.execute_reply.started": "2024-09-03T01:03:49.481017Z" } }, "outputs": [], "source": [ "df = df.loc[:, ~df.columns.duplicated()].copy()" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "execution": { "iopub.execute_input": "2024-09-03T01:03:49.574187Z", "iopub.status.busy": "2024-09-03T01:03:49.574027Z", "iopub.status.idle": "2024-09-03T01:03:49.579660Z", "shell.execute_reply": "2024-09-03T01:03:49.579221Z", "shell.execute_reply.started": "2024-09-03T01:03:49.574170Z" } }, "outputs": [ { "data": { "text/plain": [ "task\n", " 34076\n", "extend 5390\n", "Name: count, dtype: int64" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[\"task\"].value_counts()" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "execution": { "iopub.execute_input": "2024-09-03T01:03:49.580475Z", "iopub.status.busy": "2024-09-03T01:03:49.580340Z", "iopub.status.idle": "2024-09-03T01:03:49.691646Z", "shell.execute_reply": "2024-09-03T01:03:49.691175Z", "shell.execute_reply.started": "2024-09-03T01:03:49.580461Z" } }, "outputs": [ { "data": { "text/plain": [ "pos_diff_preference\n", "1.0 13718\n", "2.0 6015\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": "2024-09-03T01:03:49.692433Z", "iopub.status.busy": "2024-09-03T01:03:49.692291Z", "iopub.status.idle": "2024-09-03T01:03:49.708126Z", "shell.execute_reply": "2024-09-03T01:03:49.707623Z", "shell.execute_reply.started": "2024-09-03T01:03:49.692419Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "positive param_experiment\n", "max_cfg_12_ntag_3 4451\n", "max_cfg_12 1130\n", "cfg_12 1056\n", "n_repeat_tags_3 945\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": "2024-09-03T01:03:49.708962Z", "iopub.status.busy": "2024-09-03T01:03:49.708823Z", "iopub.status.idle": "2024-09-03T01:03:49.898587Z", "shell.execute_reply": "2024-09-03T01:03:49.898106Z", "shell.execute_reply.started": "2024-09-03T01:03:49.708949Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "-0.3284\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "df[df[\"preference\"]][\"cer_diff_preference\"].hist(bins=50)\n", "print(df[df[\"preference\"]][\"cer_diff_preference\"].quantile(0.1))\n", "plt.show()" ] }, { "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": "2024-09-03T01:03:49.899409Z", "iopub.status.busy": "2024-09-03T01:03:49.899256Z", "iopub.status.idle": "2024-09-03T01:03:50.162807Z", "shell.execute_reply": "2024-09-03T01:03:50.162220Z", "shell.execute_reply.started": "2024-09-03T01:03:49.899393Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "2216\n", "good_continue_at\n", "True 39466\n", "Name: count, dtype: int64\n", "\n", " Check some basics... \n", " preference\n", "False 19733\n", "True 19733\n", "Name: count, dtype: int64 is_30b\n", "True 39466\n", "Name: count, dtype: int64 model_name\n", "chirp-v3p5-engine-t-3 39466\n", "Name: count, dtype: int64 preference model_name \n", "False chirp-v3p5-engine-t-3 19733\n", "True chirp-v3p5-engine-t-3 19733\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", "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[\"is_30b\"].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()" ] }, { "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": "2024-09-03T01:03:50.163678Z", "iopub.status.busy": "2024-09-03T01:03:50.163523Z", "iopub.status.idle": "2024-09-03T01:03:50.248475Z", "shell.execute_reply": "2024-09-03T01:03:50.247902Z", "shell.execute_reply.started": "2024-09-03T01:03:50.163662Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "negative 19733 positive 17882\n", "total pair requests 19733 --> selected pair requests 17882 frac 0.906\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", "neg_filter_selection_mask = (\n", " (~df[\"preference\"]) # get basics aligned\n", " & (df[\"reaction_play_count\"] >= 1) # has to be played once\n", " # & (df[\"play_count\"] <= 3) # if it is actually bad, shouldn't be listened often\n", " & (df[\"duration\"] >= 10) # can't be too short, otherwise it is obvious\n", " & (df[\"duration\"] <= 240) # can't be badly long\n", " & (df[\"has_continue_and_start_continue_at\"].isna()) # won't have any continues\n", " & (df[\"upvote_count\"] == 0) # won't have any likes\n", " & ((df[\"norm_play_frac\"] <= 3.1))\n", " # & (df[\"dislike_count\"] >= 1) # this is kinda strict\n", " # & (\n", " # (df_slice[\"is_in_playlist\"] == False)\n", " # & (df_slice[\"concat_in_playlist\"] == False)\n", " # ) # can't be part of a playlist -- otherwise there are some like signal in it?\n", ")\n", "pos_filter_selectin_mask = (\n", " (df[\"preference\"]) # get basics aligned\n", " & (\n", " df[\"good_continue_at\"]\n", " ) # if continue, needs to continue off a certain percentage\n", " & (df[\"reaction_play_count\"] >= 1)\n", " & (df[\"play_rel_diff\"] >= 0) # this is more like quality assurance\n", " & (df[\"duration\"] >= 10) # can't be too short, otherwise it is obvious\n", " & (df[\"duration\"] <= 240) # can't be badly long\n", " & (df[\"dislike_count\"] == 0) # can't have dislikes\n", " & (df[\"flag_count\"] == 0) # can't have issues\n", " & (\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\"])\n", " & (df[\"reaction_play_count\"] >= normal_pos_play_count)\n", " # & (df[\"norm_play_frac\"] >= 1.5)\n", " )\n", " \n", " )\n", " # & (df[\"norm_play_frac\"] >= 1.9)\n", " # & (df[\"user_n_clips\"] >= 40) # user needs to have genereated at least 20\n", " # & (df[\"duration_rel_diff\"] < 10) # positive isn't just longer\n", " & ((df[\"upvote_count\"] >= 1) | (df[\"reaction_play_count\"] >= 5)| (df[\"concat_play_counts\"] >= 5))\n", " # & (df[\"pos_diff_preference\"] == 2)\n", " & (df[\"task\"] != \"infill\")\n", " & ((df[\"cer_diff_preference\"] < 0.2) & (df[\"cer\"] < 0.95)) # cut on hoot cer difference and abs cer\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", " \"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", ")" ] }, { "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": "2024-09-03T01:03:50.249316Z", "iopub.status.busy": "2024-09-03T01:03:50.249166Z", "iopub.status.idle": "2024-09-03T01:03:50.359407Z", "shell.execute_reply": "2024-09-03T01:03:50.358824Z", "shell.execute_reply.started": "2024-09-03T01:03:50.249301Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "30b_t3_v7_cycle requests 17882 clips 35764 total khrs 1.915; N gpus for 1000 iters 2.235; 2 gpus for x iters 1117.625; n unique users 14174 n pro users 8122\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\"2 gpus for x iters {df_slice.shape[0] / 8 / 2 / 2:.