Upload 5 files

genre_finder/test_some_codes/full_data_train.ipynb

@@ -0,0 +1,1748 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "id": "initial_id",
+   "metadata": {
+    "collapsed": true,
+    "ExecuteTime": {
+     "end_time": "2026-03-30T14:23:35.971085700Z",
+     "start_time": "2026-03-30T14:23:35.219016Z"
+    }
+   },
+   "source": [
+    "import pandas as pd\n",
+    "path1 = r\"assets/model_ready_dataset1.csv\"\n",
+    "path2 = r\"assets/model_ready_dataset_final.csv\"\n",
+    "path3 = r\"assets/model_ready_dataset_final1.csv\"\n",
+    "df1 = pd.read_csv(path1)\n",
+    "df2 = pd.read_csv(path2)\n",
+    "df3 = pd.read_csv(path3)\n",
+    "\n",
+    "df1.shape"
+   ],
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(34197, 28)"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 16
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T14:23:39.069360700Z",
+     "start_time": "2026-03-30T14:23:38.981358300Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df2.shape",
+   "id": "3a56566012bc4d20",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(41935, 28)"
+      ]
+     },
+     "execution_count": 17,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 17
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T14:23:41.625962100Z",
+     "start_time": "2026-03-30T14:23:41.550176200Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df3.shape",
+   "id": "cfaa8e83df91c960",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(49037, 28)"
+      ]
+     },
+     "execution_count": 18,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 18
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T14:23:42.129422500Z",
+     "start_time": "2026-03-30T14:23:42.060636Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df1[\"genre\"].value_counts()",
+   "id": "1bb37e9a1a4a63da",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "genre\n",
+       "Hip Hop Rap               21428\n",
+       "Folk Singer Songwriter     6917\n",
+       "Alternative Indie          5274\n",
+       "Gospel Christian            578\n",
+       "Name: count, dtype: int64"
+      ]
+     },
+     "execution_count": 19,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 19
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T14:23:44.650008100Z",
+     "start_time": "2026-03-30T14:23:44.586650900Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df2[\"genre\"].value_counts()",
+   "id": "2ef006202738872c",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "genre\n",
+       "Rock       29105\n",
+       "Country    12499\n",
+       "Punk         252\n",
+       "Metal         79\n",
+       "Name: count, dtype: int64"
+      ]
+     },
+     "execution_count": 20,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 20
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T14:23:45.215165Z",
+     "start_time": "2026-03-30T14:23:45.151818Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df3[\"genre\"].value_counts()",
+   "id": "e24fa51f0ef3da67",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "genre\n",
+       "EDM         18903\n",
+       "Pop         15967\n",
+       "RnB Soul    14167\n",
+       "Name: count, dtype: int64"
+      ]
+     },
+     "execution_count": 21,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 21
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T14:26:12.355726900Z",
+     "start_time": "2026-03-30T14:26:12.287068500Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "final_df = pd.concat([df1, df2, df3], ignore_index=True)\n",
+    "print(final_df.shape)"
+   ],
+   "id": "b13f17240912f389",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "(125169, 28)\n"
+     ]
+    }
+   ],
+   "execution_count": 22
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T14:27:02.483708400Z",
+     "start_time": "2026-03-30T14:26:58.546324800Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "final_df.to_csv(\"assets/all_genre.csv\",index=False)",
+   "id": "15fda3454dffbe41",
+   "outputs": [],
+   "execution_count": 23
+  },
+  {
+   "metadata": {},
+   "cell_type": "markdown",
+   "source": "### train the model",
+   "id": "20ac2df31c79f4da"
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T16:46:45.958963800Z",
+     "start_time": "2026-03-30T16:46:44.686791300Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import re\n",
+    "import ast\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from sklearn.preprocessing import StandardScaler, MultiLabelBinarizer\n",
+    "from sklearn.multioutput import MultiOutputClassifier\n",
+    "from xgboost import XGBClassifier\n",
+    "from sklearn.metrics import f1_score\n",
+    "\n",
+    "path = r\"assets/all_genre.csv\"\n",
+    "df = pd.read_csv(path)\n",
+    "df.sample(3)"
+   ],
+   "id": "13b7635596d96c8f",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "                             _id              genre  \\\n",
+       "87997   6996ea8d82f9a4b71aea132e                EDM   \n",
+       "104069  6954f094fe0a759b1bbc1c2f                Pop   \n",
+       "23363   6953758110ee1ae3f18eedf6  Alternative Indie   \n",
+       "\n",
+       "                                           genre_subgenre  \\\n",
+       "87997                                          EDM --- []   \n",
+       "104069          Pop --- ['Indie Electronic', 'Indie Pop']   \n",
+       "23363   Alternative Indie --- ['Folk Rock', 'Indie Pop...   \n",
+       "\n",
+       "        trimmed_audio_duration_sec  syllable_count  word_count  \\\n",
+       "87997                   143.144558           357.0       283.0   \n",
+       "104069                  183.665918           418.0       370.0   \n",
+       "23363                   193.548390           282.0       243.0   \n",
+       "\n",
+       "        spectral_contrast_mean (mix)  chroma_mean (mix)  \\\n",
+       "87997                      19.047967           0.483924   \n",
+       "104069                     19.945393           0.493085   \n",
+       "23363                      18.924223           0.461779   \n",
+       "\n",
+       "        melody_variability (vocals)  rhythm_onset_rate (mix)  ...  \\\n",
+       "87997                      0.428562                 4.645653  ...   \n",
+       "104069                     0.558270                 4.594444  ...   \n",
+       "23363                      0.462750                 2.633333  ...   \n",
+       "\n",
+       "        zero_crossing_rate (mix)  vocab_richness  \\\n",
+       "87997                   0.073138           0.258   \n",
+       "104069                  0.059187           0.224   \n",
+       "23363                   0.086377           0.387   \n",
+       "\n",
+       "        loudness_integrated_lufs custom (mix)  readability_score  \\\n",
+       "87997                              -12.788278               36.1   \n",
+       "104069                             -14.280496               45.4   \n",
+       "23363                              -12.252384               44.8   \n",
+       "\n",
+       "        energy_essentia (mix)  energy_librosa (mix)  rms_energy_mean (mix)  \\\n",
+       "87997                0.355957              0.550532               0.569507   \n",
+       "104069               0.276840              0.480446               0.480615   \n",
+       "23363                0.324193              0.525521               0.545800   \n",
+       "\n",
+       "        sentiment_score  melody_complexity (vocals)  avg_word_length  \n",
+       "87997             0.117                    2.333333             3.67  \n",
+       "104069            0.069                    2.666667             3.39  \n",
+       "23363            -0.064                    2.416667             3.62  \n",
+       "\n",
+       "[3 rows x 28 columns]"
+      ],
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>_id</th>\n",
+       "      <th>genre</th>\n",
+       "      <th>genre_subgenre</th>\n",
+       "      <th>trimmed_audio_duration_sec</th>\n",
+       "      <th>syllable_count</th>\n",
+       "      <th>word_count</th>\n",
+       "      <th>spectral_contrast_mean (mix)</th>\n",
+       "      <th>chroma_mean (mix)</th>\n",
+       "      <th>melody_variability (vocals)</th>\n",
+       "      <th>rhythm_onset_rate (mix)</th>\n",
+       "      <th>...</th>\n",
+       "      <th>zero_crossing_rate (mix)</th>\n",
+       "      <th>vocab_richness</th>\n",
+       "      <th>loudness_integrated_lufs custom (mix)</th>\n",
+       "      <th>readability_score</th>\n",
+       "      <th>energy_essentia (mix)</th>\n",
+       "      <th>energy_librosa (mix)</th>\n",
+       "      <th>rms_energy_mean (mix)</th>\n",
+       "      <th>sentiment_score</th>\n",
+       "      <th>melody_complexity (vocals)</th>\n",
+       "      <th>avg_word_length</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>87997</th>\n",
+       "      <td>6996ea8d82f9a4b71aea132e</td>\n",
+       "      <td>EDM</td>\n",
+       "      <td>EDM --- []</td>\n",
+       "      <td>143.144558</td>\n",
+       "      <td>357.0</td>\n",
+       "      <td>283.0</td>\n",
+       "      <td>19.047967</td>\n",
+       "      <td>0.483924</td>\n",
+       "      <td>0.428562</td>\n",
+       "      <td>4.645653</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.073138</td>\n",
+       "      <td>0.258</td>\n",
+       "      <td>-12.788278</td>\n",
+       "      <td>36.1</td>\n",
+       "      <td>0.355957</td>\n",
+       "      <td>0.550532</td>\n",
+       "      <td>0.569507</td>\n",
+       "      <td>0.117</td>\n",
+       "      <td>2.333333</td>\n",
+       "      <td>3.67</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>104069</th>\n",
+       "      <td>6954f094fe0a759b1bbc1c2f</td>\n",
+       "      <td>Pop</td>\n",
+       "      <td>Pop --- ['Indie Electronic', 'Indie Pop']</td>\n",
+       "      <td>183.665918</td>\n",
+       "      <td>418.0</td>\n",
+       "      <td>370.0</td>\n",
+       "      <td>19.945393</td>\n",
+       "      <td>0.493085</td>\n",
+       "      <td>0.558270</td>\n",
+       "      <td>4.594444</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.059187</td>\n",
+       "      <td>0.224</td>\n",
+       "      <td>-14.280496</td>\n",
+       "      <td>45.4</td>\n",
+       "      <td>0.276840</td>\n",
+       "      <td>0.480446</td>\n",
+       "      <td>0.480615</td>\n",
+       "      <td>0.069</td>\n",
+       "      <td>2.666667</td>\n",
+       "      <td>3.39</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>23363</th>\n",
+       "      <td>6953758110ee1ae3f18eedf6</td>\n",
+       "      <td>Alternative Indie</td>\n",
+       "      <td>Alternative Indie --- ['Folk Rock', 'Indie Pop...</td>\n",
+       "      <td>193.548390</td>\n",
+       "      <td>282.0</td>\n",
+       "      <td>243.0</td>\n",
+       "      <td>18.924223</td>\n",
+       "      <td>0.461779</td>\n",
+       "      <td>0.462750</td>\n",
+       "      <td>2.633333</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.086377</td>\n",
+       "      <td>0.387</td>\n",
+       "      <td>-12.252384</td>\n",
+       "      <td>44.8</td>\n",
+       "      <td>0.324193</td>\n",
+       "      <td>0.525521</td>\n",
+       "      <td>0.545800</td>\n",
+       "      <td>-0.064</td>\n",
+       "      <td>2.416667</td>\n",
+       "      <td>3.62</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>3 rows × 28 columns</p>\n",
+       "</div>"
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 1
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T16:46:45.992830500Z",
+     "start_time": "2026-03-30T16:46:45.967978200Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df.shape",
+   "id": "aff9220ee6839f04",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(125169, 28)"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 2
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T16:46:47.001568100Z",
+     "start_time": "2026-03-30T16:46:46.857630300Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "def parse_subgenres_fast(text):\n",
+    "    if pd.isna(text) or text == \"\":\n",
+    "        return []\n",
+    "    text = str(text)\n",
+    "    parts = text.split('---')\n",
+    "    main_genre = parts[0].strip()\n",
+    "    if len(parts) > 1:\n",
+    "        sub = parts[1].strip()\n",
+    "        sub = sub.strip('[]')\n",
+    "        if sub:\n",
+    "            subgenres = [s.strip().strip(\"'\").strip('\"') for s in sub.split(',')]\n",
+    "        else:\n",
+    "            subgenres = []\n",
+    "    else:\n",
+    "        subgenres = []\n",
+    "    return [main_genre] + subgenres\n",
+    "df['subgenres_list'] = df['genre_subgenre'].apply(parse_subgenres_fast)"
+   ],
+   "id": "5ccadaac034f64dc",
+   "outputs": [],
+   "execution_count": 3
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T16:46:47.886086100Z",
+     "start_time": "2026-03-30T16:46:47.825067400Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df['subgenres_list'][0]",
+   "id": "1e5e62299b79eeac",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['Hip Hop Rap',\n",
+       " 'Alternative Hip Hop',\n",
+       " 'Boom Bap',\n",
+       " 'Contemporary Hip Hop',\n",
+       " 'Trap']"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 4
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T16:46:48.597653700Z",
+     "start_time": "2026-03-30T16:46:48.481941800Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "mlb = MultiLabelBinarizer()\n",
+    "y = mlb.fit_transform(df['subgenres_list'])\n",
+    "y"
+   ],
+   "id": "7683931669716dbb",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([[0, 0, 0, ..., 0, 0, 0],\n",
+       "       [0, 0, 0, ..., 0, 0, 0],\n",
+       "       [0, 0, 0, ..., 0, 0, 0],\n",
+       "       ...,\n",
+       "       [0, 0, 0, ..., 0, 0, 0],\n",
+       "       [0, 0, 0, ..., 0, 0, 0],\n",
+       "       [0, 0, 0, ..., 0, 0, 0]], shape=(125169, 163))"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 5
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T16:57:04.551225300Z",
+     "start_time": "2026-03-30T16:46:49.634662300Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "drop_cols = ['_id', 'genre', 'genre_subgenre', 'subgenres_list', 'labels',\n",
+    "             'trimmed_audio_duration_sec', 'syllable_count', 'word_count',\n",
+    "             'sentiment_score', 'avg_word_length']\n",
+    "\n",
+    "feature_cols = [c for c in df.columns if c not in drop_cols]\n",
+    "X = df[feature_cols].copy()\n",
+    "X = X.select_dtypes(include=[np.number]).fillna(0)\n",
+    "\n",
+    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
+    "\n",
+    "scaler = StandardScaler()\n",
+    "X_train_scaled = scaler.fit_transform(X_train)\n",
+    "X_test_scaled = scaler.transform(X_test)\n",
+    "\n",
+    "model = MultiOutputClassifier(\n",
+    "    XGBClassifier(\n",
+    "        n_estimators=1700,\n",
+    "        learning_rate=0.03,\n",
+    "        max_depth=8,\n",
+    "        subsample=0.85,\n",
+    "        colsample_bytree=0.8,\n",
+    "        random_state=42,\n",
+    "        eval_metric='logloss'\n",
+    "    ),\n",
+    "    n_jobs=-1\n",
+    ")\n",
+    "\n",
+    "model.fit(X_train_scaled, y_train)\n",
+    "\n",
+    "def get_accuracy(X_scaled, y_true, name=\"\"):\n",
+    "    y_pred = model.predict(X_scaled)\n",
+    "    subset_acc = np.mean(np.all(y_pred == y_true, axis=1))\n",
+    "    micro_f1 = f1_score(y_true, y_pred, average='micro')\n",
+    "    print(f\"{name} Subset Accuracy: {subset_acc:.4f}\")\n",
+    "    print(f\"{name} Micro F1 Score: {micro_f1:.4f}\")\n",
+    "\n",
+    "print(\"=== MODEL ACCURACY ===\")\n",
+    "get_accuracy(X_train_scaled, y_train, \"Train\")\n",
+    "get_accuracy(X_test_scaled, y_test, \"Test\")\n",
+    "\n",
+    "def manual_predict(row):\n",
+    "    input_df = pd.DataFrame([row])\n",
+    "    input_numeric = input_df.select_dtypes(include=[np.number]).fillna(0)\n",
+    "\n",
+    "    for col in X.columns:\n",
+    "        if col not in input_numeric.columns:\n",
+    "            input_numeric[col] = 0\n",
+    "    input_numeric = input_numeric[X.columns]\n",
+    "    input_scaled = scaler.transform(input_numeric)\n",
+    "    probas = model.predict_proba(input_scaled)\n",
+    "    threshold = 0.20\n",
+    "    pred_binary = np.zeros((1, y.shape[1]), dtype=int)\n",
+    "    for i in range(len(probas)):\n",
+    "        if probas[i][0][1] > threshold:\n",
+    "            pred_binary[0, i] = 1\n",
+    "    subgenres = mlb.inverse_transform(pred_binary)[0]\n",
+    "    if len(subgenres) == 0:\n",
+    "        return f\"None\"\n",
+    "    return f\"{', '.join(sorted(subgenres))}\""
+   ],
+   "id": "2e87bee87e8284f1",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "=== MODEL ACCURACY ===\n",
+      "Train Subset Accuracy: 0.8709\n",
+      "Train Micro F1 Score: 0.9673\n",
+      "Test Subset Accuracy: 0.1465\n",
+      "Test Micro F1 Score: 0.3423\n"
+     ]
+    }
+   ],
+   "execution_count": 6
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T16:57:34.255351600Z",
+     "start_time": "2026-03-30T16:57:27.678874900Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "import joblib\n",
+    "bundle = {\n",
+    "    \"model\": model,\n",
+    "    \"scaler\": scaler,\n",
+    "    \"mlb\": mlb,\n",
+    "    \"feature_cols\": X.columns.tolist()\n",
+    "}\n",
+    "joblib.dump(bundle, \"train_model/all_genre.pkl\")\n",
+    "print(\"Model saved successfully!\")"
+   ],
+   "id": "53c2d801062301d7",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model saved successfully!\n"
+     ]
+    }
+   ],
+   "execution_count": 7
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T17:00:16.116265100Z",
+     "start_time": "2026-03-30T17:00:16.024652100Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df[[\"subgenres_list\",\"genre\"]].iloc[508,]",
+   "id": "29ae3adc40dc0a2d",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "subgenres_list    [Hip Hop Rap, Contemporary Hip Hop, Trap]\n",
+       "genre                                           Hip Hop Rap\n",
+       "Name: 508, dtype: object"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 8
+  },
+  {
+   "metadata": {},
+   "cell_type": "markdown",
+   "source": "## 1) Second apporoch",
+   "id": "5b64bdd097a9d631"
+  },
+  {
+   "metadata": {},
+   "cell_type": "markdown",
+   "source": "## 1) SECOND CONCEPTS",
+   "id": "3f8160d92c2cd86f"
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T05:30:08.866729200Z",
+     "start_time": "2026-03-31T05:30:08.800975400Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "#! pip install tensorflow",
+   "id": "60c350c541592fb8",
+   "outputs": [],
+   "execution_count": 3
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T09:27:58.269252700Z",
