{ "cells": [ { "cell_type": "code", "execution_count": null, "id": "0", "metadata": {}, "outputs": [], "source": [ "from typing import Any\n", "import pandas as pd" ] }, { "cell_type": "code", "execution_count": null, "id": "1", "metadata": {}, "outputs": [], "source": [ "def map_control_sliders_to_df(df: pd.DataFrame) -> pd.DataFrame:\n", " \"\"\"Extracts 'style_weight' and 'weirdness_constraint' from the 'control_sliders' dict in the 'metadata' column\n", " and adds them as new columns to the DataFrame.\n", "\n", " Args:\n", " df (pd.DataFrame): DataFrame with a 'metadata' column containing a 'control_sliders' dict.\n", "\n", " Returns:\n", " pd.DataFrame: DataFrame with added 'style_weight' and 'weirdness_constraint' columns.\n", "\n", " Raises:\n", " KeyError: If 'control_sliders', 'style_weight', or 'weirdness_constraint' are missing in any row.\n", " TypeError: If the extracted values are not floats.\n", "\n", " Example:\n", " >>> import pandas as pd\n", " >>> data = [{'metadata': {'control_sliders': {'style_weight': 0.89, 'weirdness_constraint': 0.8}}}]\n", " >>> df = pd.DataFrame(data)\n", " >>> df = map_control_sliders_to_df(df)\n", " >>> df[['style_weight', 'weirdness_constraint']].iloc[0].tolist()\n", " [0.89, 0.8]\n", " \"\"\"\n", " # Vectorized extraction for performance\n", " sliders = df[\"metadata\"].map(lambda m: m.get(\"control_sliders\", {}))\n", " style_weight = sliders.map(lambda s: s.get(\"style_weight\", None))\n", " weirdness_constraint = sliders.map(lambda s: s.get(\"weirdness_constraint\", None))\n", " audio_weight = sliders.map(lambda s: s.get(\"audio_weight\", None))\n", "\n", " df[\"style_weight\"] = style_weight\n", " df[\"weirdness_constraint\"] = weirdness_constraint\n", " df[\"audio_weight\"] = audio_weight\n", " return df" ] }, { "cell_type": "code", "execution_count": null, "id": "2", "metadata": {}, "outputs": [], "source": [ "df = pd.read_pickle(\"/home/tony/Data/Preference/crow_t1/interesting_clips_crow_t1_20250930.pkl\")" ] }, { "cell_type": "code", "execution_count": null, "id": "3", "metadata": {}, "outputs": [], "source": [ "df = map_control_sliders_to_df(df)\n", "df.head()\n" ] }, { "cell_type": "code", "execution_count": null, "id": "4", "metadata": {}, "outputs": [], "source": [ "# Convert list of pairs to a dataframe\n", "# Each row represents one pair with prefixed columns for each item\n", "pairs_data = []\n", "for pair in pairs:\n", " item1, item2 = pair\n", " row = {}\n", " # Add item1 columns with prefix\n", " for key, value in item1.items():\n", " row[f'item1_{key}'] = value\n", " # Add item2 columns with prefix\n", " for key, value in item2.items():\n", " row[f'item2_{key}'] = value\n", " pairs_data.append(row)\n", "\n", "df_pairs = pd.DataFrame(pairs_data)\n", "print(f\"Pairs dataframe shape: {df_pairs.shape}\")\n", "df_pairs.head()\n" ] }, { "cell_type": "code", "execution_count": null, "id": "5", "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": null, "id": "6", "metadata": {}, "outputs": [], "source": [ "# Remove rows where all three columns are NaN\n", "#df = df.dropna(subset=['style_weight', 'weirdness_constraint', 'audio_weight'], how='all')\n", "df.head()\n" ] }, { "cell_type": "code", "execution_count": null, "id": "7", "metadata": {}, "outputs": [], "source": [ "# Condense dataframe into list of pairs\n", "# Keep specified columns for each pair\n", "columns_to_keep = ['pos_preference', 'neg_preference', 'style_weight', 'weirdness_constraint', 'audio_weight', 'metadata']\n", "\n", "pairs = []\n", "for i in range(0, len(df), 2):\n", " if i + 1 < len(df): # Make sure we have a complete pair\n", " pair = (\n", " df.iloc[i][columns_to_keep].to_dict(),\n", " df.iloc[i + 1][columns_to_keep].to_dict()\n", " )\n", " pairs.append(pair)\n", "\n", "print(f\"Number of pairs: {len(pairs)}\")\n", "print(f\"First pair:\")\n", "print(pairs[0])\n" ] }, { "cell_type": "code", "execution_count": null, "id": "8", "metadata": {}, "outputs": [], "source": [ "df_pos = df_clean[df_clean['pos_preference'] == 1]\n", "df_neg = df_clean[df_clean['neg_preference'] == 1]\n", "print(len(df_pos))\n", "print(len(df_neg))\n" ] }, { "cell_type": "code", "execution_count": null, "id": "9", "metadata": {}, "outputs": [], "source": [ "pos_style = df_pos.dropna(subset=['style_weight'])['style_weight']\n", "neg_style = df_neg.dropna(subset=['style_weight'])['style_weight']\n", "pos_weirdness = df_pos.dropna(subset=['weirdness_constraint'])['weirdness_constraint']\n", "neg_weirdness = df_neg.dropna(subset=['weirdness_constraint'])['weirdness_constraint']\n", "pos_audio = df_pos.dropna(subset=['audio_weight'])['audio_weight']\n", "neg_audio = df_neg.dropna(subset=['audio_weight'])['audio_weight']\n" ] }, { "cell_type": "code", "execution_count": null, "id": "10", "metadata": {}, "outputs": [], "source": [ "pos_style.hist(bins=20)" ] }, { "cell_type": "code", "execution_count": null, "id": "11", "metadata": {}, "outputs": [], "source": [ "neg_style.hist(bins=20)" ] }, { "cell_type": "code", "execution_count": null, "id": "12", "metadata": {}, "outputs": [], "source": [ "pos_weirdness.hist(bins=20)" ] }, { "cell_type": "code", "execution_count": null, "id": "13", "metadata": {}, "outputs": [], "source": [ "neg_weirdness.hist(bins=20)" ] }, { "cell_type": "code", "execution_count": null, "id": "14", "metadata": {}, "outputs": [], "source": [ "pos_audio.hist(bins=20)" ] }, { "cell_type": "code", "execution_count": null, "id": "15", "metadata": {}, "outputs": [], "source": [ "neg_audio.hist(bins=20)" ] }, { "cell_type": "code", "execution_count": null, "id": "16", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "suno_clean", "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" } }, "nbformat": 4, "nbformat_minor": 5 }