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", "# 76171 152342 total khrs 2.880 n gpus for 1250 iters 3.809\n", "# v10 has 78866\n", "# v14 has 110402" ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "execution": { "iopub.execute_input": "2024-09-03T01:03:50.360247Z", "iopub.status.busy": "2024-09-03T01:03:50.360097Z", "iopub.status.idle": "2024-09-03T01:03:50.362367Z", "shell.execute_reply": "2024-09-03T01:03:50.361961Z", "shell.execute_reply.started": "2024-09-03T01:03:50.360232Z" } }, "outputs": [], "source": [ "# v1 requests 17681 clips 35362 total khrs 1.815; N gpus for 1500 iters 1.473; 4 gpus for x iters 552.531; n unique users 15179 n pro users 5983" ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.277006Z", "start_time": "2024-05-16T13:59:41.252105Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:50.363043Z", "iopub.status.busy": "2024-09-03T01:03:50.362905Z", "iopub.status.idle": "2024-09-03T01:03:50.404688Z", "shell.execute_reply": "2024-09-03T01:03:50.404214Z", "shell.execute_reply.started": "2024-09-03T01:03:50.363029Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "positive in playlist (2537, 101)\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)" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.323409Z", "start_time": "2024-05-16T13:59:41.278278Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:50.405419Z", "iopub.status.busy": "2024-09-03T01:03:50.405279Z", "iopub.status.idle": "2024-09-03T01:03:50.436782Z", "shell.execute_reply": "2024-09-03T01:03:50.436396Z", "shell.execute_reply.started": "2024-09-03T01:03:50.405406Z" } }, "outputs": [], "source": [ "# interesting_clips_must_be_positive_mask = (\n", "# (df_slice[\"upvoted\"] == True)\n", "# | (df_slice[\"has_action\"] == True)\n", "# | (df_slice[\"part_of_concat\"] == True)\n", "# )\n", "# interesting_clips_must_be_not_negative_mask = (df_slice[\"downvoted\"] == False) # & (df_slice[\"dislike_count\"] < 1)\n", "# interesting_clips_mask = interesting_clips_must_be_positive_mask & interesting_clips_must_be_not_negative_mask\n", "# assert interesting_clips_mask.eq(df_slice[\"preference\"]).all()" ] }, { "cell_type": "code", "execution_count": 21, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.392244Z", "start_time": "2024-05-16T13:59:41.324472Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:50.439648Z", "iopub.status.busy": "2024-09-03T01:03:50.439309Z", "iopub.status.idle": "2024-09-03T01:03:50.471509Z", "shell.execute_reply": "2024-09-03T01:03:50.471117Z", "shell.execute_reply.started": "2024-09-03T01:03:50.439632Z" } }, "outputs": [], "source": [ "# save positive ids\n", "# positive_preference_ids = df_slice[df_slice[\"preference\"] == False][\"s3_id\"].to_json(orient='values')\n", "# with open('/home/tony/Data/Preference/7b_v2/7v_v20_full_recut_id_negative.json', 'w') as file:\n", "# file.write(positive_preference_ids)" ] }, { "cell_type": "code", "execution_count": 22, "metadata": { "execution": { "iopub.execute_input": "2024-09-03T01:03:50.472346Z", "iopub.status.busy": "2024-09-03T01:03:50.472042Z", "iopub.status.idle": "2024-09-03T01:03:50.512139Z", "shell.execute_reply": "2024-09-03T01:03:50.511759Z", "shell.execute_reply.started": "2024-09-03T01:03:50.472332Z" } }, "outputs": [], "source": [ "# df_slice_2 = pd.read_csv(\"/home/tony/Data/Preference/30b_v1/interesting_clips_v4_t_1_20240808_slice.csv\")\n", "# df_total = pd.concat([df_slice, df_slice_2])\n", "# print(df_total.shape)" ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "execution": { "iopub.execute_input": "2024-09-03T01:03:50.512799Z", "iopub.status.busy": "2024-09-03T01:03:50.512671Z", "iopub.status.idle": "2024-09-03T01:03:50.548537Z", "shell.execute_reply": "2024-09-03T01:03:50.548155Z", "shell.execute_reply.started": "2024-09-03T01:03:50.512787Z" } }, "outputs": [], "source": [ "# print(df_slices_2.shape, df_slice.shape)" ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "execution": { "iopub.execute_input": "2024-09-03T01:03:50.549167Z", "iopub.status.busy": "2024-09-03T01:03:50.549046Z", "iopub.status.idle": "2024-09-03T01:03:50.583238Z", "shell.execute_reply": "2024-09-03T01:03:50.582827Z", "shell.execute_reply.started": "2024-09-03T01:03:50.549154Z" } }, "outputs": [], "source": [ "# print(df_total.shape)\n", "# df_slice = df_total.copy()" ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T14:00:20.866354Z", "start_time": "2024-05-16T14:00:12.443344Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:50.583898Z", "iopub.status.busy": "2024-09-03T01:03:50.583767Z", "iopub.status.idle": "2024-09-03T01:03:50.619096Z", "shell.execute_reply": "2024-09-03T01:03:50.618716Z", "shell.execute_reply.started": "2024-09-03T01:03:50.583885Z" } }, "outputs": [], "source": [ "# df_slice.to_csv(\"/home/tony/Data/Preference/30b_v2/interesting_clips_v4_t_3_20240902_slice.csv\", index=False)\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": 26, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.932296Z", "start_time": "2024-05-16T13:59:41.932287Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:50.619721Z", "iopub.status.busy": "2024-09-03T01:03:50.619586Z", "iopub.status.idle": "2024-09-03T01:03:50.714994Z", "shell.execute_reply": "2024-09-03T01:03:50.714557Z", "shell.execute_reply.started": "2024-09-03T01:03:50.619707Z" } }, "outputs": [ { "data": { "text/plain": [ "(15727, 17882)" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# don't have continue at\n", "df_slice[\"request_id\"] = df_slice[\"request_id\"].astype(str)\n", "df_slice[df_slice[\"continue_at\"].isna()][\"request_id\"].nunique(), df_slice[\"request_id\"].nunique()" ] }, { "cell_type": "code", "execution_count": 27, "metadata": { "execution": { "iopub.execute_input": "2024-09-03T01:03:50.715734Z", "iopub.status.busy": "2024-09-03T01:03:50.715589Z", "iopub.status.idle": "2024-09-03T01:03:50.768797Z", "shell.execute_reply": "2024-09-03T01:03:50.768193Z", "shell.execute_reply.started": "2024-09-03T01:03:50.715720Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(35764, 