+     "start_time": "2026-03-31T09:27:56.493881900Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import re\n",
+    "import ast\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from sklearn.preprocessing import StandardScaler, MultiLabelBinarizer\n",
+    "from sklearn.multioutput import MultiOutputClassifier\n",
+    "from xgboost import XGBClassifier\n",
+    "from sklearn.metrics import f1_score"
+   ],
+   "id": "717885273a039bc0",
+   "outputs": [],
+   "execution_count": 1
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T09:28:02.080288200Z",
+     "start_time": "2026-03-31T09:27:59.859812300Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "df = pd.read_csv(\"final_clean_data.csv\")\n",
+    "df['genre_subgenre'] = df['genre_subgenre'].apply(lambda x: ast.literal_eval(x) if isinstance(x, str) else x)"
+   ],
+   "id": "6460a59542844d74",
+   "outputs": [],
+   "execution_count": 2
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T09:28:44.762378100Z",
+     "start_time": "2026-03-31T09:28:44.664699400Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df.sample(3)",
+   "id": "a5951d9f9252687e",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "                                           genre_subgenre  \\\n",
+       "55138   [Indie Folk, Dream Pop, Indie Pop, Folk Pop, I...   \n",
+       "34570                    [Classic Alternative Rock, Rock]   \n",
+       "115446            [Dream Pop, Indie Pop, Lo Fi, RnB Soul]   \n",
+       "\n",
+       "        spectral_contrast_mean (mix)  chroma_mean (mix)  \\\n",
+       "55138                      18.953288           0.438801   \n",
+       "34570                      17.648178           0.593524   \n",
+       "115446                     19.939022           0.460538   \n",
+       "\n",
+       "        melody_variability (vocals)  rhythm_onset_rate (mix)  \\\n",
+       "55138                      0.276532                 5.561111   \n",
+       "34570                      0.828343                 1.917868   \n",
+       "115446                     0.366296                 2.050000   \n",
+       "\n",
+       "        spectral_centroid_mean custom (mix)  repetition_score custom (mix)  \\\n",
+       "55138                              0.136025                       0.006977   \n",
+       "34570                              0.192934                       0.009650   \n",
+       "115446                             0.137038                       0.007738   \n",
+       "\n",
+       "        pitch_std (mix)  tempo_bpm_corrected (mix)  tempo_bpm_original (mix)  \\\n",
+       "55138         48.063896                 162.070480                162.070480   \n",
+       "34570        132.527191                 161.965561                161.965561   \n",
+       "115446       108.319832                 139.820786                139.820786   \n",
+       "\n",
+       "        ...  mfcc_mean_2 (mix)  loudness_range_lu custom (mix)  \\\n",
+       "55138   ...         -27.844322                        6.587286   \n",
+       "34570   ...         -38.153713                        0.697901   \n",
+       "115446  ...          21.716513                        5.794365   \n",
+       "\n",
+       "        zero_crossing_rate (mix)  vocab_richness  \\\n",
+       "55138                   0.058096           0.478   \n",
+       "34570                   0.103853           0.480   \n",
+       "115446                  0.046103           0.622   \n",
+       "\n",
+       "        loudness_integrated_lufs custom (mix)  readability_score  \\\n",
+       "55138                              -15.621501              107.0   \n",
+       "34570                              -10.773592               37.6   \n",
+       "115446                             -17.407324               54.4   \n",
+       "\n",
+       "        energy_essentia (mix)  energy_librosa (mix)  rms_energy_mean (mix)  \\\n",
+       "55138                0.201458              0.412518               0.417358   \n",
+       "34570                0.511921              0.698029               0.706971   \n",
+       "115446               0.147544              0.324240               0.332860   \n",
+       "\n",
+       "        melody_complexity (vocals)  \n",
+       "55138                     1.500000  \n",
+       "34570                     2.916667  \n",
+       "115446                    1.916667  \n",
+       "\n",
+       "[3 rows x 21 columns]"
+      ],
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>genre_subgenre</th>\n",
+       "      <th>spectral_contrast_mean (mix)</th>\n",
+       "      <th>chroma_mean (mix)</th>\n",
+       "      <th>melody_variability (vocals)</th>\n",
+       "      <th>rhythm_onset_rate (mix)</th>\n",
+       "      <th>spectral_centroid_mean custom (mix)</th>\n",
+       "      <th>repetition_score custom (mix)</th>\n",
+       "      <th>pitch_std (mix)</th>\n",
+       "      <th>tempo_bpm_corrected (mix)</th>\n",
+       "      <th>tempo_bpm_original (mix)</th>\n",
+       "      <th>...</th>\n",
+       "      <th>mfcc_mean_2 (mix)</th>\n",
+       "      <th>loudness_range_lu custom (mix)</th>\n",
+       "      <th>zero_crossing_rate (mix)</th>\n",
+       "      <th>vocab_richness</th>\n",
+       "      <th>loudness_integrated_lufs custom (mix)</th>\n",
+       "      <th>readability_score</th>\n",
+       "      <th>energy_essentia (mix)</th>\n",
+       "      <th>energy_librosa (mix)</th>\n",
+       "      <th>rms_energy_mean (mix)</th>\n",
+       "      <th>melody_complexity (vocals)</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>55138</th>\n",
+       "      <td>[Indie Folk, Dream Pop, Indie Pop, Folk Pop, I...</td>\n",
+       "      <td>18.953288</td>\n",
+       "      <td>0.438801</td>\n",
+       "      <td>0.276532</td>\n",
+       "      <td>5.561111</td>\n",
+       "      <td>0.136025</td>\n",
+       "      <td>0.006977</td>\n",
+       "      <td>48.063896</td>\n",
+       "      <td>162.070480</td>\n",
+       "      <td>162.070480</td>\n",
+       "      <td>...</td>\n",
+       "      <td>-27.844322</td>\n",
+       "      <td>6.587286</td>\n",
+       "      <td>0.058096</td>\n",
+       "      <td>0.478</td>\n",
+       "      <td>-15.621501</td>\n",
+       "      <td>107.0</td>\n",
+       "      <td>0.201458</td>\n",
+       "      <td>0.412518</td>\n",
+       "      <td>0.417358</td>\n",
+       "      <td>1.500000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>34570</th>\n",
+       "      <td>[Classic Alternative Rock, Rock]</td>\n",
+       "      <td>17.648178</td>\n",
+       "      <td>0.593524</td>\n",
+       "      <td>0.828343</td>\n",
+       "      <td>1.917868</td>\n",
+       "      <td>0.192934</td>\n",
+       "      <td>0.009650</td>\n",
+       "      <td>132.527191</td>\n",
+       "      <td>161.965561</td>\n",
+       "      <td>161.965561</td>\n",
+       "      <td>...</td>\n",
+       "      <td>-38.153713</td>\n",
+       "      <td>0.697901</td>\n",
+       "      <td>0.103853</td>\n",
+       "      <td>0.480</td>\n",
+       "      <td>-10.773592</td>\n",
+       "      <td>37.6</td>\n",
+       "      <td>0.511921</td>\n",
+       "      <td>0.698029</td>\n",
+       "      <td>0.706971</td>\n",
+       "      <td>2.916667</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>115446</th>\n",
+       "      <td>[Dream Pop, Indie Pop, Lo Fi, RnB Soul]</td>\n",
+       "      <td>19.939022</td>\n",
+       "      <td>0.460538</td>\n",
+       "      <td>0.366296</td>\n",
+       "      <td>2.050000</td>\n",
+       "      <td>0.137038</td>\n",
+       "      <td>0.007738</td>\n",
+       "      <td>108.319832</td>\n",
+       "      <td>139.820786</td>\n",
+       "      <td>139.820786</td>\n",
+       "      <td>...</td>\n",
+       "      <td>21.716513</td>\n",
+       "      <td>5.794365</td>\n",
+       "      <td>0.046103</td>\n",
+       "      <td>0.622</td>\n",
+       "      <td>-17.407324</td>\n",
+       "      <td>54.4</td>\n",
+       "      <td>0.147544</td>\n",
+       "      <td>0.324240</td>\n",
+       "      <td>0.332860</td>\n",
+       "      <td>1.916667</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>3 rows × 21 columns</p>\n",
+       "</div>"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 3
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T09:28:47.156213Z",
+     "start_time": "2026-03-31T09:28:47.071559100Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "X= df.drop(columns=[\"genre_subgenre\"])\n",
+    "X"
+   ],
+   "id": "9d4d262a6e6158b",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "        spectral_contrast_mean (mix)  chroma_mean (mix)  \\\n",
+       "0                          20.440557           0.464204   \n",
+       "1                          19.657228           0.424237   \n",
+       "2                          20.115561           0.454093   \n",
+       "3                          19.659283           0.443467   \n",
+       "4                          20.855880           0.461875   \n",
+       "...                              ...                ...   \n",
+       "125164                     20.244104           0.503082   \n",
+       "125165                     19.474557           0.567127   \n",
+       "125166                     19.522079           0.492525   \n",
+       "125167                     19.721407           0.529660   \n",
+       "125168                     20.478757           0.498923   \n",
+       "\n",
+       "        melody_variability (vocals)  rhythm_onset_rate (mix)  \\\n",
+       "0                          0.434021                 3.250000   \n",
+       "1                          0.351001                 2.425576   \n",
+       "2                          0.327000                 2.738889   \n",
+       "3                          0.360642                 3.500000   \n",
+       "4                          0.767493                 4.983333   \n",
+       "...                             ...                      ...   \n",
+       "125164                     0.872020                 3.916667   \n",
+       "125165                     0.567603                 4.211111   \n",
+       "125166                     0.795226                 3.722222   \n",
+       "125167                     0.640377                 3.994444   \n",
+       "125168                     0.570758                 3.436631   \n",
+       "\n",
+       "        spectral_centroid_mean custom (mix)  repetition_score custom (mix)  \\\n",
+       "0                                  0.132002                       0.006988   \n",
+       "1                                  0.139757                       0.007367   \n",
+       "2                                  0.111282                       0.006671   \n",
+       "3                                  0.159247                       0.007202   \n",
+       "4                                  0.155302                       0.007290   \n",
+       "...                                     ...                            ...   \n",
+       "125164                             0.162460                       0.008042   \n",
+       "125165                             0.202201                       0.007874   \n",
+       "125166                             0.180673                       0.006816   \n",
+       "125167                             0.191156                       0.006488   \n",
+       "125168                             0.166860                       0.010107   \n",
+       "\n",
+       "        pitch_std (mix)  tempo_bpm_corrected (mix)  tempo_bpm_original (mix)  \\\n",
+       "0             69.616264                 152.006775                152.006775   \n",
+       "1            156.358124                 117.943665                117.943665   \n",
+       "2             69.311798                 124.344086                124.344086   \n",
+       "3             65.514191                 109.909401                109.909401   \n",
+       "4            127.873909                 151.955536                151.955536   \n",
+       "...                 ...                        ...                       ...   \n",
+       "125164        83.902977                  81.075050                 81.075050   \n",
+       "125165        82.328941                 101.035919                101.035919   \n",
+       "125166       179.515854                 111.822731                111.822731   \n",
+       "125167        64.050621                  93.118805                 93.118805   \n",
+       "125168       128.814209                 123.944077                123.944077   \n",
+       "\n",
+       "        mfcc_mean_1 (mix)  mfcc_mean_2 (mix)  loudness_range_lu custom (mix)  \\\n",
+       "0              153.058243          -7.368185                        1.143722   \n",
+       "1              148.055832         -49.435444                        4.708925   \n",
+       "2              166.170502          -5.852028                        1.948263   \n",
+       "3              134.211624         -27.524824                        1.244652   \n",
+       "4              129.783264         -11.940094                        3.452597   \n",
+       "...                   ...                ...                             ...   \n",
+       "125164         117.612518           9.144649                        2.602306   \n",
+       "125165          94.410416         -11.153146                        1.345105   \n",
+       "125166         115.851357         -48.314766                        3.907774   \n",
+       "125167         111.312195         -16.979046                        1.437813   \n",
+       "125168         124.521019         -20.625500                        6.944749   \n",
+       "\n",
+       "        zero_crossing_rate (mix)  vocab_richness  \\\n",
+       "0                       0.045159           0.305   \n",
+       "1                       0.070352           0.352   \n",
+       "2                       0.036463           0.400   \n",
+       "3                       0.064695           0.365   \n",
+       "4                       0.056507           0.281   \n",
+       "...                          ...             ...   \n",
+       "125164                  0.060365           0.389   \n",
+       "125165                  0.077860           0.416   \n",
+       "125166                  0.095063           0.436   \n",
+       "125167                  0.087692           0.411   \n",
+       "125168                  0.075311           0.541   \n",
+       "\n",
+       "        loudness_integrated_lufs custom (mix)  readability_score  \\\n",
+       "0                                  -12.104649               20.9   \n",
+       "1                                  -11.814197               89.4   \n",
+       "2                                  -12.000179              107.0   \n",
+       "3                                  -12.435019              105.8   \n",
+       "4                                  -16.217091              212.3   \n",
+       "...                                       ...                ...   \n",
+       "125164                             -15.233034                2.0   \n",
+       "125165                             -13.726265               94.5   \n",
+       "125166                             -14.254233                1.8   \n",
+       "125167                             -15.104306                1.5   \n",
+       "125168                             -15.867487               21.2   \n",
+       "\n",
+       "        energy_essentia (mix)  energy_librosa (mix)  rms_energy_mean (mix)  \\\n",
+       "0                    0.349783              0.576250               0.578454   \n",
+       "1                    0.429792              0.619931               0.632507   \n",
+       "2                    0.325114              0.520390               0.579053   \n",
+       "3                    0.407647              0.617191               0.617197   \n",
+       "4                    0.255311              0.446953               0.454712   \n",
+       "...                       ...                   ...                    ...   \n",
+       "125164               0.202741              0.378823               0.384110   \n",
+       "125165               0.275752              0.470764               0.481895   \n",
+       "125166               0.265776              0.457695               0.472769   \n",
+       "125167               0.185114              0.360564               0.403058   \n",
+       "125168               0.280897              0.475058               0.475100   \n",
+       "\n",
+       "        melody_complexity (vocals)  \n",
+       "0                         2.250000  \n",
+       "1                         1.916667  \n",
+       "2                         1.833333  \n",
+       "3                         1.916667  \n",
+       "4                         3.500000  \n",
+       "...                            ...  \n",
+       "125164                    3.666667  \n",
+       "125165                    2.833333  \n",
+       "125166                    3.583333  \n",
+       "125167                    3.416667  \n",
+       "125168                    2.666667  \n",
+       "\n",
+       "[125169 rows x 20 columns]"
+      ],
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>spectral_contrast_mean (mix)</th>\n",
+       "      <th>chroma_mean (mix)</th>\n",
+       "      <th>melody_variability (vocals)</th>\n",
+       "      <th>rhythm_onset_rate (mix)</th>\n",
+       "      <th>spectral_centroid_mean custom (mix)</th>\n",
+       "      <th>repetition_score custom (mix)</th>\n",
+       "      <th>pitch_std (mix)</th>\n",
+       "      <th>tempo_bpm_corrected (mix)</th>\n",
+       "      <th>tempo_bpm_original (mix)</th>\n",
+       "      <th>mfcc_mean_1 (mix)</th>\n",
+       "      <th>mfcc_mean_2 (mix)</th>\n",
+       "      <th>loudness_range_lu custom (mix)</th>\n",
+       "      <th>zero_crossing_rate (mix)</th>\n",
+       "      <th>vocab_richness</th>\n",
+       "      <th>loudness_integrated_lufs custom (mix)</th>\n",
+       "      <th>readability_score</th>\n",
+       "      <th>energy_essentia (mix)</th>\n",
+       "      <th>energy_librosa (mix)</th>\n",
+       "      <th>rms_energy_mean (mix)</th>\n",
+       "      <th>melody_complexity (vocals)</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>20.440557</td>\n",
+       "      <td>0.464204</td>\n",
+       "      <td>0.434021</td>\n",
+       "      <td>3.250000</td>\n",
+       "      <td>0.132002</td>\n",
+       "      <td>0.006988</td>\n",
+       "      <td>69.616264</td>\n",
+       "      <td>152.006775</td>\n",
+       "      <td>152.006775</td>\n",
+       "      <td>153.058243</td>\n",
+       "      <td>-7.368185</td>\n",
+       "      <td>1.143722</td>\n",