101)\n", "(35764, 101)\n", "(35764, 101)\n" ] } ], "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": 28, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.932966Z", "start_time": "2024-05-16T13:59:41.932957Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:50.769614Z", "iopub.status.busy": "2024-09-03T01:03:50.769468Z", "iopub.status.idle": "2024-09-03T01:03:50.776111Z", "shell.execute_reply": "2024-09-03T01:03:50.775640Z", "shell.execute_reply.started": "2024-09-03T01:03:50.769599Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "17882\n" ] } ], "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": 29, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.933558Z", "start_time": "2024-05-16T13:59:41.933550Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:50.776848Z", "iopub.status.busy": "2024-09-03T01:03:50.776712Z", "iopub.status.idle": "2024-09-03T01:03:50.809709Z", "shell.execute_reply": "2024-09-03T01:03:50.809323Z", "shell.execute_reply.started": "2024-09-03T01:03:50.776835Z" } }, "outputs": [], "source": [ "# df_slice.to_csv(\"/home/tony/Data/Preference/30b_v2/interesting_clips_v4_t_3_20240902_slice.csv\", index=False)" ] }, { "cell_type": "code", "execution_count": 30, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.934277Z", "start_time": "2024-05-16T13:59:41.934268Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:50.810391Z", "iopub.status.busy": "2024-09-03T01:03:50.810259Z", "iopub.status.idle": "2024-09-03T01:03:50.926222Z", "shell.execute_reply": "2024-09-03T01:03:50.925698Z", "shell.execute_reply.started": "2024-09-03T01:03:50.810379Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "17703 179\n", "(35406, 102) (358, 102)\n" ] } ], "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", "\n", "print(train_df.shape, val_df.shape)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Actually make" ] }, { "cell_type": "code", "execution_count": 31, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.934954Z", "start_time": "2024-05-16T13:59:41.934946Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:50.926957Z", "iopub.status.busy": "2024-09-03T01:03:50.926819Z", "iopub.status.idle": "2024-09-03T01:03:50.928995Z", "shell.execute_reply": "2024-09-03T01:03:50.928607Z", "shell.execute_reply.started": "2024-09-03T01:03:50.926943Z" } }, "outputs": [], "source": [ "# val_df[[\"request_id\", \"metadata\", \"updated_at\", \"user_id\", \"preference\"]].head()" ] }, { "cell_type": "code", "execution_count": 32, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.935620Z", "start_time": "2024-05-16T13:59:41.935613Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:50.929668Z", "iopub.status.busy": "2024-09-03T01:03:50.929531Z", "iopub.status.idle": "2024-09-03T01:03:52.306595Z", "shell.execute_reply": "2024-09-03T01:03:52.306007Z", "shell.execute_reply.started": "2024-09-03T01:03:50.929655Z" } }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████| 35406/35406 [00:01<00:00, 26444.15it/s]" ] }, { "name": "stdout", "output_type": "stream", "text": [ "1,896 hours of 35406 clips, 2.212875 nodes, 368.8125 iters\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "\n" ] } ], "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 / 2 / 6} iters\"\n", ")" ] }, { "cell_type": "code", "execution_count": 33, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.936268Z", "start_time": "2024-05-16T13:59:41.936260Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:52.307442Z", "iopub.status.busy": "2024-09-03T01:03:52.307290Z", "iopub.status.idle": "2024-09-03T01:03:56.574817Z", "shell.execute_reply": "2024-09-03T01:03:56.574259Z", "shell.execute_reply.started": "2024-09-03T01:03:52.307427Z" } }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████| 358/358 [00:04<00:00, 85.49it/s]" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Total 358 clips, 3 different prompts\n", "10 hours of False\n", "10 hours of True\n", "Done\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "\n" ] } ], "source": [ "make_dataset(val_df, OUT_DATA_DIR, is_val=True, npz_dir=NPZ_DIR)" ] }, { "cell_type": "code", "execution_count": 34, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.936964Z", "start_time": "2024-05-16T13:59:41.936957Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:03:56.575645Z", "iopub.status.busy": "2024-09-03T01:03:56.575494Z", "iopub.status.idle": "2024-09-03T01:12:10.107735Z", "shell.execute_reply": "2024-09-03T01:12:10.107150Z", "shell.execute_reply.started": "2024-09-03T01:03:56.575630Z" } }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████| 35406/35406 [08:13<00:00, 71.76it/s]\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Total 35406 clips, 260 different prompts\n", "963 hours of False\n", "932 hours of True\n", "Done\n" ] } ], "source": [ "make_dataset(train_df, OUT_DATA_DIR, is_val=False, npz_dir=NPZ_DIR)" ] }, { "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": 35, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.937879Z", "start_time": "2024-05-16T13:59:41.937870Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:12:10.108694Z", "iopub.status.busy": "2024-09-03T01:12:10.108522Z", "iopub.status.idle": "2024-09-03T01:12:10.209204Z", "shell.execute_reply": "2024-09-03T01:12:10.208705Z", "shell.execute_reply.started": "2024-09-03T01:12:10.108677Z" } }, "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, 6016, 13)\n", "assert len(mm) == len(test_metas)\n", "assert mm[:100, :, 0].min() >= 0\n", "assert mm[:100, :, 0].max() <= 4000\n", "assert mm[:100, :, 1:].min() >= 0\n", "assert mm[:100, :, 1:].max() <= 2048" ] }, { "cell_type": "code", "execution_count": 36, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.938629Z", "start_time": "2024-05-16T13:59:41.938621Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:12:10.209980Z", "iopub.status.busy": "2024-09-03T01:12:10.209834Z", "iopub.status.idle": "2024-09-03T01:12:10.212336Z", "shell.execute_reply": "2024-09-03T01:12:10.211926Z", "shell.execute_reply.started": "2024-09-03T01:12:10.209965Z" } }, "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": 37, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.939205Z", "start_time": "2024-05-16T13:59:41.939198Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:12:10.213027Z", "iopub.status.busy": "2024-09-03T01:12:10.212891Z", "iopub.status.idle": "2024-09-03T01:12:10.249113Z", "shell.execute_reply": "2024-09-03T01:12:10.248710Z", "shell.execute_reply.started": "2024-09-03T01:12:10.213013Z" } }, "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": 38, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.939977Z", "start_time": "2024-05-16T13:59:41.939969Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:12:10.249787Z", "iopub.status.busy": "2024-09-03T01:12:10.249657Z", "iopub.status.idle": "2024-09-03T01:12:10.287782Z", "shell.execute_reply": "2024-09-03T01:12:10.287398Z", "shell.execute_reply.started": "2024-09-03T01:12:10.249773Z" } }, "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": 39, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.940610Z", "start_time": "2024-05-16T13:59:41.940603Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:12:10.288461Z", "iopub.status.busy": "2024-09-03T01:12:10.288328Z", "iopub.status.idle": "2024-09-03T01:12:10.326900Z", "shell.execute_reply": "2024-09-03T01:12:10.326519Z", "shell.execute_reply.started": "2024-09-03T01:12:10.288448Z" } }, "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": 40, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.941167Z", "start_time": "2024-05-16T13:59:41.941159Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:12:10.327557Z", "iopub.status.busy": "2024-09-03T01:12:10.327421Z", "iopub.status.idle": "2024-09-03T01:12:10.367451Z", "shell.execute_reply": "2024-09-03T01:12:10.367075Z", "shell.execute_reply.started": "2024-09-03T01:12:10.327543Z" } }, "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": 41, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.941801Z", "start_time": "2024-05-16T13:59:41.941793Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:12:10.368239Z", "iopub.status.busy": "2024-09-03T01:12:10.368107Z", "iopub.status.idle": "2024-09-03T01:12:10.405552Z", "shell.execute_reply": "2024-09-03T01:12:10.405108Z", "shell.execute_reply.started": "2024-09-03T01:12:10.368226Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "179 0\n" ] } ], "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": 42, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.942520Z", "start_time": "2024-05-16T13:59:41.942511Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:12:10.406238Z", "iopub.status.busy": "2024-09-03T01:12:10.406102Z", "iopub.status.idle": "2024-09-03T01:12:10.444244Z", "shell.execute_reply": "2024-09-03T01:12:10.443806Z", "shell.execute_reply.started": "2024-09-03T01:12:10.406224Z" } }, "outputs": [], "source": [ "train_info = read_json(os.path.join(OUT_DATA_DIR, f\"info_tr.json\"))" ] }, { "cell_type": "code", "execution_count": 43, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.943072Z", "start_time": "2024-05-16T13:59:41.943065Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:12:10.444952Z", "iopub.status.busy": "2024-09-03T01:12:10.444818Z", "iopub.status.idle": "2024-09-03T01:12:10.477392Z", "shell.execute_reply": "2024-09-03T01:12:10.476986Z", "shell.execute_reply.started": "2024-09-03T01:12:10.444938Z" } }, "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)" ] }, { "cell_type": "code", "execution_count": 44, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.944246Z", "start_time": "2024-05-16T13:59:41.944237Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:12:10.478051Z", "iopub.status.busy": "2024-09-03T01:12:10.477918Z", "iopub.status.idle": "2024-09-03T01:12:10.517609Z", "shell.execute_reply": "2024-09-03T01:12:10.517192Z", "shell.execute_reply.started": "2024-09-03T01:12:10.478038Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "total samples 35406 (35406, 102)\n" ] } ], "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": 45, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.945249Z", "start_time": "2024-05-16T13:59:41.945241Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:12:10.518282Z", "iopub.status.busy": "2024-09-03T01:12:10.518153Z", "iopub.status.idle": "2024-09-03T01:12:10.555461Z", "shell.execute_reply": "2024-09-03T01:12:10.555049Z", "shell.execute_reply.started": "2024-09-03T01:12:10.518269Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "1 epoch per batch 2, total 553.21875\n" ] } ], "source": [ "print(\"1 epoch per batch 2, total\", total_iters / 8 / 2 / 4)" ] }, { "cell_type": "code", "execution_count": 46, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.945972Z", "start_time": "2024-05-16T13:59:41.945964Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:12:10.556320Z", "iopub.status.busy": "2024-09-03T01:12:10.556189Z", "iopub.status.idle": "2024-09-03T01:12:10.770480Z", "shell.execute_reply": "2024-09-03T01:12:10.769898Z", "shell.execute_reply.started": "2024-09-03T01:12:10.556307Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Submitted batch job 2207\n" ] } ], "source": [ "!cd /home/tony/Work/tony/slurm/30b_dpo && sbatch sbatch_ipo_30b" ] }, { "cell_type": "code", "execution_count": 47, "metadata": { "execution": { "iopub.execute_input": "2024-09-03T01:12:10.771461Z", "iopub.status.busy": "2024-09-03T01:12:10.771301Z", "iopub.status.idle": "2024-09-03T01:12:10.782805Z", "shell.execute_reply": "2024-09-03T01:12:10.782369Z", "shell.execute_reply.started": "2024-09-03T01:12:10.771445Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Cache kept!\n" ] } ], "source": [ "import shutil\n", "\n", "# Basic file copy\n", "shutil.copy('/home/tony/Work/tony/Preference/make_dataset_13b_v4_t3.ipynb', os.path.join(OUT_DATA_DIR, \"make_dataset.ipynb\"))\n", "print(\"Cache kept!\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# some gymathtics loading prev data" ] }, { "cell_type": "code", "execution_count": 48, "metadata": { "ExecuteTime": { "end_time": "2024-05-16T13:59:41.946562Z", "start_time": "2024-05-16T13:59:41.946555Z" }, "execution": { "iopub.execute_input": "2024-09-03T01:12:10.783565Z", "iopub.status.busy": "2024-09-03T01:12:10.783420Z", "iopub.status.idle": "2024-09-03T01:12:10.812246Z", "shell.execute_reply": "2024-09-03T01:12:10.811864Z", "shell.execute_reply.started": "2024-09-03T01:12:10.783551Z" } }, "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": 49, "metadata": { "execution": { "iopub.execute_input": "2024-09-03T01:12:10.812934Z", "iopub.status.busy": "2024-09-03T01:12:10.812805Z", "iopub.status.idle": "2024-09-03T01:12:10.902102Z", "shell.execute_reply": "2024-09-03T01:12:10.901696Z", "shell.execute_reply.started": "2024-09-03T01:12:10.812920Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
level_0indexid_xcreated_atupdated_attime_usedmetadatauser_idstatusrequest_idis_generateds3_idupvote_countbatch_indexmodel_nameprompt_textis_deletedimage_s3_idis_publicdislike_countflag_countplay_countskip_counttitlecontinued_parentdurationsourceclip_typetaskis_pro_useris_in_playlisthas_stemsuser_n_clipsupvoteddownvotedhas_continuedpart_of_concathas_actionflaggeddeletedpos_preferenceneg_preferencediff_preferencepreferencereaction_play_countreaction_pro_play_counttotal_start_stotal_clip_sconcat_play_countsconcat_in_playlistconcat_likesconcat_dislikesstr_idid_ytotal_play_countauto_play_counttotal_play_timetotal_play_count_hourauto_play_count_hourtotal_play_time_hourp_datep_hournorm_play_fraccleaned_prompt_textceris_30btagstypeis_botpromptstreampriorityexperimentgpt_promptnegative_tagsrefund_creditsmake_instrumentalexperiment_versiongpt_description_promptparam_experimentinfillhistoryoptionscontinue_atcontinued_from_prompthas_vocalconcat_historyis_audio_upload_tos_acceptedcontinued_aligned_promptaudio_prompt_idstem_typestem_from_iderror_typeerror_messagepos_diff_preferencecer_diff_preferenceidoriginal_duration_shas_continue_and_start_continue_atgood_continue_atduration_rel_diffplay_rel_diff
0636206800781ca5b266-fadd-4bb3-8db0-13274ee96c202024-08-31 04:31:44.622797+00:002024-08-31 04:31:44.622804+00:00107.232122{'tags': '80s, Synthwave, punk wave, ele...14886814complete00d00a69-e7f7-49cf-acee-81e4e5943f65True1ca5b266-fadd-4bb3-8db0-13274ee96c2001chirp-v3p5-engine-t-3Free Palestine\\nFree Palestine\\n\\n[Piano Solo]...Falseimage_1ca5b266-fadd-4bb3-8db0-13274ee96c20False0010Sunshine Dreams0c1e9ca4-d59c-43a7-bc1a-8b25e06049af120.00webgenextendTrueFalseFalse164FalseFalseFalseFalseFalseFalseFalseFalseFalse0False1.01.0NaNNaNNaNNoneNaNNaN1ca5b266-fadd-4bb3-8db0-13274ee96c201.984543e+111.01.057.7452530.00.00.02024-09-02190.481210free palestine free palestine long live palest...0.5126True80s, Synthwave, punk wave, electronic, ...genFalseFree Palestine\\nFree Palestine\\n\\n[Piano Solo]...True10[chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3]NoneFalseFalsev_39NoneNaNFalse[{'id': '0c1e9ca4-d59c-43a7-bc1a-8b25e06049af'...NaN17.0Free Palestine\\nFree Palestine\\n\\n[Piano Solo]...NaNNaNNaNNoneNaNNaNNaNNaNNaN0.00.51261ca5b266-fadd-4bb3-8db0-13274ee96c20NaNNaNTrue-120.00-6.0
163720680077dfc2dd38-34bc-40e6-a1f6-e6240a83776d2024-08-31 04:31:44.622684+00:002024-08-31 04:31:44.622700+00:00111.774462{'tags': '80s, Synthwave, punk wave, ele...14886814complete00d00a69-e7f7-49cf-acee-81e4e5943f65Truedfc2dd38-34bc-40e6-a1f6-e6240a83776d10chirp-v3p5-engine-t-3Free Palestine\\nFree Palestine\\n\\n[Piano Solo]...Falseimage_dfc2dd38-34bc-40e6-a1f6-e6240a83776dFalse0050Sunshine Dreams0c1e9ca4-d59c-43a7-bc1a-8b25e06049af120.00webgenextendTrueFalseFalse164TrueFalseFalseTrueFalseFalseFalseTrueFalse1True5.05.017.0136.9999794.0False1.00.0dfc2dd38-34bc-40e6-a1f6-e6240a83776d1.983854e+115.00.0439.9867120.00.00.02024-09-02193.666556free palestine free palestine long live palest...0.5462True80s, Synthwave, punk wave, electronic, ...genFalseFree Palestine\\nFree Palestine\\n\\n[Piano Solo]...True10[chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3]NoneFalseFalsev_39NoneNaNFalse[{'id': '0c1e9ca4-d59c-43a7-bc1a-8b25e06049af'...NaN17.0Free Palestine\\nFree Palestine\\n\\n[Piano Solo]...NaNNaNNaNNoneNaNNaNNaNNaNNaN1.00.0336dfc2dd38-34bc-40e6-a1f6-e6240a83776d137.0NaNTrue0.004.0
2108633363430d7ab351-b054-40b2-a2f4-877a1e478be22024-08-27 14:53:45.101812+00:002024-08-27 14:53:45.101818+00:0085.035722{'tags': 'high beat rap', 'task': None, 'type'...6026074complete01611dd1-8a94-4475-a9d8-7845b5fb4ecdTrue0d7ab351-b054-40b2-a2f4-877a1e478be201chirp-v3p5-engine-t-3[Verse]\\nCity of dreams, man, the hustle never...Falseimage_0d7ab351-b054-40b2-a2f4-877a1e478be2False1010Mumbai HustleNone95.96webgenFalseFalseFalse63FalseTrueFalseFalseFalseFalseFalseFalseTrue-1False1.0NaNNaNNaNNaNNoneNaNNaN0d7ab351-b054-40b2-a2f4-877a1e478be21.984515e+111.00.013.6480860.00.00.02024-09-02190.142227city of dreams man the hustle never ends from ...0.2231Truehigh beat rapgenFalse[Verse]\\nCity of dreams, man, the hustle never...True0[chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3]NoneNoneFalseFalsev_36Write a rap on hustling in Mumbai city in Engl...NaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaN-2.0-0.04650d7ab351-b054-40b2-a2f4-877a1e478be2NaNNaNTrue-23.04-13.0
310873336342fa48cb45-a21b-45e1-9e9d-a5a2f10c8e282024-08-27 14:53:45.101705+00:002024-08-27 14:53:45.101720+00:00191.813621{'tags': 'high beat rap', 'task': None, 'type'...6026074complete01611dd1-8a94-4475-a9d8-7845b5fb4ecdTruefa48cb45-a21b-45e1-9e9d-a5a2f10c8e2810chirp-v3p5-engine-t-3[Verse]\\nCity of dreams, man, the hustle never...Falseimage_fa48cb45-a21b-45e1-9e9d-a5a2f10c8e28False0040Mumbai HustleNone227.72webgenFalseFalseFalse63TrueFalseFalseFalseFalseFalseFalseTrueFalse1True4.0NaNNaNNaNNaNNoneNaNNaNfa48cb45-a21b-45e1-9e9d-a5a2f10c8e281.984127e+113.00.0257.2731960.00.00.02024-09-02191.129779city of dreams man the hustle never ends from ...0.2081Truehigh beat rapgenFalse[Verse]\\nCity of dreams, man, the hustle never...True0[chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3]NoneNoneFalseFalsev_36Write a rap on hustling in Mumbai city in Engl...cfg_12NaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaN2.0-0.0150fa48cb45-a21b-45e1-9e9d-a5a2f10c8e28NaNNaNTrue131.763.0