+       "      <td>0.045159</td>\n",
+       "      <td>0.305</td>\n",
+       "      <td>-12.104649</td>\n",
+       "      <td>20.9</td>\n",
+       "      <td>0.349783</td>\n",
+       "      <td>0.576250</td>\n",
+       "      <td>0.578454</td>\n",
+       "      <td>2.250000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>19.657228</td>\n",
+       "      <td>0.424237</td>\n",
+       "      <td>0.351001</td>\n",
+       "      <td>2.425576</td>\n",
+       "      <td>0.139757</td>\n",
+       "      <td>0.007367</td>\n",
+       "      <td>156.358124</td>\n",
+       "      <td>117.943665</td>\n",
+       "      <td>117.943665</td>\n",
+       "      <td>148.055832</td>\n",
+       "      <td>-49.435444</td>\n",
+       "      <td>4.708925</td>\n",
+       "      <td>0.070352</td>\n",
+       "      <td>0.352</td>\n",
+       "      <td>-11.814197</td>\n",
+       "      <td>89.4</td>\n",
+       "      <td>0.429792</td>\n",
+       "      <td>0.619931</td>\n",
+       "      <td>0.632507</td>\n",
+       "      <td>1.916667</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>20.115561</td>\n",
+       "      <td>0.454093</td>\n",
+       "      <td>0.327000</td>\n",
+       "      <td>2.738889</td>\n",
+       "      <td>0.111282</td>\n",
+       "      <td>0.006671</td>\n",
+       "      <td>69.311798</td>\n",
+       "      <td>124.344086</td>\n",
+       "      <td>124.344086</td>\n",
+       "      <td>166.170502</td>\n",
+       "      <td>-5.852028</td>\n",
+       "      <td>1.948263</td>\n",
+       "      <td>0.036463</td>\n",
+       "      <td>0.400</td>\n",
+       "      <td>-12.000179</td>\n",
+       "      <td>107.0</td>\n",
+       "      <td>0.325114</td>\n",
+       "      <td>0.520390</td>\n",
+       "      <td>0.579053</td>\n",
+       "      <td>1.833333</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>19.659283</td>\n",
+       "      <td>0.443467</td>\n",
+       "      <td>0.360642</td>\n",
+       "      <td>3.500000</td>\n",
+       "      <td>0.159247</td>\n",
+       "      <td>0.007202</td>\n",
+       "      <td>65.514191</td>\n",
+       "      <td>109.909401</td>\n",
+       "      <td>109.909401</td>\n",
+       "      <td>134.211624</td>\n",
+       "      <td>-27.524824</td>\n",
+       "      <td>1.244652</td>\n",
+       "      <td>0.064695</td>\n",
+       "      <td>0.365</td>\n",
+       "      <td>-12.435019</td>\n",
+       "      <td>105.8</td>\n",
+       "      <td>0.407647</td>\n",
+       "      <td>0.617191</td>\n",
+       "      <td>0.617197</td>\n",
+       "      <td>1.916667</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>20.855880</td>\n",
+       "      <td>0.461875</td>\n",
+       "      <td>0.767493</td>\n",
+       "      <td>4.983333</td>\n",
+       "      <td>0.155302</td>\n",
+       "      <td>0.007290</td>\n",
+       "      <td>127.873909</td>\n",
+       "      <td>151.955536</td>\n",
+       "      <td>151.955536</td>\n",
+       "      <td>129.783264</td>\n",
+       "      <td>-11.940094</td>\n",
+       "      <td>3.452597</td>\n",
+       "      <td>0.056507</td>\n",
+       "      <td>0.281</td>\n",
+       "      <td>-16.217091</td>\n",
+       "      <td>212.3</td>\n",
+       "      <td>0.255311</td>\n",
+       "      <td>0.446953</td>\n",
+       "      <td>0.454712</td>\n",
+       "      <td>3.500000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>125164</th>\n",
+       "      <td>20.244104</td>\n",
+       "      <td>0.503082</td>\n",
+       "      <td>0.872020</td>\n",
+       "      <td>3.916667</td>\n",
+       "      <td>0.162460</td>\n",
+       "      <td>0.008042</td>\n",
+       "      <td>83.902977</td>\n",
+       "      <td>81.075050</td>\n",
+       "      <td>81.075050</td>\n",
+       "      <td>117.612518</td>\n",
+       "      <td>9.144649</td>\n",
+       "      <td>2.602306</td>\n",
+       "      <td>0.060365</td>\n",
+       "      <td>0.389</td>\n",
+       "      <td>-15.233034</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>0.202741</td>\n",
+       "      <td>0.378823</td>\n",
+       "      <td>0.384110</td>\n",
+       "      <td>3.666667</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>125165</th>\n",
+       "      <td>19.474557</td>\n",
+       "      <td>0.567127</td>\n",
+       "      <td>0.567603</td>\n",
+       "      <td>4.211111</td>\n",
+       "      <td>0.202201</td>\n",
+       "      <td>0.007874</td>\n",
+       "      <td>82.328941</td>\n",
+       "      <td>101.035919</td>\n",
+       "      <td>101.035919</td>\n",
+       "      <td>94.410416</td>\n",
+       "      <td>-11.153146</td>\n",
+       "      <td>1.345105</td>\n",
+       "      <td>0.077860</td>\n",
+       "      <td>0.416</td>\n",
+       "      <td>-13.726265</td>\n",
+       "      <td>94.5</td>\n",
+       "      <td>0.275752</td>\n",
+       "      <td>0.470764</td>\n",
+       "      <td>0.481895</td>\n",
+       "      <td>2.833333</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>125166</th>\n",
+       "      <td>19.522079</td>\n",
+       "      <td>0.492525</td>\n",
+       "      <td>0.795226</td>\n",
+       "      <td>3.722222</td>\n",
+       "      <td>0.180673</td>\n",
+       "      <td>0.006816</td>\n",
+       "      <td>179.515854</td>\n",
+       "      <td>111.822731</td>\n",
+       "      <td>111.822731</td>\n",
+       "      <td>115.851357</td>\n",
+       "      <td>-48.314766</td>\n",
+       "      <td>3.907774</td>\n",
+       "      <td>0.095063</td>\n",
+       "      <td>0.436</td>\n",
+       "      <td>-14.254233</td>\n",
+       "      <td>1.8</td>\n",
+       "      <td>0.265776</td>\n",
+       "      <td>0.457695</td>\n",
+       "      <td>0.472769</td>\n",
+       "      <td>3.583333</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>125167</th>\n",
+       "      <td>19.721407</td>\n",
+       "      <td>0.529660</td>\n",
+       "      <td>0.640377</td>\n",
+       "      <td>3.994444</td>\n",
+       "      <td>0.191156</td>\n",
+       "      <td>0.006488</td>\n",
+       "      <td>64.050621</td>\n",
+       "      <td>93.118805</td>\n",
+       "      <td>93.118805</td>\n",
+       "      <td>111.312195</td>\n",
+       "      <td>-16.979046</td>\n",
+       "      <td>1.437813</td>\n",
+       "      <td>0.087692</td>\n",
+       "      <td>0.411</td>\n",
+       "      <td>-15.104306</td>\n",
+       "      <td>1.5</td>\n",
+       "      <td>0.185114</td>\n",
+       "      <td>0.360564</td>\n",
+       "      <td>0.403058</td>\n",
+       "      <td>3.416667</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>125168</th>\n",
+       "      <td>20.478757</td>\n",
+       "      <td>0.498923</td>\n",
+       "      <td>0.570758</td>\n",
+       "      <td>3.436631</td>\n",
+       "      <td>0.166860</td>\n",
+       "      <td>0.010107</td>\n",
+       "      <td>128.814209</td>\n",
+       "      <td>123.944077</td>\n",
+       "      <td>123.944077</td>\n",
+       "      <td>124.521019</td>\n",
+       "      <td>-20.625500</td>\n",
+       "      <td>6.944749</td>\n",
+       "      <td>0.075311</td>\n",
+       "      <td>0.541</td>\n",
+       "      <td>-15.867487</td>\n",
+       "      <td>21.2</td>\n",
+       "      <td>0.280897</td>\n",
+       "      <td>0.475058</td>\n",
+       "      <td>0.475100</td>\n",
+       "      <td>2.666667</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>125169 rows × 20 columns</p>\n",
+       "</div>"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 4
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T09:28:49.838936400Z",
+     "start_time": "2026-03-31T09:28:49.621738700Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "mlb = MultiLabelBinarizer()\n",
+    "y = mlb.fit_transform(df['genre_subgenre'])\n",
+    "y"
+   ],
+   "id": "a1b09be302261710",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([[0, 0, 0, ..., 0, 0, 0],\n",
+       "       [0, 0, 0, ..., 0, 0, 0],\n",
+       "       [0, 0, 0, ..., 0, 0, 0],\n",
+       "       ...,\n",
+       "       [0, 0, 0, ..., 0, 0, 0],\n",
+       "       [0, 0, 0, ..., 0, 0, 0],\n",
+       "       [0, 0, 0, ..., 0, 0, 0]], shape=(125169, 163))"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 5
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T09:28:54.840304600Z",
+     "start_time": "2026-03-31T09:28:54.776756500Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "mlb.classes_",
+   "id": "208d2a63091460eb",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array(['Acoustic', 'Afrobeats', 'Alt', 'Alt Country', 'Alt Electronic',\n",
+       "       'Alt Folk', 'Alt Folk Christian', 'Alt Pop', 'Alt Pop Punk',\n",
+       "       'Alt Punk Rap', 'Alt Soul', 'Alternative Hip Hop',\n",
+       "       'Alternative Indie', 'Alternative RnB', 'Alternative Rock',\n",
+       "       'Ambient', 'Ambient Pop', 'Americana', 'Americana Alt', 'Bad Pop',\n",
+       "       'Baroque Pop', 'Black Metal', 'Bluegrass', 'Boom Bap',\n",
+       "       'British Punk', 'Bro Country', 'CCM', 'Chicano Rap', 'Chillwave',\n",
+       "       'Christian Rap', 'Christian Rock', 'Classic Alternative Rock',\n",
+       "       'Classic Country', 'Classic Rock', 'Conscious Rap',\n",
+       "       'Contemporary Christian', 'Contemporary Country',\n",
+       "       'Contemporary Hip Hop', 'Contemporary RnB', 'Contemporary Singer',\n",
+       "       'Contemporary Singer Songwriter', 'Country', 'Country Folk',\n",
+       "       'Country Gospel', 'Country Pop', 'Country Rock',\n",
+       "       'Country Trap / Hybrid', 'Dance', 'Dance Pop', 'Death Metal',\n",
+       "       'Deep House', 'Doom Metal', 'Dream Pop', 'Drill', 'Drum n Bass',\n",
+       "       'Dubstep', 'EDM', 'East Coast Hip Hop', 'East Coast Rap',\n",
+       "       'Electro', 'Electro Pop', 'Electronic Dance', 'Electropop', 'Emo',\n",
+       "       'Emo Rap', 'Ethereal Wave', 'Folk', 'Folk Alt', 'Folk Pop',\n",
+       "       'Folk Rock', 'Folk Singer Songwriter', 'Funk', 'Funk Alt',\n",
+       "       'Funk Grunge', 'Future Bass', 'G Funk', 'Gangsta', 'Gangsta Rap',\n",
+       "       'Garage Rock', 'Glam Metal', 'Gospel', 'Gospel Christian',\n",
+       "       'Gothic Metal', 'Grunge', 'Hard Rock', 'Hardcore Hip Hop',\n",
+       "       'Hardcore Punk', 'Hardcore Rap', 'Heavy Metal', 'Hip Hop Punk',\n",
+       "       'Hip Hop Rap', 'Honky Tonk', 'House', 'Indie', 'Indie Alt',\n",
+       "       'Indie Electronic', 'Indie Folk', 'Indie Funk', 'Indie Pop',\n",
+       "       'Indie Punk', 'Indie Rock', 'J Pop', 'Jazz Pop', 'Jazz Rap',\n",
+       "       'Jazz-influenced Hip Hop', 'K Pop', 'Latin Pop', 'Lo Fi',\n",
+       "       'Mafioso Rap', 'Mainstream Pop', 'Metal', 'Metal Pop', 'Metalcore',\n",
+       "       'Modern Alternative Rock', 'Modern Neo-Traditional', 'Motown',\n",
+       "       'Neo Soul', 'Neo-Traditional', 'New Jack Swing', 'Noise Pop',\n",
+       "       'Nu Metal', 'Old School Hip Hop', 'Outlaw Country',\n",
+       "       'Political Hip Hop', 'Pop', 'Pop Alt', 'Pop Country', 'Pop Indie',\n",
+       "       'Pop Metal', 'Pop Punk', 'Pop Rap', 'Pop Rock', 'Post Punk',\n",
+       "       'Post-Rock', 'Praise and Worship', 'Progressive Rock',\n",
+       "       'Psychedelic Rock', 'Punk', 'Punk Metal', 'Punk Rock', 'Red Dirt',\n",
+       "       'RnB Soul', 'Rock', 'Shoegaze', 'Songwriter', 'Soul',\n",
+       "       'Southern Gospel', 'Southern Hip Hop', 'Southern Rap',\n",
+       "       'Southern Rock', 'Synth Pop', 'Techno', 'Teen Pop',\n",
+       "       'Texas Country', 'Thrash Metal', 'Traditional Folk', 'Trance',\n",
+       "       'Trap', 'UK Garage', 'Underground Hip Hop', 'Underground Rap',\n",
+       "       'Urban Gospel', 'West Coast Hip Hop'], dtype=object)"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 6
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T09:42:33.539216600Z",
+     "start_time": "2026-03-31T09:29:38.234766500Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.20, random_state=42)\n",
+    "\n",
+    "scaler = StandardScaler()\n",
+    "X_train_scaled = scaler.fit_transform(X_train)\n",
+    "X_test_scaled = scaler.transform(X_test)\n",
+    "\n",
+    "model = MultiOutputClassifier(\n",
+    "    XGBClassifier(\n",
+    "        n_estimators=1600,\n",
+    "        learning_rate=0.03,\n",
+    "        max_depth=8,\n",
+    "        subsample=0.85,\n",
+    "        colsample_bytree=0.8,\n",
+    "        random_state=42,\n",
+    "        eval_metric='logloss'\n",
+    "    ),\n",
+    "    n_jobs=-1\n",
+    ")\n",
+    "\n",
+    "model.fit(X_train_scaled, y_train)\n",
+    "\n",
+    "def get_accuracy(X_scaled, y_true, name=\"\"):\n",
+    "    y_pred = model.predict(X_scaled)\n",
+    "    subset_acc = np.mean(np.all(y_pred == y_true, axis=1))\n",
+    "    micro_f1 = f1_score(y_true, y_pred, average='micro')\n",
+    "    print(f\"{name} Subset Accuracy: {subset_acc:.4f}\")\n",
+    "    print(f\"{name} Micro F1 Score: {micro_f1:.4f}\")\n",
+    "\n",
+    "print(\"=== MODEL ACCURACY ===\")\n",
+    "get_accuracy(X_train_scaled, y_train, \"Train\")\n",
+    "get_accuracy(X_test_scaled, y_test, \"Test\")\n"
+   ],
+   "id": "4689dad33e9555d7",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "=== MODEL ACCURACY ===\n",
+      "Train Subset Accuracy: 0.8454\n",
+      "Train Micro F1 Score: 0.9588\n",
+      "Test Subset Accuracy: 0.1467\n",
+      "Test Micro F1 Score: 0.3416\n"
+     ]
+    }
+   ],
+   "execution_count": 8
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T09:43:06.181862900Z",
+     "start_time": "2026-03-31T09:42:59.357899600Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "import joblib\n",
+    "bundle = {\n",
+    "    \"model\": model,\n",
+    "    \"scaler\": scaler,\n",
+    "    \"mlb\": mlb,\n",
+    "    \"feature_cols\": X.columns.tolist()\n",
+    "}\n",
+    "joblib.dump(bundle, \"all_genre_and_subgenre.pkl\")\n",
+    "print(\"Model saved successfully!\")"
+   ],
+   "id": "64f64e491b4c551f",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model saved successfully!\n"
+     ]
+    }
+   ],
+   "execution_count": 9
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T09:54:22.048035300Z",
+     "start_time": "2026-03-31T09:54:22.011206500Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "def manual_predict(row):\n",
+    "    input_df = pd.DataFrame([row])\n",
+    "    input_numeric = input_df.select_dtypes(include=[np.number]).fillna(0)\n",
+    "    for col in X.columns:\n",
+    "        if col not in input_numeric.columns:\n",
+    "            input_numeric[col] = 0\n",
+    "    input_numeric = input_numeric[X.columns]\n",
+    "    input_scaled = scaler.transform(input_numeric)\n",
+    "    probas = model.predict_proba(input_scaled)\n",
+    "    threshold = 0.17\n",
+    "    pred_binary = np.zeros((1, y.shape[1]), dtype=int)\n",
+    "    print(pred_binary)\n",
+    "    for i in range(len(probas)):\n",
+    "        if probas[i][0][1] > threshold:\n",
+    "            pred_binary[0, i] = 1\n",
+    "    subgenres = mlb.inverse_transform(pred_binary)[0]\n",
+    "    if len(subgenres) == 0:\n",
+    "        return f\"None\"\n",
+    "    return subgenres"
+   ],
+   "id": "3a6b67cc34a84ed0",
+   "outputs": [],
+   "execution_count": 19
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T09:54:22.490382600Z",
+     "start_time": "2026-03-31T09:54:22.419572Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df[[\"genre_subgenre\"]].iloc[502,]",
+   "id": "64dde8703c32c743",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "genre_subgenre    [Alternative Hip Hop, Boom Bap, Contemporary H...\n",
+       "Name: 502, dtype: object"
+      ]
+     },
+     "execution_count": 20,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 20
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T09:54:23.151824900Z",
+     "start_time": "2026-03-31T09:54:23.072274100Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "X.iloc[508].to_dict()",
+   "id": "d5d2b028522a07a5",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'spectral_contrast_mean (mix)': 18.775835817338336,\n",
+       " 'chroma_mean (mix)': 0.4393191337585449,\n",
+       " 'melody_variability (vocals)': 0.7237541064882048,\n",
+       " 'rhythm_onset_rate (mix)': 4.769279513338826,\n",
+       " 'spectral_centroid_mean custom (mix)': 0.1423719443498439,\n",
+       " 'repetition_score custom (mix)': 0.0086769169570016,\n",
+       " 'pitch_std (mix)': 70.28816223144531,\n",
+       " 'tempo_bpm_corrected (mix)': 120.0334014892578,\n",
+       " 'tempo_bpm_original (mix)': 120.0334014892578,\n",
+       " 'mfcc_mean_1 (mix)': 139.3588104248047,\n",
+       " 'mfcc_mean_2 (mix)': -10.140735626220703,\n",
+       " 'loudness_range_lu custom (mix)': 11.643880844116213,\n",
+       " 'zero_crossing_rate (mix)': 0.0576565340161323,\n",
+       " 'vocab_richness': 0.47,\n",
+       " 'loudness_integrated_lufs custom (mix)': -17.81583023071289,\n",
+       " 'readability_score': 159.6,\n",
+       " 'energy_essentia (mix)': 0.1938344933977563,\n",
+       " 'energy_librosa (mix)': 0.3731466044114756,\n",
+       " 'rms_energy_mean (mix)': 0.3730348890104648,\n",
+       " 'melody_complexity (vocals)': 3.1666666666666665}"
+      ]
+     },
+     "execution_count": 21,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 21
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T09:54:24.444755Z",
+     "start_time": "2026-03-31T09:54:23.999858200Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "print(manual_predict(X.iloc[508].to_dict()))",
+   "id": "48a6edb36a09bb6c",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[[0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n",
+      "  0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n",
+      "  0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n",
+      "  0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n",
+      "  0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]]\n",
+      "('Contemporary Hip Hop', 'Hip Hop Rap', 'Trap')\n"
+     ]
+    }
+   ],
+   "execution_count": 22
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T09:49:35.281450200Z",
+     "start_time": "2026-03-31T09:49:34.980650900Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "dict_data = {'spectral_contrast_mean (mix)': 21.826375772089914, 'chroma_mean (mix)': 0.3961135149002075, 'melody_variability (vocals)': 0.38250987027377, 'rhythm_onset_rate (mix)': 3.1277777777777778, 'spectral_centroid_mean custom (mix)': 0.15499075010897467, 'repetition_score custom (mix)': 0.0074139935621249506, 'pitch_std (mix)': 74.42411041259766, 'tempo_bpm_corrected (mix)': 128.70570373535156, 'mfcc_mean_1 (mix)': 131.99278259277344, 'mfcc_mean_2 (mix)': -21.035446166992188, 'loudness_range_lu custom (mix)': 5.404609680175781, 'zero_crossing_rate (mix)': 0.06825674325227737, 'vocab_richness': 0.455, 'loudness_integrated_lufs custom (mix)': -13.08819580078125, 'readability_score': 32.8, 'energy_essentia (mix)': 0.32994533436041884, 'energy_librosa (mix)': 0.5203621226467141, 'rms_energy_mean (mix)': 0.5467776892768835, 'melody_complexity (vocals)': 2.25}\n",