41092142011716acd28f7-5dd3-4891-bee0-d1ab84c5d1c02024-08-29 20:02:16.198929+00:002024-08-29 20:02:16.198936+00:00200.582993{'tags': 'Tech house remix introperspective Ba...599094complete016250b5-11b4-456d-a3dd-61b5e396f03eTrue6acd28f7-5dd3-4891-bee0-d1ab84c5d1c001chirp-v3p5-engine-t-3[Verse]\\nTake off your clothes\\nTake it all of...Falseimage_6acd28f7-5dd3-4891-bee0-d1ab84c5d1c0False0010Today Tomorrow EverydayNone240.00webgenFalseFalseFalse36FalseFalseFalseFalseFalseFalseFalseFalseFalse0False1.0NaNNaNNaNNaNNoneNaNNaN6acd28f7-5dd3-4891-bee0-d1ab84c5d1c01.983523e+111.00.014.0180920.00.00.02024-09-02190.058409take off your clothes take it all off today to...0.8239TrueTech house remix introperspective Bass clubgenFalse[Verse]\\nTake off your clothes\\nTake it all of...True0[chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3]NoneFalseFalsev_37NoneNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaN-1.00.82396acd28f7-5dd3-4891-bee0-d1ab84c5d1c0NaNNaNTrue184.00-5.0
510931420117002fa6991-5f01-4636-a3f8-b015354affa92024-08-29 20:02:16.198793+00:002024-08-29 20:02:16.198815+00:00223.832865{'tags': 'Tech house remix introperspective Ba...599094complete016250b5-11b4-456d-a3dd-61b5e396f03eTrue02fa6991-5f01-4636-a3f8-b015354affa910chirp-v3p5-engine-t-3[Verse]\\nTake off your clothes\\nTake it all of...Falseimage_02fa6991-5f01-4636-a3f8-b015354affa9False0080Today Tomorrow EverydayNone238.96webgenFalseFalseFalse36TrueFalseFalseFalseFalseFalseFalseTrueFalse1True8.0NaNNaNNaNNaNNoneNaNNaN02fa6991-5f01-4636-a3f8-b015354affa91.985729e+118.00.0419.9771060.00.00.02024-09-02191.757521take off your clothes take it all off today to...0.4155TrueTech house remix introperspective Bass clubgenFalse[Verse]\\nTake off your clothes\\nTake it all of...True0[chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3]NoneFalseFalsev_37Nonemax_cfg_12_ntag_3NaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaN1.0-0.408402fa6991-5f01-4636-a3f8-b015354affa9NaNNaNTrue-1.047.0
6129628464188a914b325-d74c-4fec-938a-6db0944fd5842024-09-01 20:27:49.874962+00:002024-09-01 20:27:49.874968+00:00210.299322{'tags': 'blues male vocals, violin, acoustic ...1499044complete019b1def-aa22-47ee-9681-2d5347378479Truea914b325-d74c-4fec-938a-6db0944fd58401chirp-v3p5-engine-t-3[Verse]\\nUnder the moonlight glow\\nWhere shado...Falseimage_a914b325-d74c-4fec-938a-6db0944fd584False0030Gold for SoulsNone240.00webgenTrueFalseFalse339FalseFalseFalseFalseFalseFalseFalseFalseFalse0False3.03.0NaNNaNNaNNoneNaNNaNa914b325-d74c-4fec-938a-6db0944fd5841.983693e+113.01.063.3376220.00.00.02024-09-02190.263907under the moonlight glow where shadows softly ...0.1619Trueblues male vocals, violin, acoustic rock, melo...genFalse[Verse]\\nUnder the moonlight glow\\nWhere shado...True10[chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3]NoneFalseFalsev_39NoneNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaN0.00.1619a914b325-d74c-4fec-938a-6db0944fd584NaNNaNTrue116.00-4.0
7129728464187f226a18b-aed7-4e82-bcaa-5fed4126e25d2024-09-01 20:27:49.874878+00:002024-09-01 20:27:49.874889+00:00206.660231{'tags': 'blues male vocals, violin, acoustic ...1499044complete019b1def-aa22-47ee-9681-2d5347378479Truef226a18b-aed7-4e82-bcaa-5fed4126e25d10chirp-v3p5-engine-t-3[Verse]\\nUnder the moonlight glow\\nWhere shado...Falsef226a18b-aed7-4e82-bcaa-5fed4126e25d_907feb54False0050Gold for SoulsNone237.80webgenTrueFalseFalse339TrueFalseFalseFalseFalseFalseFalseTrueFalse1True5.05.0NaNNaNNaNNoneNaNNaNf226a18b-aed7-4e82-bcaa-5fed4126e25d1.985552e+115.00.0469.6740130.00.00.02024-09-02191.975080under the moonlight glow where shadows softly ...0.1123Trueblues male vocals, violin, acoustic rock, melo...genFalse[Verse]\\nUnder the moonlight glow\\nWhere shado...True10[chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3]NoneFalseFalsev_39Nonemax_cfg_12_ntag_3NaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaN1.0-0.0496f226a18b-aed7-4e82-bcaa-5fed4126e25dNaNNaNTrue-2.202.0
82644215461988e0bde2c-9f8e-4739-86a2-5160244921582024-08-31 09:52:01.959526+00:002024-08-31 09:52:01.959548+00:00215.618332{'tags': 'Krautrock, Berlin School, Kosmische ...24249474complete0340e772-5bff-4700-aa56-b825900bd0e9True8e0bde2c-9f8e-4739-86a2-51602449215800chirp-v3p5-engine-t-3[Short Instrumental Intro]\\n\\n[Verse 1]\\nFortu...Falseimage_8e0bde2c-9f8e-4739-86a2-516024492158False0020Fortunas RadNone240.00webgenTrueFalseFalse193FalseFalseFalseFalseFalseFalseFalseFalseFalse0False2.02.0NaNNaNNaNNoneNaNNaN8e0bde2c-9f8e-4739-86a2-5160244921581.984647e+112.00.0192.8965650.00.00.02024-09-02190.803736fortunas rad du drehst so schnell wie der mond...0.0647TrueKrautrock, Berlin School, Kosmische Musik, Sy...genFalse[Short Instrumental Intro]\\n\\n[Verse 1]\\nFortu...True10[chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3]NoneFalseFalsev_39Nonemax_cfg_12_ntag_3NaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaN0.00.06478e0bde2c-9f8e-4739-86a2-516024492158NaNNaNTrue0.00-3.0
9264521546199edff3263-8e98-4cf5-a9e2-547396b3bcdf2024-08-31 09:52:01.959663+00:002024-08-31 09:52:01.959672+00:00209.053880{'tags': 'Krautrock, Berlin School, Kosmische ...24249474complete0340e772-5bff-4700-aa56-b825900bd0e9Trueedff3263-8e98-4cf5-a9e2-547396b3bcdf01chirp-v3p5-engine-t-3[Short Instrumental Intro]\\n\\n[Verse 1]\\nFortu...Falseimage_edff3263-8e98-4cf5-a9e2-547396b3bcdfFalse0030Fortunas RadNone240.00webgenTrueFalseFalse193FalseFalseTrueTrueFalseFalseFalseTrueFalse1True3.03.00.0258.9999798.0False1.00.0edff3263-8e98-4cf5-a9e2-547396b3bcdf1.985132e+113.00.0277.7295970.00.00.02024-09-02191.157207fortunas rad du drehst so schnell wie der mond...0.0758TrueKrautrock, Berlin School, Kosmische Musik, Sy...genFalse[Short Instrumental Intro]\\n\\n[Verse 1]\\nFortu...True10[chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3]NoneFalseFalsev_39NoneNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaNNaN1.00.0111edff3263-8e98-4cf5-a9e2-547396b3bcdf240.0NaNTrue0.001.0
\n", "
" ], "text/plain": [ " level_0 index id_x created_at updated_at time_used metadata user_id status request_id is_generated s3_id upvote_count batch_index model_name prompt_text is_deleted image_s3_id is_public dislike_count flag_count play_count skip_count title continued_parent duration source clip_type task is_pro_user is_in_playlist has_stems user_n_clips upvoted downvoted has_continued part_of_concat has_action flagged deleted pos_preference neg_preference diff_preference preference reaction_play_count reaction_pro_play_count total_start_s total_clip_s concat_play_counts concat_in_playlist concat_likes concat_dislikes str_id \\\n", "0 636 20680078 1ca5b266-fadd-4bb3-8db0-13274ee96c20 2024-08-31 04:31:44.622797+00:00 2024-08-31 04:31:44.622804+00:00 107.232122 {'tags': '80s, Synthwave, punk wave, ele... 14886814 complete 00d00a69-e7f7-49cf-acee-81e4e5943f65 True 1ca5b266-fadd-4bb3-8db0-13274ee96c20 0 1 chirp-v3p5-engine-t-3 Free Palestine\\nFree Palestine\\n\\n[Piano Solo]... False image_1ca5b266-fadd-4bb3-8db0-13274ee96c20 False 0 0 1 0 Sunshine Dreams 0c1e9ca4-d59c-43a7-bc1a-8b25e06049af 120.00 web gen extend True False False 164 False False False False False False False False False 0 False 1.0 1.0 NaN NaN NaN None NaN NaN 1ca5b266-fadd-4bb3-8db0-13274ee96c20 \n", "1 637 20680077 dfc2dd38-34bc-40e6-a1f6-e6240a83776d 2024-08-31 04:31:44.622684+00:00 2024-08-31 04:31:44.622700+00:00 111.774462 {'tags': '80s, Synthwave, punk wave, ele... 14886814 complete 00d00a69-e7f7-49cf-acee-81e4e5943f65 True dfc2dd38-34bc-40e6-a1f6-e6240a83776d 1 0 chirp-v3p5-engine-t-3 Free Palestine\\nFree Palestine\\n\\n[Piano Solo]... False image_dfc2dd38-34bc-40e6-a1f6-e6240a83776d False 0 0 5 0 Sunshine Dreams 0c1e9ca4-d59c-43a7-bc1a-8b25e06049af 120.00 web gen extend True False False 164 True False False True False False False True False 1 True 5.0 5.0 17.0 136.999979 4.0 False 1.0 0.0 dfc2dd38-34bc-40e6-a1f6-e6240a83776d \n", "2 1086 3336343 0d7ab351-b054-40b2-a2f4-877a1e478be2 2024-08-27 14:53:45.101812+00:00 2024-08-27 14:53:45.101818+00:00 85.035722 {'tags': 'high beat rap', 'task': None, 'type'... 6026074 complete 01611dd1-8a94-4475-a9d8-7845b5fb4ecd True 0d7ab351-b054-40b2-a2f4-877a1e478be2 0 1 chirp-v3p5-engine-t-3 [Verse]\\nCity of dreams, man, the hustle never... False image_0d7ab351-b054-40b2-a2f4-877a1e478be2 False 1 0 1 0 Mumbai Hustle None 95.96 web gen False False False 63 False True False False False False False False True -1 False 1.0 NaN NaN NaN NaN None NaN NaN 0d7ab351-b054-40b2-a2f4-877a1e478be2 \n", "3 1087 3336342 fa48cb45-a21b-45e1-9e9d-a5a2f10c8e28 2024-08-27 14:53:45.101705+00:00 2024-08-27 14:53:45.101720+00:00 191.813621 {'tags': 'high beat rap', 'task': None, 'type'... 6026074 complete 01611dd1-8a94-4475-a9d8-7845b5fb4ecd True fa48cb45-a21b-45e1-9e9d-a5a2f10c8e28 1 0 chirp-v3p5-engine-t-3 [Verse]\\nCity of dreams, man, the hustle never... False image_fa48cb45-a21b-45e1-9e9d-a5a2f10c8e28 False 0 0 4 0 Mumbai Hustle None 227.72 web gen False False False 63 True False False False False False False True False 1 True 4.0 NaN NaN NaN NaN None NaN NaN fa48cb45-a21b-45e1-9e9d-a5a2f10c8e28 \n", "4 1092 14201171 6acd28f7-5dd3-4891-bee0-d1ab84c5d1c0 2024-08-29 20:02:16.198929+00:00 2024-08-29 20:02:16.198936+00:00 200.582993 {'tags': 'Tech house remix introperspective Ba... 599094 complete 016250b5-11b4-456d-a3dd-61b5e396f03e True 6acd28f7-5dd3-4891-bee0-d1ab84c5d1c0 0 1 chirp-v3p5-engine-t-3 [Verse]\\nTake off your clothes\\nTake it all of... False image_6acd28f7-5dd3-4891-bee0-d1ab84c5d1c0 False 0 0 1 0 Today Tomorrow Everyday None 240.00 web gen False False False 36 False False False False False False False False False 0 False 1.0 NaN NaN NaN NaN None NaN NaN 6acd28f7-5dd3-4891-bee0-d1ab84c5d1c0 \n", "5 1093 14201170 02fa6991-5f01-4636-a3f8-b015354affa9 2024-08-29 20:02:16.198793+00:00 2024-08-29 20:02:16.198815+00:00 223.832865 {'tags': 'Tech house remix introperspective Ba... 599094 complete 016250b5-11b4-456d-a3dd-61b5e396f03e True 02fa6991-5f01-4636-a3f8-b015354affa9 1 0 chirp-v3p5-engine-t-3 [Verse]\\nTake off your clothes\\nTake it all of... False image_02fa6991-5f01-4636-a3f8-b015354affa9 False 0 0 8 0 Today Tomorrow Everyday None 238.96 web gen False False False 36 True False False False False False False True False 1 True 8.0 NaN NaN NaN NaN None NaN NaN 02fa6991-5f01-4636-a3f8-b015354affa9 \n", "6 1296 28464188 a914b325-d74c-4fec-938a-6db0944fd584 2024-09-01 20:27:49.874962+00:00 2024-09-01 20:27:49.874968+00:00 210.299322 {'tags': 'blues male vocals, violin, acoustic ... 1499044 complete 019b1def-aa22-47ee-9681-2d5347378479 True a914b325-d74c-4fec-938a-6db0944fd584 0 1 chirp-v3p5-engine-t-3 [Verse]\\nUnder the moonlight glow\\nWhere shado... False image_a914b325-d74c-4fec-938a-6db0944fd584 False 0 0 3 0 Gold for Souls None 240.00 web gen True False False 339 False False False False False False False False False 0 False 3.0 3.0 NaN NaN NaN None NaN NaN a914b325-d74c-4fec-938a-6db0944fd584 \n", "7 1297 28464187 f226a18b-aed7-4e82-bcaa-5fed4126e25d 2024-09-01 20:27:49.874878+00:00 2024-09-01 20:27:49.874889+00:00 206.660231 {'tags': 'blues male vocals, violin, acoustic ... 1499044 complete 019b1def-aa22-47ee-9681-2d5347378479 True f226a18b-aed7-4e82-bcaa-5fed4126e25d 1 0 chirp-v3p5-engine-t-3 [Verse]\\nUnder the moonlight glow\\nWhere shado... False f226a18b-aed7-4e82-bcaa-5fed4126e25d_907feb54 False 0 0 5 0 Gold for Souls None 237.80 web gen True False False 339 True False False False False False False True False 1 True 5.0 5.0 NaN NaN NaN None NaN NaN f226a18b-aed7-4e82-bcaa-5fed4126e25d \n", "8 2644 21546198 8e0bde2c-9f8e-4739-86a2-516024492158 2024-08-31 09:52:01.959526+00:00 2024-08-31 09:52:01.959548+00:00 215.618332 {'tags': 'Krautrock, Berlin School, Kosmische ... 24249474 complete 0340e772-5bff-4700-aa56-b825900bd0e9 True 8e0bde2c-9f8e-4739-86a2-516024492158 0 0 chirp-v3p5-engine-t-3 [Short Instrumental Intro]\\n\\n[Verse 1]\\nFortu... False image_8e0bde2c-9f8e-4739-86a2-516024492158 False 0 0 2 0 Fortunas Rad None 240.00 web gen True False False 193 False False False False False False False False False 0 False 2.0 2.0 NaN NaN NaN None NaN NaN 8e0bde2c-9f8e-4739-86a2-516024492158 \n", "9 2645 21546199 edff3263-8e98-4cf5-a9e2-547396b3bcdf 2024-08-31 09:52:01.959663+00:00 2024-08-31 09:52:01.959672+00:00 209.053880 {'tags': 'Krautrock, Berlin School, Kosmische ... 