+    "\n",
+    "\n",
+    "print(manual_predict(dict_data))"
+   ],
+   "id": "2c84f939df788c50",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "('Pop', 'RnB Soul', 'Rock')\n"
+     ]
+    }
+   ],
+   "execution_count": 18
+  },
+  {
+   "metadata": {},
+   "cell_type": "code",
+   "outputs": [],
+   "execution_count": null,
+   "source": "",
+   "id": "80ec85ef86f5dbc4"
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

genre_finder/test_some_codes/main_file.ipynb

@@ -0,0 +1,2205 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "id": "initial_id",
+   "metadata": {
+    "collapsed": true,
+    "ExecuteTime": {
+     "end_time": "2026-03-30T19:07:31.853468800Z",
+     "start_time": "2026-03-30T19:07:28.033173600Z"
+    }
+   },
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import re\n",
+    "import ast\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from sklearn.preprocessing import StandardScaler, MultiLabelBinarizer\n",
+    "from sklearn.multioutput import MultiOutputClassifier\n",
+    "from xgboost import XGBClassifier\n",
+    "from sklearn.metrics import f1_score\n",
+    "pd.set_option(\"display.max_columns\", None)\n",
+    "df = pd.read_csv(\"assets/all_genre.csv\")\n",
+    "df.head()"
+   ],
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "                        _id        genre  \\\n",
+       "0  69143037d64595f86b812d77  Hip Hop Rap   \n",
+       "1  691448a64bef1dcbb1d3da1b  Hip Hop Rap   \n",
+       "2  691453c34bef1dcbb1d3de77  Hip Hop Rap   \n",
+       "3  6914609dfb8126cae4324718  Hip Hop Rap   \n",
+       "4  691af0f3979070eb7827937c  Hip Hop Rap   \n",
+       "\n",
+       "                                      genre_subgenre  \\\n",
+       "0  Hip Hop Rap --- ['Alternative Hip Hop', 'Boom ...   \n",
+       "1  Hip Hop Rap --- ['Americana', 'Contemporary Co...   \n",
+       "2  Hip Hop Rap --- ['Alternative Hip Hop', 'Boom ...   \n",
+       "3  Hip Hop Rap --- ['Alternative Hip Hop', 'Conte...   \n",
+       "4  Hip Hop Rap --- ['Alternative Hip Hop', 'Conte...   \n",
+       "\n",
+       "   trimmed_audio_duration_sec  syllable_count  word_count  \\\n",
+       "0                  183.843991           355.0       298.0   \n",
+       "1                  160.786576           285.0       236.0   \n",
+       "2                  202.320000           350.0       285.0   \n",
+       "3                  187.298798           318.0       285.0   \n",
+       "4                  185.004989           651.0       551.0   \n",
+       "\n",
+       "   spectral_contrast_mean (mix)  chroma_mean (mix)  \\\n",
+       "0                     20.440557           0.464204   \n",
+       "1                     19.657228           0.424237   \n",
+       "2                     20.115561           0.454093   \n",
+       "3                     19.659283           0.443467   \n",
+       "4                     20.855880           0.461875   \n",
+       "\n",
+       "   melody_variability (vocals)  rhythm_onset_rate (mix)  \\\n",
+       "0                     0.434021                 3.250000   \n",
+       "1                     0.351001                 2.425576   \n",
+       "2                     0.327000                 2.738889   \n",
+       "3                     0.360642                 3.500000   \n",
+       "4                     0.767493                 4.983333   \n",
+       "\n",
+       "   spectral_centroid_mean custom (mix)  repetition_score custom (mix)  \\\n",
+       "0                             0.132002                       0.006988   \n",
+       "1                             0.139757                       0.007367   \n",
+       "2                             0.111282                       0.006671   \n",
+       "3                             0.159247                       0.007202   \n",
+       "4                             0.155302                       0.007290   \n",
+       "\n",
+       "   pitch_std (mix)  tempo_bpm_corrected (mix)  tempo_bpm_original (mix)  \\\n",
+       "0        69.616264                 152.006775                152.006775   \n",
+       "1       156.358124                 117.943665                117.943665   \n",
+       "2        69.311798                 124.344086                124.344086   \n",
+       "3        65.514191                 109.909401                109.909401   \n",
+       "4       127.873909                 151.955536                151.955536   \n",
+       "\n",
+       "   mfcc_mean_1 (mix)  mfcc_mean_2 (mix)  loudness_range_lu custom (mix)  \\\n",
+       "0         153.058243          -7.368185                        1.143722   \n",
+       "1         148.055832         -49.435444                        4.708925   \n",
+       "2         166.170502          -5.852028                        1.948263   \n",
+       "3         134.211624         -27.524824                        1.244652   \n",
+       "4         129.783264         -11.940094                        3.452597   \n",
+       "\n",
+       "   zero_crossing_rate (mix)  vocab_richness  \\\n",
+       "0                  0.045159           0.305   \n",
+       "1                  0.070352           0.352   \n",
+       "2                  0.036463           0.400   \n",
+       "3                  0.064695           0.365   \n",
+       "4                  0.056507           0.281   \n",
+       "\n",
+       "   loudness_integrated_lufs custom (mix)  readability_score  \\\n",
+       "0                             -12.104649               20.9   \n",
+       "1                             -11.814197               89.4   \n",
+       "2                             -12.000179              107.0   \n",
+       "3                             -12.435019              105.8   \n",
+       "4                             -16.217091              212.3   \n",
+       "\n",
+       "   energy_essentia (mix)  energy_librosa (mix)  rms_energy_mean (mix)  \\\n",
+       "0               0.349783              0.576250               0.578454   \n",
+       "1               0.429792              0.619931               0.632507   \n",
+       "2               0.325114              0.520390               0.579053   \n",
+       "3               0.407647              0.617191               0.617197   \n",
+       "4               0.255311              0.446953               0.454712   \n",
+       "\n",
+       "   sentiment_score  melody_complexity (vocals)  avg_word_length  \n",
+       "0            0.116                    2.250000             3.76  \n",
+       "1            0.173                    1.916667             3.50  \n",
+       "2            0.158                    1.833333             3.73  \n",
+       "3            0.071                    1.916667             3.48  \n",
+       "4            0.039                    3.500000             3.56  "
+      ],
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>_id</th>\n",
+       "      <th>genre</th>\n",
+       "      <th>genre_subgenre</th>\n",
+       "      <th>trimmed_audio_duration_sec</th>\n",
+       "      <th>syllable_count</th>\n",
+       "      <th>word_count</th>\n",
+       "      <th>spectral_contrast_mean (mix)</th>\n",
+       "      <th>chroma_mean (mix)</th>\n",
+       "      <th>melody_variability (vocals)</th>\n",
+       "      <th>rhythm_onset_rate (mix)</th>\n",
+       "      <th>spectral_centroid_mean custom (mix)</th>\n",
+       "      <th>repetition_score custom (mix)</th>\n",
+       "      <th>pitch_std (mix)</th>\n",
+       "      <th>tempo_bpm_corrected (mix)</th>\n",
+       "      <th>tempo_bpm_original (mix)</th>\n",
+       "      <th>mfcc_mean_1 (mix)</th>\n",
+       "      <th>mfcc_mean_2 (mix)</th>\n",
+       "      <th>loudness_range_lu custom (mix)</th>\n",
+       "      <th>zero_crossing_rate (mix)</th>\n",
+       "      <th>vocab_richness</th>\n",
+       "      <th>loudness_integrated_lufs custom (mix)</th>\n",
+       "      <th>readability_score</th>\n",
+       "      <th>energy_essentia (mix)</th>\n",
+       "      <th>energy_librosa (mix)</th>\n",
+       "      <th>rms_energy_mean (mix)</th>\n",
+       "      <th>sentiment_score</th>\n",
+       "      <th>melody_complexity (vocals)</th>\n",
+       "      <th>avg_word_length</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>69143037d64595f86b812d77</td>\n",
+       "      <td>Hip Hop Rap</td>\n",
+       "      <td>Hip Hop Rap --- ['Alternative Hip Hop', 'Boom ...</td>\n",
+       "      <td>183.843991</td>\n",
+       "      <td>355.0</td>\n",
+       "      <td>298.0</td>\n",
+       "      <td>20.440557</td>\n",
+       "      <td>0.464204</td>\n",
+       "      <td>0.434021</td>\n",
+       "      <td>3.250000</td>\n",
+       "      <td>0.132002</td>\n",
+       "      <td>0.006988</td>\n",
+       "      <td>69.616264</td>\n",
+       "      <td>152.006775</td>\n",
+       "      <td>152.006775</td>\n",
+       "      <td>153.058243</td>\n",
+       "      <td>-7.368185</td>\n",
+       "      <td>1.143722</td>\n",
+       "      <td>0.045159</td>\n",
+       "      <td>0.305</td>\n",
+       "      <td>-12.104649</td>\n",
+       "      <td>20.9</td>\n",
+       "      <td>0.349783</td>\n",
+       "      <td>0.576250</td>\n",
+       "      <td>0.578454</td>\n",
+       "      <td>0.116</td>\n",
+       "      <td>2.250000</td>\n",
+       "      <td>3.76</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>691448a64bef1dcbb1d3da1b</td>\n",
+       "      <td>Hip Hop Rap</td>\n",
+       "      <td>Hip Hop Rap --- ['Americana', 'Contemporary Co...</td>\n",
+       "      <td>160.786576</td>\n",
+       "      <td>285.0</td>\n",
+       "      <td>236.0</td>\n",
+       "      <td>19.657228</td>\n",
+       "      <td>0.424237</td>\n",
+       "      <td>0.351001</td>\n",
+       "      <td>2.425576</td>\n",
+       "      <td>0.139757</td>\n",
+       "      <td>0.007367</td>\n",
+       "      <td>156.358124</td>\n",
+       "      <td>117.943665</td>\n",
+       "      <td>117.943665</td>\n",
+       "      <td>148.055832</td>\n",
+       "      <td>-49.435444</td>\n",
+       "      <td>4.708925</td>\n",
+       "      <td>0.070352</td>\n",
+       "      <td>0.352</td>\n",
+       "      <td>-11.814197</td>\n",
+       "      <td>89.4</td>\n",
+       "      <td>0.429792</td>\n",
+       "      <td>0.619931</td>\n",
+       "      <td>0.632507</td>\n",
+       "      <td>0.173</td>\n",
+       "      <td>1.916667</td>\n",
+       "      <td>3.50</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>691453c34bef1dcbb1d3de77</td>\n",
+       "      <td>Hip Hop Rap</td>\n",
+       "      <td>Hip Hop Rap --- ['Alternative Hip Hop', 'Boom ...</td>\n",
+       "      <td>202.320000</td>\n",
+       "      <td>350.0</td>\n",
+       "      <td>285.0</td>\n",
+       "      <td>20.115561</td>\n",
+       "      <td>0.454093</td>\n",
+       "      <td>0.327000</td>\n",
+       "      <td>2.738889</td>\n",
+       "      <td>0.111282</td>\n",
+       "      <td>0.006671</td>\n",
+       "      <td>69.311798</td>\n",
+       "      <td>124.344086</td>\n",
+       "      <td>124.344086</td>\n",
+       "      <td>166.170502</td>\n",
+       "      <td>-5.852028</td>\n",
+       "      <td>1.948263</td>\n",
+       "      <td>0.036463</td>\n",
+       "      <td>0.400</td>\n",
+       "      <td>-12.000179</td>\n",
+       "      <td>107.0</td>\n",
+       "      <td>0.325114</td>\n",
+       "      <td>0.520390</td>\n",
+       "      <td>0.579053</td>\n",
+       "      <td>0.158</td>\n",
+       "      <td>1.833333</td>\n",
+       "      <td>3.73</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>6914609dfb8126cae4324718</td>\n",
+       "      <td>Hip Hop Rap</td>\n",
+       "      <td>Hip Hop Rap --- ['Alternative Hip Hop', 'Conte...</td>\n",
+       "      <td>187.298798</td>\n",
+       "      <td>318.0</td>\n",
+       "      <td>285.0</td>\n",
+       "      <td>19.659283</td>\n",
+       "      <td>0.443467</td>\n",
+       "      <td>0.360642</td>\n",
+       "      <td>3.500000</td>\n",
+       "      <td>0.159247</td>\n",
+       "      <td>0.007202</td>\n",
+       "      <td>65.514191</td>\n",
+       "      <td>109.909401</td>\n",
+       "      <td>109.909401</td>\n",
+       "      <td>134.211624</td>\n",
+       "      <td>-27.524824</td>\n",
+       "      <td>1.244652</td>\n",
+       "      <td>0.064695</td>\n",
+       "      <td>0.365</td>\n",
+       "      <td>-12.435019</td>\n",
+       "      <td>105.8</td>\n",
+       "      <td>0.407647</td>\n",
+       "      <td>0.617191</td>\n",
+       "      <td>0.617197</td>\n",
+       "      <td>0.071</td>\n",
+       "      <td>1.916667</td>\n",
+       "      <td>3.48</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>691af0f3979070eb7827937c</td>\n",
+       "      <td>Hip Hop Rap</td>\n",
+       "      <td>Hip Hop Rap --- ['Alternative Hip Hop', 'Conte...</td>\n",
+       "      <td>185.004989</td>\n",
+       "      <td>651.0</td>\n",
+       "      <td>551.0</td>\n",
+       "      <td>20.855880</td>\n",
+       "      <td>0.461875</td>\n",
+       "      <td>0.767493</td>\n",
+       "      <td>4.983333</td>\n",
+       "      <td>0.155302</td>\n",
+       "      <td>0.007290</td>\n",
+       "      <td>127.873909</td>\n",
+       "      <td>151.955536</td>\n",
+       "      <td>151.955536</td>\n",
+       "      <td>129.783264</td>\n",
+       "      <td>-11.940094</td>\n",
+       "      <td>3.452597</td>\n",
+       "      <td>0.056507</td>\n",
+       "      <td>0.281</td>\n",
+       "      <td>-16.217091</td>\n",
+       "      <td>212.3</td>\n",
+       "      <td>0.255311</td>\n",
+       "      <td>0.446953</td>\n",
+       "      <td>0.454712</td>\n",
+       "      <td>0.039</td>\n",
+       "      <td>3.500000</td>\n",
+       "      <td>3.56</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 1
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T19:07:36.085695400Z",
+     "start_time": "2026-03-30T19:07:35.837094700Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "def parse_subgenres_fast(text):\n",
+    "    if pd.isna(text) or text == \"\":\n",
+    "        return []\n",
+    "    text = str(text)\n",
+    "    parts = text.split('---')\n",
+    "\n",
+    "    main_genre = parts[0].strip()\n",
+    "\n",
+    "    if len(parts) > 1:\n",
+    "        sub = parts[1].strip().strip('[]')\n",
+    "        subgenres = [s.strip().strip(\"'\").strip('\"') for s in sub.split(',')] if sub else []\n",
+    "    else:\n",
+    "        subgenres = []\n",
+    "\n",
+    "    return [main_genre] + subgenres\n",
+    "df['subgenres_list'] = df['genre_subgenre'].apply(parse_subgenres_fast)"
+   ],
+   "id": "67098b6deadbc3ef",
+   "outputs": [],
+   "execution_count": 2
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T19:07:39.104448300Z",
+     "start_time": "2026-03-30T19:07:39.016728300Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df['subgenres_list'][0]",
+   "id": "e0fc9a9de1291e67",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['Hip Hop Rap',\n",
+       " 'Alternative Hip Hop',\n",
+       " 'Boom Bap',\n",
+       " 'Contemporary Hip Hop',\n",
+       " 'Trap']"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 3
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "start_time": "2026-03-30T19:07:51.575299200Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "mlb = MultiLabelBinarizer()\n",
+    "y = mlb.fit_transform(df['subgenres_list'])\n",
+    "\n",
+    "drop_cols = ['_id', 'genre', 'genre_subgenre', 'subgenres_list', 'labels',\n",
+    "             'trimmed_audio_duration_sec', 'syllable_count', 'word_count',\n",
+    "             'sentiment_score', 'avg_word_length']\n",
+    "\n",
+    "feature_cols = [c for c in df.columns if c not in drop_cols]\n",
+    "X = df[feature_cols].copy()\n",
+    "X = X.select_dtypes(include=[np.number]).fillna(0)\n",
+    "\n",
+    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
+    "\n",
+    "scaler = StandardScaler()\n",
+    "X_train_scaled = scaler.fit_transform(X_train)\n",
+    "X_test_scaled = scaler.transform(X_test)\n",
+    "\n",
+    "model = MultiOutputClassifier(\n",
+    "    XGBClassifier(\n",
+    "        n_estimators=1500,\n",
+    "        learning_rate=0.03,\n",
+    "        max_depth=8,\n",
+    "        subsample=0.85,\n",
+    "        colsample_bytree=0.8,\n",
+    "        random_state=42,\n",
+    "        eval_metric='logloss'\n",
+    "    ),\n",
+    "    n_jobs=-1\n",
+    ")\n",
+    "\n",
+    "model.fit(X_train_scaled, y_train)\n",
+    "\n",
+    "\n",
+    "def get_accuracy(X_scaled, y_true, name=\"\"):\n",
+    "    y_pred = model.predict(X_scaled)\n",
+    "    subset_acc = np.mean(np.all(y_pred == y_true, axis=1))\n",
+    "    micro_f1 = f1_score(y_true, y_pred, average='micro')\n",
+    "    print(f\"{name} Subset Accuracy: {subset_acc:.4f}\")\n",
+    "    print(f\"{name} Micro F1 Score: {micro_f1:.4f}\")\n",
+    "\n",
+    "print(\"=== MODEL ACCURACY ===\")\n",
+    "get_accuracy(X_train_scaled, y_train, \"Train\")\n",
+    "get_accuracy(X_test_scaled, y_test, \"Test\")\n",
+    "\n",
+    "def manual_predict(row):\n",
+    "    input_df = pd.DataFrame([row])\n",
+    "    input_numeric = input_df.select_dtypes(include=[np.number]).fillna(0)\n",
+    "\n",
+    "    for col in X.columns:\n",
+    "        if col not in input_numeric.columns:\n",
+    "            input_numeric[col] = 0\n",
+    "\n",
+    "    input_numeric = input_numeric[X.columns]\n",
+    "    input_scaled = scaler.transform(input_numeric)\n",
+    "\n",
+    "    probas = model.predict_proba(input_scaled)\n",
+    "    threshold = 0.20\n",
+    "    pred_binary = np.zeros((1, y.shape[1]), dtype=int)\n",
+    "\n",
+    "    for i in range(len(probas)):\n",
+    "        if probas[i][0][1] > threshold:\n",
+    "            pred_binary[0, i] = 1\n",
+    "\n",
+    "    subgenres = mlb.inverse_transform(pred_binary)[0]\n",
+    "    #main_genre = row.get('genre', 'Rock')\n",
+    "\n",
+    "    if len(subgenres) == 0:\n",
+    "        return f\"None\"\n",
+    "    return f\"{', '.join(sorted(subgenres))}\""
+   ],
+   "id": "c1086565f89c486b",
+   "outputs": [],
+   "execution_count": null
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T14:45:34.721003500Z",
+     "start_time": "2026-03-27T14:45:34.665870Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "X.head()",
+   "id": "7036f929b47b9932",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "   spectral_contrast_mean (mix)  chroma_mean (mix)  \\\n",