24249474 complete 0340e772-5bff-4700-aa56-b825900bd0e9 True edff3263-8e98-4cf5-a9e2-547396b3bcdf 0 1 chirp-v3p5-engine-t-3 [Short Instrumental Intro]\\n\\n[Verse 1]\\nFortu... False image_edff3263-8e98-4cf5-a9e2-547396b3bcdf False 0 0 3 0 Fortunas Rad None 240.00 web gen True False False 193 False False True True False False False True False 1 True 3.0 3.0 0.0 258.999979 8.0 False 1.0 0.0 edff3263-8e98-4cf5-a9e2-547396b3bcdf \n", "\n", " id_y total_play_count auto_play_count total_play_time total_play_count_hour auto_play_count_hour total_play_time_hour p_date p_hour norm_play_frac cleaned_prompt_text cer is_30b tags type is_bot prompt stream priority experiment gpt_prompt negative_tags refund_credits make_instrumental experiment_version gpt_description_prompt param_experiment infill history options continue_at continued_from_prompt has_vocal concat_history is_audio_upload_tos_accepted continued_aligned_prompt audio_prompt_id stem_type stem_from_id error_type error_message pos_diff_preference cer_diff_preference id original_duration_s has_continue_and_start_continue_at good_continue_at \\\n", "0 1.984543e+11 1.0 1.0 57.745253 0.0 0.0 0.0 2024-09-02 19 0.481210 free palestine free palestine long live palest... 0.5126 True 80s, Synthwave, punk wave, electronic, ... gen False Free Palestine\\nFree Palestine\\n\\n[Piano Solo]... True 10 [chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3] None False False v_39 None NaN False [{'id': '0c1e9ca4-d59c-43a7-bc1a-8b25e06049af'... NaN 17.0 Free Palestine\\nFree Palestine\\n\\n[Piano Solo]... NaN NaN NaN None NaN NaN NaN NaN NaN 0.0 0.5126 1ca5b266-fadd-4bb3-8db0-13274ee96c20 NaN NaN True \n", "1 1.983854e+11 5.0 0.0 439.986712 0.0 0.0 0.0 2024-09-02 19 3.666556 free palestine free palestine long live palest... 0.5462 True 80s, Synthwave, punk wave, electronic, ... gen False Free Palestine\\nFree Palestine\\n\\n[Piano Solo]... True 10 [chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3] None False False v_39 None NaN False [{'id': '0c1e9ca4-d59c-43a7-bc1a-8b25e06049af'... NaN 17.0 Free Palestine\\nFree Palestine\\n\\n[Piano Solo]... NaN NaN NaN None NaN NaN NaN NaN NaN 1.0 0.0336 dfc2dd38-34bc-40e6-a1f6-e6240a83776d 137.0 NaN True \n", "2 1.984515e+11 1.0 0.0 13.648086 0.0 0.0 0.0 2024-09-02 19 0.142227 city of dreams man the hustle never ends from ... 0.2231 True high beat rap gen False [Verse]\\nCity of dreams, man, the hustle never... True 0 [chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3] None None False False v_36 Write a rap on hustling in Mumbai city in Engl... NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN -2.0 -0.0465 0d7ab351-b054-40b2-a2f4-877a1e478be2 NaN NaN True \n", "3 1.984127e+11 3.0 0.0 257.273196 0.0 0.0 0.0 2024-09-02 19 1.129779 city of dreams man the hustle never ends from ... 0.2081 True high beat rap gen False [Verse]\\nCity of dreams, man, the hustle never... True 0 [chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3] None None False False v_36 Write a rap on hustling in Mumbai city in Engl... cfg_12 NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN 2.0 -0.0150 fa48cb45-a21b-45e1-9e9d-a5a2f10c8e28 NaN NaN True \n", "4 1.983523e+11 1.0 0.0 14.018092 0.0 0.0 0.0 2024-09-02 19 0.058409 take off your clothes take it all off today to... 0.8239 True Tech house remix introperspective Bass club gen False [Verse]\\nTake off your clothes\\nTake it all of... True 0 [chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3] None False False v_37 None NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN -1.0 0.8239 6acd28f7-5dd3-4891-bee0-d1ab84c5d1c0 NaN NaN True \n", "5 1.985729e+11 8.0 0.0 419.977106 0.0 0.0 0.0 2024-09-02 19 1.757521 take off your clothes take it all off today to... 0.4155 True Tech house remix introperspective Bass club gen False [Verse]\\nTake off your clothes\\nTake it all of... True 0 [chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3] None False False v_37 None max_cfg_12_ntag_3 NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN 1.0 -0.4084 02fa6991-5f01-4636-a3f8-b015354affa9 NaN NaN True \n", "6 1.983693e+11 3.0 1.0 63.337622 0.0 0.0 0.0 2024-09-02 19 0.263907 under the moonlight glow where shadows softly ... 0.1619 True blues male vocals, violin, acoustic rock, melo... gen False [Verse]\\nUnder the moonlight glow\\nWhere shado... True 10 [chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3] None False False v_39 None NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN 0.0 0.1619 a914b325-d74c-4fec-938a-6db0944fd584 NaN NaN True \n", "7 1.985552e+11 5.0 0.0 469.674013 0.0 0.0 0.0 2024-09-02 19 1.975080 under the moonlight glow where shadows softly ... 0.1123 True blues male vocals, violin, acoustic rock, melo... gen False [Verse]\\nUnder the moonlight glow\\nWhere shado... True 10 [chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3] None False False v_39 None max_cfg_12_ntag_3 NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN 1.0 -0.0496 f226a18b-aed7-4e82-bcaa-5fed4126e25d NaN NaN True \n", "8 1.984647e+11 2.0 0.0 192.896565 0.0 0.0 0.0 2024-09-02 19 0.803736 fortunas rad du drehst so schnell wie der mond... 0.0647 True Krautrock, Berlin School, Kosmische Musik, Sy... gen False [Short Instrumental Intro]\\n\\n[Verse 1]\\nFortu... True 10 [chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3] None False False v_39 None max_cfg_12_ntag_3 NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN 0.0 0.0647 8e0bde2c-9f8e-4739-86a2-516024492158 NaN NaN True \n", "9 1.985132e+11 3.0 0.0 277.729597 0.0 0.0 0.0 2024-09-02 19 1.157207 fortunas rad du drehst so schnell wie der mond... 0.0758 True Krautrock, Berlin School, Kosmische Musik, Sy... gen False [Short Instrumental Intro]\\n\\n[Verse 1]\\nFortu... True 10 [chirp-v3p5-engine-t-3, chirp-v3p5-engine-t-3] None False False v_39 None NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN 1.0 0.0111 edff3263-8e98-4cf5-a9e2-547396b3bcdf 240.0 NaN True \n", "\n", " duration_rel_diff play_rel_diff \n", "0 -120.00 -6.0 \n", "1 0.00 4.0 \n", "2 -23.04 -13.0 \n", "3 131.76 3.0 \n", "4 184.00 -5.0 \n", "5 -1.04 7.0 \n", "6 116.00 -4.0 \n", "7 -2.20 2.0 \n", "8 0.00 -3.0 \n", "9 0.00 1.0 " ] }, "execution_count": 49, "metadata": {}, "output_type": "execute_result" } ], "source": [ "val_df.head(n=10)" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.10.14" }, "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 }