+       "0                     18.938621           0.462013   \n",
+       "1                     18.361412           0.515950   \n",
+       "2                     17.545991           0.505095   \n",
+       "3                     19.065528           0.532723   \n",
+       "4                     18.549622           0.560616   \n",
+       "\n",
+       "   melody_variability (vocals)  rhythm_onset_rate (mix)  \\\n",
+       "0                     0.887239                 0.550994   \n",
+       "1                     0.553844                 1.785610   \n",
+       "2                     0.887161                 1.691063   \n",
+       "3                     0.917722                 4.237960   \n",
+       "4                     0.967337                 0.847676   \n",
+       "\n",
+       "   spectral_centroid_mean custom (mix)  repetition_score custom (mix)  \\\n",
+       "0                             0.124953                       0.015320   \n",
+       "1                             0.178827                       0.010278   \n",
+       "2                             0.115442                       0.010617   \n",
+       "3                             0.134564                       0.009808   \n",
+       "4                             0.179751                       0.008122   \n",
+       "\n",
+       "   pitch_std (mix)  tempo_bpm_corrected (mix)  tempo_bpm_original (mix)  \\\n",
+       "0       170.196686                 117.459671                117.459671   \n",
+       "1       174.526428                  93.132881                 93.132881   \n",
+       "2        98.435074                 105.041779                105.041779   \n",
+       "3       144.853897                  94.973419                 94.973419   \n",
+       "4        94.283890                 108.527031                108.527031   \n",
+       "\n",
+       "   mfcc_mean_1 (mix)  mfcc_mean_2 (mix)  loudness_range_lu custom (mix)  \\\n",
+       "0         159.208038         -47.135399                       17.099110   \n",
+       "1         124.982330         -54.048897                       10.356828   \n",
+       "2         178.284927         -41.860600                        4.583549   \n",
+       "3         166.110214         -79.556442                        3.334763   \n",
+       "4         118.163094         -54.811043                        0.855955   \n",
+       "\n",
+       "   zero_crossing_rate (mix)  vocab_richness  \\\n",
+       "0                  0.065750           0.842   \n",
+       "1                  0.115698           0.312   \n",
+       "2                  0.050672           0.462   \n",
+       "3                  0.085350           0.273   \n",
+       "4                  0.104765           0.350   \n",
+       "\n",
+       "   loudness_integrated_lufs custom (mix)  readability_score  \\\n",
+       "0                             -17.450430                1.6   \n",
+       "1                             -14.291489                5.2   \n",
+       "2                             -14.021948               18.8   \n",
+       "3                             -11.496668                1.7   \n",
+       "4                             -10.031396               52.0   \n",
+       "\n",
+       "   energy_essentia (mix)  energy_librosa (mix)  rms_energy_mean (mix)  \\\n",
+       "0               0.400498              0.541834               0.545006   \n",
+       "1               0.286286              0.491097               0.492117   \n",
+       "2               0.330273              0.539636               0.541554   \n",
+       "3               0.525465              0.693286               0.702081   \n",
+       "4               0.498986              0.696552               0.698920   \n",
+       "\n",
+       "   melody_complexity (vocals)  \n",
+       "0                    2.833333  \n",
+       "1                    2.333333  \n",
+       "2                    3.083333  \n",
+       "3                    3.166667  \n",
+       "4                    3.500000  "
+      ],
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>spectral_contrast_mean (mix)</th>\n",
+       "      <th>chroma_mean (mix)</th>\n",
+       "      <th>melody_variability (vocals)</th>\n",
+       "      <th>rhythm_onset_rate (mix)</th>\n",
+       "      <th>spectral_centroid_mean custom (mix)</th>\n",
+       "      <th>repetition_score custom (mix)</th>\n",
+       "      <th>pitch_std (mix)</th>\n",
+       "      <th>tempo_bpm_corrected (mix)</th>\n",
+       "      <th>tempo_bpm_original (mix)</th>\n",
+       "      <th>mfcc_mean_1 (mix)</th>\n",
+       "      <th>mfcc_mean_2 (mix)</th>\n",
+       "      <th>loudness_range_lu custom (mix)</th>\n",
+       "      <th>zero_crossing_rate (mix)</th>\n",
+       "      <th>vocab_richness</th>\n",
+       "      <th>loudness_integrated_lufs custom (mix)</th>\n",
+       "      <th>readability_score</th>\n",
+       "      <th>energy_essentia (mix)</th>\n",
+       "      <th>energy_librosa (mix)</th>\n",
+       "      <th>rms_energy_mean (mix)</th>\n",
+       "      <th>melody_complexity (vocals)</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>18.938621</td>\n",
+       "      <td>0.462013</td>\n",
+       "      <td>0.887239</td>\n",
+       "      <td>0.550994</td>\n",
+       "      <td>0.124953</td>\n",
+       "      <td>0.015320</td>\n",
+       "      <td>170.196686</td>\n",
+       "      <td>117.459671</td>\n",
+       "      <td>117.459671</td>\n",
+       "      <td>159.208038</td>\n",
+       "      <td>-47.135399</td>\n",
+       "      <td>17.099110</td>\n",
+       "      <td>0.065750</td>\n",
+       "      <td>0.842</td>\n",
+       "      <td>-17.450430</td>\n",
+       "      <td>1.6</td>\n",
+       "      <td>0.400498</td>\n",
+       "      <td>0.541834</td>\n",
+       "      <td>0.545006</td>\n",
+       "      <td>2.833333</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>18.361412</td>\n",
+       "      <td>0.515950</td>\n",
+       "      <td>0.553844</td>\n",
+       "      <td>1.785610</td>\n",
+       "      <td>0.178827</td>\n",
+       "      <td>0.010278</td>\n",
+       "      <td>174.526428</td>\n",
+       "      <td>93.132881</td>\n",
+       "      <td>93.132881</td>\n",
+       "      <td>124.982330</td>\n",
+       "      <td>-54.048897</td>\n",
+       "      <td>10.356828</td>\n",
+       "      <td>0.115698</td>\n",
+       "      <td>0.312</td>\n",
+       "      <td>-14.291489</td>\n",
+       "      <td>5.2</td>\n",
+       "      <td>0.286286</td>\n",
+       "      <td>0.491097</td>\n",
+       "      <td>0.492117</td>\n",
+       "      <td>2.333333</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>17.545991</td>\n",
+       "      <td>0.505095</td>\n",
+       "      <td>0.887161</td>\n",
+       "      <td>1.691063</td>\n",
+       "      <td>0.115442</td>\n",
+       "      <td>0.010617</td>\n",
+       "      <td>98.435074</td>\n",
+       "      <td>105.041779</td>\n",
+       "      <td>105.041779</td>\n",
+       "      <td>178.284927</td>\n",
+       "      <td>-41.860600</td>\n",
+       "      <td>4.583549</td>\n",
+       "      <td>0.050672</td>\n",
+       "      <td>0.462</td>\n",
+       "      <td>-14.021948</td>\n",
+       "      <td>18.8</td>\n",
+       "      <td>0.330273</td>\n",
+       "      <td>0.539636</td>\n",
+       "      <td>0.541554</td>\n",
+       "      <td>3.083333</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>19.065528</td>\n",
+       "      <td>0.532723</td>\n",
+       "      <td>0.917722</td>\n",
+       "      <td>4.237960</td>\n",
+       "      <td>0.134564</td>\n",
+       "      <td>0.009808</td>\n",
+       "      <td>144.853897</td>\n",
+       "      <td>94.973419</td>\n",
+       "      <td>94.973419</td>\n",
+       "      <td>166.110214</td>\n",
+       "      <td>-79.556442</td>\n",
+       "      <td>3.334763</td>\n",
+       "      <td>0.085350</td>\n",
+       "      <td>0.273</td>\n",
+       "      <td>-11.496668</td>\n",
+       "      <td>1.7</td>\n",
+       "      <td>0.525465</td>\n",
+       "      <td>0.693286</td>\n",
+       "      <td>0.702081</td>\n",
+       "      <td>3.166667</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>18.549622</td>\n",
+       "      <td>0.560616</td>\n",
+       "      <td>0.967337</td>\n",
+       "      <td>0.847676</td>\n",
+       "      <td>0.179751</td>\n",
+       "      <td>0.008122</td>\n",
+       "      <td>94.283890</td>\n",
+       "      <td>108.527031</td>\n",
+       "      <td>108.527031</td>\n",
+       "      <td>118.163094</td>\n",
+       "      <td>-54.811043</td>\n",
+       "      <td>0.855955</td>\n",
+       "      <td>0.104765</td>\n",
+       "      <td>0.350</td>\n",
+       "      <td>-10.031396</td>\n",
+       "      <td>52.0</td>\n",
+       "      <td>0.498986</td>\n",
+       "      <td>0.696552</td>\n",
+       "      <td>0.698920</td>\n",
+       "      <td>3.500000</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 6
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T14:55:56.357032800Z",
+     "start_time": "2026-03-27T14:55:56.275652600Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "X.shape",
+   "id": "f7d2011c6b5f8699",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(41935, 20)"
+      ]
+     },
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 15
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T15:10:02.865313500Z",
+     "start_time": "2026-03-27T15:10:02.807027900Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df[[\"subgenres_list\",\"genre\"]].iloc[502,]",
+   "id": "d462801f323e88a",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "subgenres_list    [Modern Alternative Rock, Indie Rock]\n",
+       "genre                                              Rock\n",
+       "Name: 502, dtype: object"
+      ]
+     },
+     "execution_count": 43,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 43
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T15:13:50.114563800Z",
+     "start_time": "2026-03-27T15:13:50.060507800Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "pd.set_option('display.max_columns', None)\n",
+    "df[(df[\"genre\"]==\"Metal\")].head(2)"
+   ],
+   "id": "1fbd1d7876fb8996",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "                            _id  genre  \\\n",
+       "41856  691b00908e8610886093475b  Metal   \n",
+       "41857  691c4b322bf3d72bd0be023e  Metal   \n",
+       "\n",
+       "                                          genre_subgenre  \\\n",
+       "41856  Metal --- ['Black Metal', 'Death Metal', 'Doom...   \n",
+       "41857          Metal --- ['Heavy Metal', 'Thrash Metal']   \n",
+       "\n",
+       "       trimmed_audio_duration_sec  syllable_count  word_count  \\\n",
+       "41856                  185.602971           185.0       158.0   \n",
+       "41857                  131.819683           353.0       253.0   \n",
+       "\n",
+       "       spectral_contrast_mean (mix)  chroma_mean (mix)  \\\n",
+       "41856                     20.152063           0.408171   \n",
+       "41857                     19.495984           0.517214   \n",
+       "\n",
+       "       melody_variability (vocals)  rhythm_onset_rate (mix)  \\\n",
+       "41856                     0.374856                 2.027778   \n",
+       "41857                     0.475671                 5.803382   \n",
+       "\n",
+       "       spectral_centroid_mean custom (mix)  repetition_score custom (mix)  \\\n",
+       "41856                             0.093899                       0.007882   \n",
+       "41857                             0.168338                       0.007993   \n",
+       "\n",
+       "       pitch_std (mix)  tempo_bpm_corrected (mix)  tempo_bpm_original (mix)  \\\n",
+       "41856       108.278755                 104.605293                104.605293   \n",
+       "41857       134.086761                 138.027847                138.027847   \n",
+       "\n",
+       "       mfcc_mean_1 (mix)  mfcc_mean_2 (mix)  loudness_range_lu custom (mix)  \\\n",
+       "41856         163.308655          22.928051                        7.131201   \n",
+       "41857         113.692780         -26.532072                        2.994879   \n",
+       "\n",
+       "       zero_crossing_rate (mix)  vocab_richness  \\\n",
+       "41856                  0.026933           0.582   \n",
+       "41857                  0.069337           0.352   \n",
+       "\n",
+       "       loudness_integrated_lufs custom (mix)  readability_score  \\\n",
+       "41856                             -18.566132               27.8   \n",
+       "41857                             -13.850066                3.0   \n",
+       "\n",
+       "       energy_essentia (mix)  energy_librosa (mix)  rms_energy_mean (mix)  \\\n",
+       "41856               0.153738              0.335069               0.346289   \n",
+       "41857               0.266081              0.441120               0.467461   \n",
+       "\n",
+       "       sentiment_score  melody_complexity (vocals)  avg_word_length  \\\n",
+       "41856            0.050                    1.916667             3.97   \n",
+       "41857            0.128                    2.250000             4.24   \n",
+       "\n",
+       "                                          subgenres_list  \n",
+       "41856  [Black Metal, Death Metal, Doom Metal, Heavy M...  \n",
+       "41857                        [Heavy Metal, Thrash Metal]  "
+      ],
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>_id</th>\n",
+       "      <th>genre</th>\n",
+       "      <th>genre_subgenre</th>\n",
+       "      <th>trimmed_audio_duration_sec</th>\n",
+       "      <th>syllable_count</th>\n",
+       "      <th>word_count</th>\n",
+       "      <th>spectral_contrast_mean (mix)</th>\n",
+       "      <th>chroma_mean (mix)</th>\n",
+       "      <th>melody_variability (vocals)</th>\n",
+       "      <th>rhythm_onset_rate (mix)</th>\n",
+       "      <th>spectral_centroid_mean custom (mix)</th>\n",
+       "      <th>repetition_score custom (mix)</th>\n",
+       "      <th>pitch_std (mix)</th>\n",
+       "      <th>tempo_bpm_corrected (mix)</th>\n",
+       "      <th>tempo_bpm_original (mix)</th>\n",
+       "      <th>mfcc_mean_1 (mix)</th>\n",
+       "      <th>mfcc_mean_2 (mix)</th>\n",
+       "      <th>loudness_range_lu custom (mix)</th>\n",
+       "      <th>zero_crossing_rate (mix)</th>\n",
+       "      <th>vocab_richness</th>\n",
+       "      <th>loudness_integrated_lufs custom (mix)</th>\n",
+       "      <th>readability_score</th>\n",
+       "      <th>energy_essentia (mix)</th>\n",
+       "      <th>energy_librosa (mix)</th>\n",
+       "      <th>rms_energy_mean (mix)</th>\n",
+       "      <th>sentiment_score</th>\n",
+       "      <th>melody_complexity (vocals)</th>\n",
+       "      <th>avg_word_length</th>\n",
+       "      <th>subgenres_list</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>41856</th>\n",
+       "      <td>691b00908e8610886093475b</td>\n",
+       "      <td>Metal</td>\n",
+       "      <td>Metal --- ['Black Metal', 'Death Metal', 'Doom...</td>\n",
+       "      <td>185.602971</td>\n",
+       "      <td>185.0</td>\n",
+       "      <td>158.0</td>\n",
+       "      <td>20.152063</td>\n",
+       "      <td>0.408171</td>\n",
+       "      <td>0.374856</td>\n",
+       "      <td>2.027778</td>\n",
+       "      <td>0.093899</td>\n",
+       "      <td>0.007882</td>\n",
+       "      <td>108.278755</td>\n",
+       "      <td>104.605293</td>\n",
+       "      <td>104.605293</td>\n",
+       "      <td>163.308655</td>\n",
+       "      <td>22.928051</td>\n",
+       "      <td>7.131201</td>\n",
+       "      <td>0.026933</td>\n",
+       "      <td>0.582</td>\n",
+       "      <td>-18.566132</td>\n",
+       "      <td>27.8</td>\n",
+       "      <td>0.153738</td>\n",
+       "      <td>0.335069</td>\n",
+       "      <td>0.346289</td>\n",
+       "      <td>0.050</td>\n",
+       "      <td>1.916667</td>\n",
+       "      <td>3.97</td>\n",
+       "      <td>[Black Metal, Death Metal, Doom Metal, Heavy M...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>41857</th>\n",
+       "      <td>691c4b322bf3d72bd0be023e</td>\n",
+       "      <td>Metal</td>\n",
+       "      <td>Metal --- ['Heavy Metal', 'Thrash Metal']</td>\n",
+       "      <td>131.819683</td>\n",
+       "      <td>353.0</td>\n",
+       "      <td>253.0</td>\n",
+       "      <td>19.495984</td>\n",
+       "      <td>0.517214</td>\n",
+       "      <td>0.475671</td>\n",
+       "      <td>5.803382</td>\n",
+       "      <td>0.168338</td>\n",
+       "      <td>0.007993</td>\n",
+       "      <td>134.086761</td>\n",
+       "      <td>138.027847</td>\n",
+       "      <td>138.027847</td>\n",
+       "      <td>113.692780</td>\n",
+       "      <td>-26.532072</td>\n",
+       "      <td>2.994879</td>\n",
+       "      <td>0.069337</td>\n",
+       "      <td>0.352</td>\n",
+       "      <td>-13.850066</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>0.266081</td>\n",
+       "      <td>0.441120</td>\n",
+       "      <td>0.467461</td>\n",
+       "      <td>0.128</td>\n",
+       "      <td>2.250000</td>\n",
+       "      <td>4.24</td>\n",
+       "      <td>[Heavy Metal, Thrash Metal]</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ]
+     },
+     "execution_count": 50,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 50
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T15:24:32.000214700Z",
+     "start_time": "2026-03-27T15:24:31.938664600Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "X.iloc[41856].to_dict()",
+   "id": "37ef9304bdae300c",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'spectral_contrast_mean (mix)': 20.1520629356514,\n",
+       " 'chroma_mean (mix)': 0.408170998096466,\n",
+       " 'melody_variability (vocals)': 0.3748557548841115,\n",
+       " 'rhythm_onset_rate (mix)': 2.0277777777777777,\n",
+       " 'spectral_centroid_mean custom (mix)': 0.0938994687064515,\n",
+       " 'repetition_score custom (mix)': 0.0078822056596288,\n",
+       " 'pitch_std (mix)': 108.27875518798828,\n",
+       " 'tempo_bpm_corrected (mix)': 104.60529327392578,\n",
+       " 'tempo_bpm_original (mix)': 104.60529327392578,\n",
+       " 'mfcc_mean_1 (mix)': 163.30865478515625,\n",
+       " 'mfcc_mean_2 (mix)': 22.928050994873047,\n",
+       " 'loudness_range_lu custom (mix)': 7.131200790405273,\n",
+       " 'zero_crossing_rate (mix)': 0.02693298086524,\n",
+       " 'vocab_richness': 0.582,\n",
+       " 'loudness_integrated_lufs custom (mix)': -18.566131591796875,\n",
+       " 'readability_score': 27.8,\n",
+       " 'energy_essentia (mix)': 0.1537376961168162,\n",
+       " 'energy_librosa (mix)': 0.335069430512918,\n",
+       " 'rms_energy_mean (mix)': 0.3462888819034169,\n",
+       " 'melody_complexity (vocals)': 1.9166666666666667}"
+      ]
+     },
+     "execution_count": 58,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 58
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T15:32:09.375740200Z",
+     "start_time": "2026-03-27T15:32:08.891439Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "print(\"\\n=== FIRST ROW PREDICTION ===\")\n",
+    "print(manual_predict(X.iloc[508].to_dict()))\n",
+    "\n",
+    "print(\"\\n=== LAST ROW PREDICTION ===\")\n",
+    "print(manual_predict(X.iloc[-1].to_dict()))"
+   ],
+   "id": "fef35c101a9a6ad",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "=== FIRST ROW PREDICTION ===\n",
+      "Indie Pop, Indie Rock, Modern Alternative Rock\n",
+      "\n",
+      "=== LAST ROW PREDICTION ===\n",
+      "Gothic Metal, Heavy Metal, Metalcore\n"
+     ]
+    }
+   ],
+   "execution_count": 63
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T15:31:39.900012500Z",
+     "start_time": "2026-03-27T15:31:39.638999300Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "audio_features_ = {\n",
+    "    'spectral_contrast_mean (mix)': 17.245,\n",
+    "    'chroma_mean (mix)': 0.482,\n",
+    "    'melody_variability (vocals)': 0.612,\n",
+    "    'rhythm_onset_rate (mix)': 2.145,\n",
+    "    'spectral_centroid_mean custom (mix)': 0.165,\n",
+    "    'repetition_score custom (mix)': 0.015,\n",
+    "    'pitch_std (mix)': 120.75,\n",
+    "    'tempo_bpm_corrected (mix)': 98.34,\n",
+    "    'tempo_bpm_original (mix)': 100.12,\n",
+    "    'mfcc_mean_1 (mix)': 128.56,\n",
+    "    'mfcc_mean_2 (mix)': -35.78,\n",
+    "    'loudness_range_lu custom (mix)': 8.45,\n",
+    "    'zero_crossing_rate (mix)': 0.089,\n",
+    "    'vocab_richness': 0.29,\n",
+    "    'loudness_integrated_lufs custom (mix)': -12.85,\n",
+    "    'readability_score': 48.5,\n",
+    "    'energy_essentia (mix)': 0.41,\n",
+    "    'energy_librosa (mix)': 0.52,\n",
+    "    'rms_energy_mean (mix)': 0.50,\n",
+    "    'melody_complexity (vocals)': 2.75\n",
+    "}\n",
+    "\n",
+    "answer = manual_predict(audio_features_)\n",
+    "print(answer)"
+   ],
+   "id": "6d8f315863783a68",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Indie Pop, Indie Rock, Modern Alternative Rock\n"
+     ]
+    }
+   ],
+   "execution_count": 61
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T15:31:41.650948200Z",
+     "start_time": "2026-03-27T15:31:41.421694900Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "fake_audio_features_2 = {'spectral_contrast_mean (mix)': 20.1520629356514,\n",
+    " 'chroma_mean (mix)': 0.408170998096466,\n",
+    " 'melody_variability (vocals)': 0.3748557548841115,\n",
+    " 'rhythm_onset_rate (mix)': 2.0277777777777777,\n",
+    " 'spectral_centroid_mean custom (mix)': 0.0938994687064515,\n",
+    " 'repetition_score custom (mix)': 0.0078822056596288,\n",
+    " 'pitch_std (mix)': 108.27875518798828,\n",
+    " 'tempo_bpm_corrected (mix)': 104.60529327392578,\n",
+    " 'tempo_bpm_original (mix)': 104.60529327392578,\n",
+    " 'mfcc_mean_1 (mix)': 163.30865478515625,\n",
+    " 'mfcc_mean_2 (mix)': 22.928050994873047,\n",
+    " 'loudness_range_lu custom (mix)': 7.131200790405273,\n",
+    " 'zero_crossing_rate (mix)': 0.02693298086524,\n",
+    " 'vocab_richness': 0.582,\n",
+    " 'loudness_integrated_lufs custom (mix)': -18.566131591796875,\n",
+    " 'readability_score': 27.8,\n",
+    " 'energy_essentia (mix)': 0.1537376961168162,\n",
+    " 'energy_librosa (mix)': 0.335069430512918,\n",
+    " 'rms_energy_mean (mix)': 0.3462888819034169,\n",
+    " 'melody_complexity (vocals)': 1.9166666666666667}\n",
+    "answer = manual_predict(fake_audio_features_2)\n",
+    "print(answer)"
+   ],
+   "id": "b998289b1d156dc2",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Black Metal, Death Metal, Doom Metal, Heavy Metal, Metalcore, Thrash Metal\n"
+     ]
+    }
+   ],
+   "execution_count": 62
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T15:24:02.082478Z",
+     "start_time": "2026-03-27T15:24:01.848007Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "fake_metal_audio = {\n",
+    "    'spectral_contrast_mean (mix)': 24.85,\n",
+    "    'chroma_mean (mix)': 0.38,\n",
+    "    'melody_variability (vocals)': 0.72,\n",
+    "    'rhythm_onset_rate (mix)': 4.95,\n",
+    "    'spectral_centroid_mean custom (mix)': 0.29,\n",
+    "    'repetition_score custom (mix)': 0.006,\n",
+    "    'pitch_std (mix)': 210.45,\n",
+    "    'tempo_bpm_corrected (mix)': 145.60,\n",
+    "    'tempo_bpm_original (mix)': 147.20,\n",
+    "    'mfcc_mean_1 (mix)': 142.75,\n",
+    "    'mfcc_mean_2 (mix)': -18.34,\n",
+    "    'loudness_range_lu custom (mix)': 6.25,\n",
+    "    'zero_crossing_rate (mix)': 0.18,\n",
+    "    'vocab_richness': 0.27,\n",
+    "    'loudness_integrated_lufs custom (mix)': -8.95,\n",
+    "    'readability_score': 42.0,\n",
+    "    'energy_essentia (mix)': 0.78,\n",
+    "    'energy_librosa (mix)': 0.82,\n",
+    "    'rms_energy_mean (mix)': 0.80,\n",
+    "    'melody_complexity (vocals)': 3.95\n",
+    "}\n",
+    "\n",
+    "answer = manual_predict(fake_metal_audio)\n",
+    "print(answer)"
+   ],
+   "id": "b265ee987709c2a3",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Rock -- Americana\n"
+     ]
+    }
+   ],
+   "execution_count": 57
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T15:38:00.278309800Z",
+     "start_time": "2026-03-27T15:37:57.734946600Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "",
+   "id": "228a47b066312e75",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "✅ Model saved as genre_model.pkl\n"
+     ]
+    }
+   ],
+   "execution_count": 64
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T15:49:22.832888800Z",
+     "start_time": "2026-03-27T15:49:20.228927800Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "import joblib\n",
+    "bundle = {\n",
+    "    \"model\": model,\n",
+    "    \"scaler\": scaler,\n",
+    "    \"mlb\": mlb,\n",
+    "    \"feature_cols\": X.columns.tolist()\n",
+    "}\n",
+    "joblib.dump(bundle, \"genre_model1.pkl\")\n",
+    "print(\"Model saved successfully!\")"
+   ],
+   "id": "694614771e1c62a9",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model saved successfully!\n"
+     ]
+    }
+   ],
+   "execution_count": 68
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T15:49:54.380812800Z",
+     "start_time": "2026-03-27T15:49:51.308892Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "import joblib\n",
+    "bundle = joblib.load(\"genre_model1.pkl\")\n",
+    "model = bundle[\"model\"]\n",
+    "scaler = bundle[\"scaler\"]\n",
+    "mlb = bundle[\"mlb\"]\n",
+    "feature_cols = bundle[\"feature_cols\"]"
+   ],
+   "id": "bb9f24b4e0947c22",
+   "outputs": [],
+   "execution_count": 70
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T15:51:05.220558400Z",
+     "start_time": "2026-03-27T15:51:05.157646Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "def predict_from_loaded(row, model, scaler, mlb, feature_cols):\n",
+    "    import pandas as pd\n",
+    "    import numpy as np\n",
+    "    input_df = pd.DataFrame([row])\n",
+    "    input_numeric = input_df.select_dtypes(include=[np.number]).fillna(0)\n",
+    "    for col in feature_cols:\n",
+    "        if col not in input_numeric.columns:\n",
+    "            input_numeric[col] = 0\n",
+    "    input_numeric = input_numeric[feature_cols]\n",
+    "    input_scaled = scaler.transform(input_numeric)\n",
+    "    probas = model.predict_proba(input_scaled)\n",
+    "    threshold = 0.20\n",
+    "    pred_binary = np.zeros((1, len(probas)), dtype=int)\n",
+    "    for i in range(len(probas)):\n",
+    "        if probas[i][0][1] > threshold:\n",
+    "            pred_binary[0, i] = 1\n",
+    "    subgenres = mlb.inverse_transform(pred_binary)[0]\n",
+    "    if len(subgenres) == 0:\n",
+    "        return \"None\"\n",
+    "    return \", \".join(sorted(subgenres))"
+   ],
+   "id": "5391f7697e93bcad",
+   "outputs": [],
+   "execution_count": 71
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-27T15:51:42.665671400Z",
+     "start_time": "2026-03-27T15:51:42.509361Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "audio_features_ = {'spectral_contrast_mean (mix)': 20.1520629356514,\n",
+    " 'chroma_mean (mix)': 0.408170998096466,\n",
+    " 'melody_variability (vocals)': 0.3748557548841115,\n",
+    " 'rhythm_onset_rate (mix)': 2.0277777777777777,\n",
+    " 'spectral_centroid_mean custom (mix)': 0.0938994687064515,\n",
+    " 'repetition_score custom (mix)': 0.0078822056596288,\n",
+    " 'pitch_std (mix)': 108.27875518798828,\n",
+    " 'tempo_bpm_corrected (mix)': 104.60529327392578,\n",
+    " 'tempo_bpm_original (mix)': 104.60529327392578,\n",
+    " 'mfcc_mean_1 (mix)': 163.30865478515625,\n",
+    " 'mfcc_mean_2 (mix)': 22.928050994873047,\n",
+    " 'loudness_range_lu custom (mix)': 7.131200790405273,\n",
+    " 'zero_crossing_rate (mix)': 0.02693298086524,\n",
+    " 'vocab_richness': 0.582,\n",
+    " 'loudness_integrated_lufs custom (mix)': -18.566131591796875,\n",
+    " 'readability_score': 27.8,\n",
+    " 'energy_essentia (mix)': 0.1537376961168162,\n",
+    " 'energy_librosa (mix)': 0.335069430512918,\n",
+    " 'rms_energy_mean (mix)': 0.3462888819034169,\n",
+    " 'melody_complexity (vocals)': 1.9166666666666667}\n",
+    "\n",
+    "result = predict_from_loaded(audio_features_, model, scaler, mlb, feature_cols)\n",
+    "print(result)"
+   ],
+   "id": "cbc80439273b814f",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Black Metal, Death Metal, Doom Metal, Heavy Metal, Metalcore, Thrash Metal\n"
+     ]
+    }
+   ],
+   "execution_count": 73
+  },
+  {
+   "metadata": {},
+   "cell_type": "markdown",
+   "source": "### 1) All Data",
+   "id": "9a576cffaecfd507"
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-30T09:59:36.204702500Z",
+     "start_time": "2026-03-30T09:59:34.858890200Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "import pandas as pd\n",
+    "path = r\"C:\\Users\\VigneshSubramani\\Downloads\\features_dataset.csv\"\n",
+    "pd.set_option(\"display.max_columns\", None)\n",
+    "data = pd.read_csv(path,low_memory=False)\n",
+    "data"
+   ],
+   "id": "16dcec42226f1c5d",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "                                                track_url        genre  \\\n",
+       "0       https://open.spotify.com/track/38RJG8nLAwGycit...      Country   \n",
+       "1       https://open.spotify.com/track/2uqYupMHANxnwge...      Country   \n",
+       "2       https://open.spotify.com/track/48X5k2vce5rXckg...      Country   \n",
+       "3       https://open.spotify.com/track/3SaS0Va6o6jYkWv...      Country   \n",
+       "4       https://open.spotify.com/track/09hVlIVHtEE1nLU...      Country   \n",
+       "...                                                   ...          ...   \n",
+       "138685  https://open.spotify.com/track/6rZno3nqeT7hv2P...  Latin Urban   \n",
+       "138686  https://open.spotify.com/track/0O4ZJKR2QDLGrBZ...  Latin Urban   \n",
+       "138687  https://open.spotify.com/track/4oV82xuYkYgyM2n...  Latin Urban   \n",
+       "138688  https://open.spotify.com/track/3jztoVZMBEaTX0b...  Latin Urban   \n",
+       "138689  https://open.spotify.com/track/3mUr9hl2VhqorCu...  Latin Urban   \n",
+       "\n",
+       "                       name  features.melody_complexity (vocals)  \\\n",
+       "0             Forever To Me                             3.500000   \n",
+       "1                    Austin                             3.250000   \n",
+       "2                Dirt Cheap                             2.750000   \n",
+       "3             I Am Not Okay                             2.333333   \n",
+       "4              Burn It Down                             3.166667   \n",
+       "...                     ...                                  ...   \n",
+       "138685            YOGURCITO                                  NaN   \n",
+       "138686  Me Arriesgo Contigo                                  NaN   \n",
+       "138687              Vocales                                  NaN   \n",
+       "138688               Tímida                                  NaN   \n",
+       "138689      Romeo Y Julieta                                  NaN   \n",
+       "\n",
+       "        features.melody_range (vocals)  features.melody_variability (vocals)  \\\n",
+       "0                                 41.0                              0.751713   \n",
+       "1                                 35.0                              0.715152   \n",
+       "2                                 34.0                              0.453325   \n",
+       "3                                 34.0                              0.436993   \n",
+       "4                                 38.0                              0.531330   \n",
+       "...                                ...                                   ...   \n",
+       "138685                             NaN                                   NaN   \n",
+       "138686                             NaN                                   NaN   \n",
+       "138687                             NaN                                   NaN   \n",
+       "138688                             NaN                                   NaN   \n",
+       "138689                             NaN                                   NaN   \n",
+       "\n",
+       "        features.tempo_bpm_original (mix)  features.danceability custom (mix)  \\\n",
+       "0                              141.384491                            0.580731   \n",
+       "1                              105.666534                            0.747039   \n",
+       "2                              110.124130                            0.694952   \n",
+       "3                              141.088547                            0.602134   \n",
+       "4                              148.782501                            0.603468   \n",
+       "...                                   ...                                 ...   \n",
+       "138685                                NaN                                 NaN   \n",
+       "138686                                NaN                                 NaN   \n",
+       "138687                                NaN                                 NaN   \n",
+       "138688                                NaN                                 NaN   \n",
+       "138689                                NaN                                 NaN   \n",
+       "\n",
+       "        features.loudness_integrated_lufs custom (mix)  \\\n",
+       "0                                           -15.467463   \n",
+       "1                                           -12.137867   \n",
+       "2                                           -14.214710   \n",
+       "3                                           -12.991043   \n",
+       "4                                           -12.433587   \n",
+       "...                                                ...   \n",
+       "138685                                             NaN   \n",
+       "138686                                             NaN   \n",
+       "138687                                             NaN   \n",
+       "138688                                             NaN   \n",
+       "138689                                             NaN   \n",
+       "\n",
+       "        features.loudness_range_lu custom (mix)  \\\n",
+       "0                                     11.282878   \n",
+       "1                                      3.665868   \n",
+       "2                                      8.500771   \n",
+       "3                                      3.841528   \n",
+       "4                                      5.801809   \n",
+       "...                                         ...   \n",
+       "138685                                      NaN   \n",
+       "138686                                      NaN   \n",
+       "138687                                      NaN   \n",
+       "138688                                      NaN   \n",
+       "138689                                      NaN   \n",
+       "\n",
+       "        features.energy_librosa (mix)  features.energy_librosa_std (mix)  \\\n",
+       "0                            0.419014                           0.191760   \n",
+       "1                            0.469624                           0.169444   \n",
+       "2                            0.427623                           0.194988   \n",
+       "3                            0.524690                           0.185032   \n",
+       "4                            0.566157                           0.145061   \n",
+       "...                               ...                                ...   \n",
+       "138685                            NaN                                NaN   \n",
+       "138686                            NaN                                NaN   \n",
+       "138687                            NaN                                NaN   \n",
+       "138688                            NaN                                NaN   \n",
+       "138689                            NaN                                NaN   \n",
+       "\n",
+       "        features.energy_essentia (mix)  features.energy_essentia_std (mix)  \\\n",
+       "0                             0.225257                            0.167297   \n",
+       "1                             0.255954                            0.162414   \n",
+       "2                             0.226500                            0.159222   \n",
+       "3                             0.321317                            0.172458   \n",
+       "4                             0.355688                            0.138871   \n",
+       "...                                ...                                 ...   \n",
+       "138685                             NaN                                 NaN   \n",
+       "138686                             NaN                                 NaN   \n",
+       "138687                             NaN                                 NaN   \n",
+       "138688                             NaN                                 NaN   \n",
+       "138689                             NaN                                 NaN   \n",
+       "\n",
+       "        features.energy_combined (mix)  \\\n",
+       "0                             0.322135   \n",
+       "1                             0.362789   \n",
+       "2                             0.327062   \n",
+       "3                             0.423004   \n",
+       "4                             0.460922   \n",
+       "...                                ...   \n",
+       "138685                             NaN   \n",
+       "138686                             NaN   \n",
+       "138687                             NaN   \n",
+       "138688                             NaN   \n",
+       "138689                             NaN   \n",
+       "\n",
+       "        features.spectral_centroid_mean custom (mix)  \\\n",
+       "0                                           0.128541   \n",
+       "1                                           0.121481   \n",
+       "2                                           0.120138   \n",
+       "3                                           0.140651   \n",
+       "4                                           0.127909   \n",
+       "...                                              ...   \n",
+       "138685                                           NaN   \n",
+       "138686                                           NaN   \n",
+       "138687                                           NaN   \n",
+       "138688                                           NaN   \n",
+       "138689                                           NaN   \n",
+       "\n",
+       "        features.mfcc_mean_1 (mix)  features.mfcc_mean_2 (mix)  \\\n",
+       "0                       144.864594                   -7.119251   \n",
+       "1                       160.318756                   -3.738415   \n",
+       "2                       156.815781                  -13.236840   \n",
+       "3                       143.763428                  -21.374804   \n",
+       "4                       156.998581                  -24.430223   \n",
+       "...                            ...                         ...   \n",
+       "138685                         NaN                         NaN   \n",
+       "138686                         NaN                         NaN   \n",
+       "138687                         NaN                         NaN   \n",
+       "138688                         NaN                         NaN   \n",
+       "138689                         NaN                         NaN   \n",
+       "\n",
+       "        features.chroma_mean (mix)  features.spectral_contrast_mean (mix)  \\\n",
+       "0                         0.482208                              19.678119   \n",
+       "1                         0.501811                              19.144755   \n",
+       "2                         0.452644                              19.641929   \n",
+       "3                         0.434028                              19.570383   \n",
+       "4                         0.493009                              19.436073   \n",
+       "...                            ...                                    ...   \n",
+       "138685                         NaN                                    NaN   \n",
+       "138686                         NaN                                    NaN   \n",
+       "138687                         NaN                                    NaN   \n",
+       "138688                         NaN                                    NaN   \n",
+       "138689                         NaN                                    NaN   \n",
+       "\n",
+       "        features.repetition_score custom (mix)  features.pitch_mean (mix)  \\\n",
+       "0                                     0.007355                 188.741852   \n",
+       "1                                     0.007224                 225.259796   \n",
+       "2                                     0.006500                 184.371185   \n",
+       "3                                     0.006950                 232.608765   \n",
+       "4                                     0.005922                 241.072388   \n",
+       "...                                        ...                        ...   \n",
+       "138685                                     NaN                        NaN   \n",
+       "138686                                     NaN                        NaN   \n",
+       "138687                                     NaN                        NaN   \n",
+       "138688                                     NaN                        NaN   \n",
+       "138689                                     NaN                        NaN   \n",
+       "\n",
+       "        features.pitch_std (mix)  features.rms_energy_mean (mix)  \\\n",
+       "0                      54.632015                        0.432170   \n",
+       "1                      90.800995                        0.477567   \n",
+       "2                      49.438427                        0.436157   \n",
+       "3                      60.514282                        0.537289   \n",
+       "4                      98.557098                        0.579285   \n",
+       "...                          ...                             ...   \n",
+       "138685                       NaN                             NaN   \n",
+       "138686                       NaN                             NaN   \n",
+       "138687                       NaN                             NaN   \n",
+       "138688                       NaN                             NaN   \n",
+       "138689                       NaN                             NaN   \n",
+       "\n",
+       "        features.rms_energy_std (mix)  features.zero_crossing_rate (mix)  \\\n",
+       "0                            0.194504                           0.042678   \n",
+       "1                            0.166947                           0.040959   \n",
+       "2                            0.192781                           0.046456   \n",
+       "3                            0.180169                           0.056903   \n",
+       "4                            0.140620                           0.053897   \n",
+       "...                               ...                                ...   \n",
+       "138685                            NaN                                NaN   \n",
+       "138686                            NaN                                NaN   \n",
+       "138687                            NaN                                NaN   \n",
+       "138688                            NaN                                NaN   \n",
+       "138689                            NaN                                NaN   \n",
+       "\n",
+       "                  sub_genre[0]          sub_genre[1]  sub_genre[2]  \\\n",
+       "0         Contemporary Country                   NaN           NaN   \n",
+       "1                          NaN                   NaN           NaN   \n",
+       "2       Modern Neo-Traditional  Contemporary Country           NaN   \n",
+       "3                    Americana          Country Rock  Country Folk   \n",
+       "4                          NaN                   NaN           NaN   \n",
+       "...                        ...                   ...           ...   \n",
+       "138685             Latin Urban                   NaN           NaN   \n",
+       "138686             Latin Urban                   NaN           NaN   \n",
+       "138687             Latin Urban                   NaN           NaN   \n",
+       "138688             Latin Urban                   NaN           NaN   \n",
+       "138689             Latin Urban                   NaN           NaN   \n",
+       "\n",
+       "                sub_genre[3] sub_genre[4] sub_genre[5] sub_genre[6]  \\\n",
+       "0                        NaN          NaN          NaN          NaN   \n",
+       "1                        NaN          NaN          NaN          NaN   \n",
+       "2                        NaN          NaN          NaN          NaN   \n",
+       "3       Contemporary Country          NaN          NaN          NaN   \n",
+       "4                        NaN          NaN          NaN          NaN   \n",
+       "...                      ...          ...          ...          ...   \n",
+       "138685                   NaN          NaN          NaN          NaN   \n",
+       "138686                   NaN          NaN          NaN          NaN   \n",
+       "138687                   NaN          NaN          NaN          NaN   \n",
+       "138688                   NaN          NaN          NaN          NaN   \n",
+       "138689                   NaN          NaN          NaN          NaN   \n",
+       "\n",
+       "       sub_genre[7] sub_genre[8] sub_genre[9] sub_genre[10] sub_genre[11]  \\\n",
+       "0               NaN          NaN          NaN           NaN           NaN   \n",
+       "1               NaN          NaN          NaN           NaN           NaN   \n",
+       "2               NaN          NaN          NaN           NaN           NaN   \n",
+       "3               NaN          NaN          NaN           NaN           NaN   \n",
+       "4               NaN          NaN          NaN           NaN           NaN   \n",
+       "...             ...          ...          ...           ...           ...   \n",
+       "138685          NaN          NaN          NaN           NaN           NaN   \n",
+       "138686          NaN          NaN          NaN           NaN           NaN   \n",
+       "138687          NaN          NaN          NaN           NaN           NaN   \n",
+       "138688          NaN          NaN          NaN           NaN           NaN   \n",
+       "138689          NaN          NaN          NaN           NaN           NaN   \n",
+       "\n",
+       "       sub_genre[12] sub_genre[13] sub_genre[14] sub_genre[15]  \n",
+       "0                NaN           NaN           NaN           NaN  \n",
+       "1                NaN           NaN           NaN           NaN  \n",
+       "2                NaN           NaN           NaN           NaN  \n",
+       "3                NaN           NaN           NaN           NaN  \n",
+       "4                NaN           NaN           NaN           NaN  \n",
+       "...              ...           ...           ...           ...  \n",
+       "138685           NaN           NaN           NaN           NaN  \n",
+       "138686           NaN           NaN           NaN           NaN  \n",
+       "138687           NaN           NaN           NaN           NaN  \n",
+       "138688           NaN           NaN           NaN           NaN  \n",
+       "138689           NaN           NaN           NaN           NaN  \n",
+       "\n",
+       "[138690 rows x 42 columns]"
+      ],
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>track_url</th>\n",
+       "      <th>genre</th>\n",
+       "      <th>name</th>\n",
+       "      <th>features.melody_complexity (vocals)</th>\n",
+       "      <th>features.melody_range (vocals)</th>\n",
+       "      <th>features.melody_variability (vocals)</th>\n",
+       "      <th>features.tempo_bpm_original (mix)</th>\n",
+       "      <th>features.danceability custom (mix)</th>\n",
+       "      <th>features.loudness_integrated_lufs custom (mix)</th>\n",
+       "      <th>features.loudness_range_lu custom (mix)</th>\n",
+       "      <th>features.energy_librosa (mix)</th>\n",
+       "      <th>features.energy_librosa_std (mix)</th>\n",
+       "      <th>features.energy_essentia (mix)</th>\n",
+       "      <th>features.energy_essentia_std (mix)</th>\n",
+       "      <th>features.energy_combined (mix)</th>\n",
+       "      <th>features.spectral_centroid_mean custom (mix)</th>\n",
+       "      <th>features.mfcc_mean_1 (mix)</th>\n",
+       "      <th>features.mfcc_mean_2 (mix)</th>\n",
+       "      <th>features.chroma_mean (mix)</th>\n",
+       "      <th>features.spectral_contrast_mean (mix)</th>\n",
+       "      <th>features.repetition_score custom (mix)</th>\n",
+       "      <th>features.pitch_mean (mix)</th>\n",
+       "      <th>features.pitch_std (mix)</th>\n",
+       "      <th>features.rms_energy_mean (mix)</th>\n",
+       "      <th>features.rms_energy_std (mix)</th>\n",
+       "      <th>features.zero_crossing_rate (mix)</th>\n",
+       "      <th>sub_genre[0]</th>\n",
+       "      <th>sub_genre[1]</th>\n",
+       "      <th>sub_genre[2]</th>\n",
+       "      <th>sub_genre[3]</th>\n",
+       "      <th>sub_genre[4]</th>\n",
+       "      <th>sub_genre[5]</th>\n",
+       "      <th>sub_genre[6]</th>\n",
+       "      <th>sub_genre[7]</th>\n",
+       "      <th>sub_genre[8]</th>\n",
+       "      <th>sub_genre[9]</th>\n",
+       "      <th>sub_genre[10]</th>\n",
+       "      <th>sub_genre[11]</th>\n",
+       "      <th>sub_genre[12]</th>\n",
+       "      <th>sub_genre[13]</th>\n",
+       "      <th>sub_genre[14]</th>\n",
+       "      <th>sub_genre[15]</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>https://open.spotify.com/track/38RJG8nLAwGycit...</td>\n",
+       "      <td>Country</td>\n",
+       "      <td>Forever To Me</td>\n",
+       "      <td>3.500000</td>\n",
+       "      <td>41.0</td>\n",
+       "      <td>0.751713</td>\n",
+       "      <td>141.384491</td>\n",
+       "      <td>0.580731</td>\n",
+       "      <td>-15.467463</td>\n",
+       "      <td>11.282878</td>\n",
+       "      <td>0.419014</td>\n",
+       "      <td>0.191760</td>\n",
+       "      <td>0.225257</td>\n",
+       "      <td>0.167297</td>\n",
+       "      <td>0.322135</td>\n",
+       "      <td>0.128541</td>\n",
+       "      <td>144.864594</td>\n",
+       "      <td>-7.119251</td>\n",
+       "      <td>0.482208</td>\n",
+       "      <td>19.678119</td>\n",
+       "      <td>0.007355</td>\n",
+       "      <td>188.741852</td>\n",
+       "      <td>54.632015</td>\n",
+       "      <td>0.432170</td>\n",
+       "      <td>0.194504</td>\n",
+       "      <td>0.042678</td>\n",
+       "      <td>Contemporary Country</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>https://open.spotify.com/track/2uqYupMHANxnwge...</td>\n",
+       "      <td>Country</td>\n",
+       "      <td>Austin</td>\n",
+       "      <td>3.250000</td>\n",
+       "      <td>35.0</td>\n",
+       "      <td>0.715152</td>\n",
+       "      <td>105.666534</td>\n",
+       "      <td>0.747039</td>\n",
+       "      <td>-12.137867</td>\n",
+       "      <td>3.665868</td>\n",
+       "      <td>0.469624</td>\n",
+       "      <td>0.169444</td>\n",
+       "      <td>0.255954</td>\n",
+       "      <td>0.162414</td>\n",
+       "      <td>0.362789</td>\n",
+       "      <td>0.121481</td>\n",
+       "      <td>160.318756</td>\n",
+       "      <td>-3.738415</td>\n",
+       "      <td>0.501811</td>\n",
+       "      <td>19.144755</td>\n",
+       "      <td>0.007224</td>\n",
+       "      <td>225.259796</td>\n",
+       "      <td>90.800995</td>\n",
+       "      <td>0.477567</td>\n",
+       "      <td>0.166947</td>\n",
+       "      <td>0.040959</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>https://open.spotify.com/track/48X5k2vce5rXckg...</td>\n",
+       "      <td>Country</td>\n",
+       "      <td>Dirt Cheap</td>\n",
+       "      <td>2.750000</td>\n",
+       "      <td>34.0</td>\n",
+       "      <td>0.453325</td>\n",
+       "      <td>110.124130</td>\n",
+       "      <td>0.694952</td>\n",
+       "      <td>-14.214710</td>\n",
+       "      <td>8.500771</td>\n",
+       "      <td>0.427623</td>\n",
+       "      <td>0.194988</td>\n",
+       "      <td>0.226500</td>\n",
+       "      <td>0.159222</td>\n",
+       "      <td>0.327062</td>\n",
+       "      <td>0.120138</td>\n",
+       "      <td>156.815781</td>\n",
+       "      <td>-13.236840</td>\n",
+       "      <td>0.452644</td>\n",
+       "      <td>19.641929</td>\n",
+       "      <td>0.006500</td>\n",
+       "      <td>184.371185</td>\n",
+       "      <td>49.438427</td>\n",
+       "      <td>0.436157</td>\n",
+       "      <td>0.192781</td>\n",
+       "      <td>0.046456</td>\n",
+       "      <td>Modern Neo-Traditional</td>\n",
+       "      <td>Contemporary Country</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>https://open.spotify.com/track/3SaS0Va6o6jYkWv...</td>\n",
+       "      <td>Country</td>\n",
+       "      <td>I Am Not Okay</td>\n",
+       "      <td>2.333333</td>\n",
+       "      <td>34.0</td>\n",
+       "      <td>0.436993</td>\n",
+       "      <td>141.088547</td>\n",
+       "      <td>0.602134</td>\n",
+       "      <td>-12.991043</td>\n",
+       "      <td>3.841528</td>\n",
+       "      <td>0.524690</td>\n",
+       "      <td>0.185032</td>\n",
+       "      <td>0.321317</td>\n",
+       "      <td>0.172458</td>\n",
+       "      <td>0.423004</td>\n",
+       "      <td>0.140651</td>\n",
+       "      <td>143.763428</td>\n",
+       "      <td>-21.374804</td>\n",
+       "      <td>0.434028</td>\n",
+       "      <td>19.570383</td>\n",
+       "      <td>0.006950</td>\n",
+       "      <td>232.608765</td>\n",
+       "      <td>60.514282</td>\n",
+       "      <td>0.537289</td>\n",
+       "      <td>0.180169</td>\n",
+       "      <td>0.056903</td>\n",
+       "      <td>Americana</td>\n",
+       "      <td>Country Rock</td>\n",
+       "      <td>Country Folk</td>\n",
+       "      <td>Contemporary Country</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>https://open.spotify.com/track/09hVlIVHtEE1nLU...</td>\n",
+       "      <td>Country</td>\n",
+       "      <td>Burn It Down</td>\n",
+       "      <td>3.166667</td>\n",
+       "      <td>38.0</td>\n",
+       "      <td>0.531330</td>\n",
+       "      <td>148.782501</td>\n",
+       "      <td>0.603468</td>\n",
+       "      <td>-12.433587</td>\n",
+       "      <td>5.801809</td>\n",
+       "      <td>0.566157</td>\n",
+       "      <td>0.145061</td>\n",
+       "      <td>0.355688</td>\n",
+       "      <td>0.138871</td>\n",
+       "      <td>0.460922</td>\n",
+       "      <td>0.127909</td>\n",
+       "      <td>156.998581</td>\n",
+       "      <td>-24.430223</td>\n",
+       "      <td>0.493009</td>\n",
+       "      <td>19.436073</td>\n",
+       "      <td>0.005922</td>\n",
+       "      <td>241.072388</td>\n",
+       "      <td>98.557098</td>\n",
+       "      <td>0.579285</td>\n",
+       "      <td>0.140620</td>\n",
+       "      <td>0.053897</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>138685</th>\n",
+       "      <td>https://open.spotify.com/track/6rZno3nqeT7hv2P...</td>\n",
+       "      <td>Latin Urban</td>\n",
+       "      <td>YOGURCITO</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>Latin Urban</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>138686</th>\n",
+       "      <td>https://open.spotify.com/track/0O4ZJKR2QDLGrBZ...</td>\n",
+       "      <td>Latin Urban</td>\n",
+       "      <td>Me Arriesgo Contigo</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>Latin Urban</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>138687</th>\n",
+       "      <td>https://open.spotify.com/track/4oV82xuYkYgyM2n...</td>\n",
+       "      <td>Latin Urban</td>\n",
+       "      <td>Vocales</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>Latin Urban</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>138688</th>\n",
+       "      <td>https://open.spotify.com/track/3jztoVZMBEaTX0b...</td>\n",
+       "      <td>Latin Urban</td>\n",
+       "      <td>Tímida</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>Latin Urban</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>138689</th>\n",
+       "      <td>https://open.spotify.com/track/3mUr9hl2VhqorCu...</td>\n",
+       "      <td>Latin Urban</td>\n",
+       "      <td>Romeo Y Julieta</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>Latin Urban</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>138690 rows × 42 columns</p>\n",
+       "</div>"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 2
+  },
+  {
+   "metadata": {},
+   "cell_type": "code",
+   "outputs": [],
+   "execution_count": null,
+   "source": "",
+   "id": "a73b0a2127420c43"
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

genre_finder/test_some_codes/new_method.ipynb

@@ -0,0 +1,1523 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "id": "initial_id",
+   "metadata": {
+    "collapsed": true,
+    "ExecuteTime": {
+     "end_time": "2026-03-31T03:54:55.631220900Z",
+     "start_time": "2026-03-31T03:54:54.937932400Z"
+    }
+   },
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "pd.set_option(\"display.max_columns\", None)\n",
+    "df = pd.read_csv(\"assets/all_genre.csv\")\n",
+    "df.head()"
+   ],
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "                        _id        genre  \\\n",
+       "0  69143037d64595f86b812d77  Hip Hop Rap   \n",
+       "1  691448a64bef1dcbb1d3da1b  Hip Hop Rap   \n",
+       "2  691453c34bef1dcbb1d3de77  Hip Hop Rap   \n",
+       "3  6914609dfb8126cae4324718  Hip Hop Rap   \n",
+       "4  691af0f3979070eb7827937c  Hip Hop Rap   \n",
+       "\n",
+       "                                      genre_subgenre  \\\n",
+       "0  Hip Hop Rap --- ['Alternative Hip Hop', 'Boom ...   \n",
+       "1  Hip Hop Rap --- ['Americana', 'Contemporary Co...   \n",
+       "2  Hip Hop Rap --- ['Alternative Hip Hop', 'Boom ...   \n",
+       "3  Hip Hop Rap --- ['Alternative Hip Hop', 'Conte...   \n",
+       "4  Hip Hop Rap --- ['Alternative Hip Hop', 'Conte...   \n",
+       "\n",
+       "   trimmed_audio_duration_sec  syllable_count  word_count  \\\n",
+       "0                  183.843991           355.0       298.0   \n",
+       "1                  160.786576           285.0       236.0   \n",
+       "2                  202.320000           350.0       285.0   \n",
+       "3                  187.298798           318.0       285.0   \n",
+       "4                  185.004989           651.0       551.0   \n",
+       "\n",
+       "   spectral_contrast_mean (mix)  chroma_mean (mix)  \\\n",
+       "0                     20.440557           0.464204   \n",
+       "1                     19.657228           0.424237   \n",
+       "2                     20.115561           0.454093   \n",
+       "3                     19.659283           0.443467   \n",
+       "4                     20.855880           0.461875   \n",
+       "\n",
+       "   melody_variability (vocals)  rhythm_onset_rate (mix)  \\\n",
+       "0                     0.434021                 3.250000   \n",
+       "1                     0.351001                 2.425576   \n",
+       "2                     0.327000                 2.738889   \n",
+       "3                     0.360642                 3.500000   \n",
+       "4                     0.767493                 4.983333   \n",
+       "\n",
+       "   spectral_centroid_mean custom (mix)  repetition_score custom (mix)  \\\n",
+       "0                             0.132002                       0.006988   \n",
+       "1                             0.139757                       0.007367   \n",
+       "2                             0.111282                       0.006671   \n",
+       "3                             0.159247                       0.007202   \n",
+       "4                             0.155302                       0.007290   \n",
+       "\n",
+       "   pitch_std (mix)  tempo_bpm_corrected (mix)  tempo_bpm_original (mix)  \\\n",
+       "0        69.616264                 152.006775                152.006775   \n",
+       "1       156.358124                 117.943665                117.943665   \n",
+       "2        69.311798                 124.344086                124.344086   \n",
+       "3        65.514191                 109.909401                109.909401   \n",
+       "4       127.873909                 151.955536                151.955536   \n",
+       "\n",
+       "   mfcc_mean_1 (mix)  mfcc_mean_2 (mix)  loudness_range_lu custom (mix)  \\\n",
+       "0         153.058243          -7.368185                        1.143722   \n",
+       "1         148.055832         -49.435444                        4.708925   \n",
+       "2         166.170502          -5.852028                        1.948263   \n",
+       "3         134.211624         -27.524824                        1.244652   \n",
+       "4         129.783264         -11.940094                        3.452597   \n",
+       "\n",
+       "   zero_crossing_rate (mix)  vocab_richness  \\\n",
+       "0                  0.045159           0.305   \n",
+       "1                  0.070352           0.352   \n",
+       "2                  0.036463           0.400   \n",
+       "3                  0.064695           0.365   \n",
+       "4                  0.056507           0.281   \n",
+       "\n",
+       "   loudness_integrated_lufs custom (mix)  readability_score  \\\n",
+       "0                             -12.104649               20.9   \n",
+       "1                             -11.814197               89.4   \n",
+       "2                             -12.000179              107.0   \n",
+       "3                             -12.435019              105.8   \n",
+       "4                             -16.217091              212.3   \n",
+       "\n",
+       "   energy_essentia (mix)  energy_librosa (mix)  rms_energy_mean (mix)  \\\n",
+       "0               0.349783              0.576250               0.578454   \n",
+       "1               0.429792              0.619931               0.632507   \n",
+       "2               0.325114              0.520390               0.579053   \n",
+       "3               0.407647              0.617191               0.617197   \n",
+       "4               0.255311              0.446953               0.454712   \n",
+       "\n",
+       "   sentiment_score  melody_complexity (vocals)  avg_word_length  \n",
+       "0            0.116                    2.250000             3.76  \n",
+       "1            0.173                    1.916667             3.50  \n",
+       "2            0.158                    1.833333             3.73  \n",
+       "3            0.071                    1.916667             3.48  \n",
+       "4            0.039                    3.500000             3.56  "
+      ],
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>_id</th>\n",
+       "      <th>genre</th>\n",
+       "      <th>genre_subgenre</th>\n",
+       "      <th>trimmed_audio_duration_sec</th>\n",
+       "      <th>syllable_count</th>\n",
+       "      <th>word_count</th>\n",
+       "      <th>spectral_contrast_mean (mix)</th>\n",
+       "      <th>chroma_mean (mix)</th>\n",
+       "      <th>melody_variability (vocals)</th>\n",
+       "      <th>rhythm_onset_rate (mix)</th>\n",
+       "      <th>spectral_centroid_mean custom (mix)</th>\n",
+       "      <th>repetition_score custom (mix)</th>\n",
+       "      <th>pitch_std (mix)</th>\n",
+       "      <th>tempo_bpm_corrected (mix)</th>\n",
+       "      <th>tempo_bpm_original (mix)</th>\n",
+       "      <th>mfcc_mean_1 (mix)</th>\n",
+       "      <th>mfcc_mean_2 (mix)</th>\n",
+       "      <th>loudness_range_lu custom (mix)</th>\n",
+       "      <th>zero_crossing_rate (mix)</th>\n",
+       "      <th>vocab_richness</th>\n",
+       "      <th>loudness_integrated_lufs custom (mix)</th>\n",
+       "      <th>readability_score</th>\n",
+       "      <th>energy_essentia (mix)</th>\n",
+       "      <th>energy_librosa (mix)</th>\n",
+       "      <th>rms_energy_mean (mix)</th>\n",
+       "      <th>sentiment_score</th>\n",
+       "      <th>melody_complexity (vocals)</th>\n",
+       "      <th>avg_word_length</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>69143037d64595f86b812d77</td>\n",
+       "      <td>Hip Hop Rap</td>\n",
+       "      <td>Hip Hop Rap --- ['Alternative Hip Hop', 'Boom ...</td>\n",
+       "      <td>183.843991</td>\n",
+       "      <td>355.0</td>\n",
+       "      <td>298.0</td>\n",
+       "      <td>20.440557</td>\n",
+       "      <td>0.464204</td>\n",
+       "      <td>0.434021</td>\n",
+       "      <td>3.250000</td>\n",
+       "      <td>0.132002</td>\n",
+       "      <td>0.006988</td>\n",
+       "      <td>69.616264</td>\n",
+       "      <td>152.006775</td>\n",
+       "      <td>152.006775</td>\n",
+       "      <td>153.058243</td>\n",
+       "      <td>-7.368185</td>\n",
+       "      <td>1.143722</td>\n",
+       "      <td>0.045159</td>\n",
+       "      <td>0.305</td>\n",
+       "      <td>-12.104649</td>\n",
+       "      <td>20.9</td>\n",
+       "      <td>0.349783</td>\n",
+       "      <td>0.576250</td>\n",
+       "      <td>0.578454</td>\n",
+       "      <td>0.116</td>\n",
+       "      <td>2.250000</td>\n",
+       "      <td>3.76</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>691448a64bef1dcbb1d3da1b</td>\n",
+       "      <td>Hip Hop Rap</td>\n",
+       "      <td>Hip Hop Rap --- ['Americana', 'Contemporary Co...</td>\n",
+       "      <td>160.786576</td>\n",
+       "      <td>285.0</td>\n",
+       "      <td>236.0</td>\n",
+       "      <td>19.657228</td>\n",
+       "      <td>0.424237</td>\n",
+       "      <td>0.351001</td>\n",
+       "      <td>2.425576</td>\n",
+       "      <td>0.139757</td>\n",
+       "      <td>0.007367</td>\n",
+       "      <td>156.358124</td>\n",
+       "      <td>117.943665</td>\n",
+       "      <td>117.943665</td>\n",
+       "      <td>148.055832</td>\n",
+       "      <td>-49.435444</td>\n",
+       "      <td>4.708925</td>\n",
+       "      <td>0.070352</td>\n",
+       "      <td>0.352</td>\n",
+       "      <td>-11.814197</td>\n",
+       "      <td>89.4</td>\n",
+       "      <td>0.429792</td>\n",
+       "      <td>0.619931</td>\n",
+       "      <td>0.632507</td>\n",
+       "      <td>0.173</td>\n",
+       "      <td>1.916667</td>\n",
+       "      <td>3.50</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>691453c34bef1dcbb1d3de77</td>\n",
+       "      <td>Hip Hop Rap</td>\n",
+       "      <td>Hip Hop Rap --- ['Alternative Hip Hop', 'Boom ...</td>\n",
+       "      <td>202.320000</td>\n",
+       "      <td>350.0</td>\n",
+       "      <td>285.0</td>\n",
+       "      <td>20.115561</td>\n",
+       "      <td>0.454093</td>\n",
+       "      <td>0.327000</td>\n",
+       "      <td>2.738889</td>\n",
+       "      <td>0.111282</td>\n",
+       "      <td>0.006671</td>\n",
+       "      <td>69.311798</td>\n",
+       "      <td>124.344086</td>\n",
+       "      <td>124.344086</td>\n",
+       "      <td>166.170502</td>\n",
+       "      <td>-5.852028</td>\n",
+       "      <td>1.948263</td>\n",
+       "      <td>0.036463</td>\n",
+       "      <td>0.400</td>\n",
+       "      <td>-12.000179</td>\n",
+       "      <td>107.0</td>\n",
+       "      <td>0.325114</td>\n",
+       "      <td>0.520390</td>\n",
+       "      <td>0.579053</td>\n",
+       "      <td>0.158</td>\n",
+       "      <td>1.833333</td>\n",
+       "      <td>3.73</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>6914609dfb8126cae4324718</td>\n",
+       "      <td>Hip Hop Rap</td>\n",
+       "      <td>Hip Hop Rap --- ['Alternative Hip Hop', 'Conte...</td>\n",
+       "      <td>187.298798</td>\n",
+       "      <td>318.0</td>\n",
+       "      <td>285.0</td>\n",
+       "      <td>19.659283</td>\n",
+       "      <td>0.443467</td>\n",
+       "      <td>0.360642</td>\n",
+       "      <td>3.500000</td>\n",
+       "      <td>0.159247</td>\n",
+       "      <td>0.007202</td>\n",
+       "      <td>65.514191</td>\n",
+       "      <td>109.909401</td>\n",
+       "      <td>109.909401</td>\n",
+       "      <td>134.211624</td>\n",
+       "      <td>-27.524824</td>\n",
+       "      <td>1.244652</td>\n",
+       "      <td>0.064695</td>\n",
+       "      <td>0.365</td>\n",
+       "      <td>-12.435019</td>\n",
+       "      <td>105.8</td>\n",
+       "      <td>0.407647</td>\n",
+       "      <td>0.617191</td>\n",
+       "      <td>0.617197</td>\n",
+       "      <td>0.071</td>\n",
+       "      <td>1.916667</td>\n",
+       "      <td>3.48</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>691af0f3979070eb7827937c</td>\n",
+       "      <td>Hip Hop Rap</td>\n",
+       "      <td>Hip Hop Rap --- ['Alternative Hip Hop', 'Conte...</td>\n",
+       "      <td>185.004989</td>\n",
+       "      <td>651.0</td>\n",
+       "      <td>551.0</td>\n",
+       "      <td>20.855880</td>\n",
+       "      <td>0.461875</td>\n",
+       "      <td>0.767493</td>\n",
+       "      <td>4.983333</td>\n",
+       "      <td>0.155302</td>\n",
+       "      <td>0.007290</td>\n",
+       "      <td>127.873909</td>\n",
+       "      <td>151.955536</td>\n",
+       "      <td>151.955536</td>\n",
+       "      <td>129.783264</td>\n",
+       "      <td>-11.940094</td>\n",
+       "      <td>3.452597</td>\n",
+       "      <td>0.056507</td>\n",
+       "      <td>0.281</td>\n",
+       "      <td>-16.217091</td>\n",
+       "      <td>212.3</td>\n",
+       "      <td>0.255311</td>\n",
+       "      <td>0.446953</td>\n",
+       "      <td>0.454712</td>\n",
+       "      <td>0.039</td>\n",
+       "      <td>3.500000</td>\n",
+       "      <td>3.56</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 1
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T03:54:55.696551500Z",
+     "start_time": "2026-03-31T03:54:55.635800100Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df[\"genre\"].value_counts()",
+   "id": "9fe14275a70aaae0",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "genre\n",
+       "Rock                      29105\n",
+       "Hip Hop Rap               21428\n",
+       "EDM                       18903\n",
+       "Pop                       15967\n",
+       "RnB Soul                  14167\n",
+       "Country                   12499\n",
+       "Folk Singer Songwriter     6917\n",
+       "Alternative Indie          5274\n",
+       "Gospel Christian            578\n",
+       "Punk                        252\n",
+       "Metal                        79\n",
+       "Name: count, dtype: int64"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 2
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T03:54:55.725419Z",
+     "start_time": "2026-03-31T03:54:55.699556300Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df['genre_subgenre'][0]",
+   "id": "7412f3d028afc4e3",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "\"Hip Hop Rap --- ['Alternative Hip Hop', 'Boom Bap', 'Contemporary Hip Hop', 'Trap']\""
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 3
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T03:55:00.359660300Z",
+     "start_time": "2026-03-31T03:54:59.377188700Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "import ast\n",
+    "\n",
+    "def transform(x):\n",
+    "    main, sub = x.split(' --- ')\n",
+    "    sub_list = ast.literal_eval(sub)\n",
+    "    return sub_list + [main]\n",
+    "\n",
+    "df['genre_subgenre'] = df['genre_subgenre'].apply(transform)"
+   ],
+   "id": "928d1aa5fab137df",
+   "outputs": [],
+   "execution_count": 4
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T03:55:01.168965400Z",
+     "start_time": "2026-03-31T03:55:01.111013400Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "drop_cols = ['_id', 'genre','labels',\n",
+    "             'trimmed_audio_duration_sec', 'syllable_count', 'word_count',\n",
+    "             'sentiment_score', 'avg_word_length']\n",
+    "\n",
+    "feature_cols = [c for c in df.columns if c not in drop_cols]\n",
+    "df = df[feature_cols].copy()\n",
+    "df.shape"
+   ],
+   "id": "7b947270eb4fe3b0",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(125169, 21)"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 5
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T03:55:02.488063800Z",
+     "start_time": "2026-03-31T03:55:02.451762900Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df.head()",
+   "id": "1827b0bcaf4b3af3",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "                                      genre_subgenre  \\\n",
+       "0  [Alternative Hip Hop, Boom Bap, Contemporary H...   \n",
+       "1  [Americana, Contemporary Country, Bro Country,...   \n",
+       "2  [Alternative Hip Hop, Boom Bap, Contemporary H...   \n",
+       "3  [Alternative Hip Hop, Contemporary Hip Hop, Hi...   \n",
+       "4  [Alternative Hip Hop, Contemporary Hip Hop, Tr...   \n",
+       "\n",
+       "   spectral_contrast_mean (mix)  chroma_mean (mix)  \\\n",
+       "0                     20.440557           0.464204   \n",
+       "1                     19.657228           0.424237   \n",
+       "2                     20.115561           0.454093   \n",
+       "3                     19.659283           0.443467   \n",
+       "4                     20.855880           0.461875   \n",
+       "\n",
+       "   melody_variability (vocals)  rhythm_onset_rate (mix)  \\\n",
+       "0                     0.434021                 3.250000   \n",
+       "1                     0.351001                 2.425576   \n",
+       "2                     0.327000                 2.738889   \n",
+       "3                     0.360642                 3.500000   \n",
+       "4                     0.767493                 4.983333   \n",
+       "\n",
+       "   spectral_centroid_mean custom (mix)  repetition_score custom (mix)  \\\n",
+       "0                             0.132002                       0.006988   \n",
+       "1                             0.139757                       0.007367   \n",
+       "2                             0.111282                       0.006671   \n",
+       "3                             0.159247                       0.007202   \n",
+       "4                             0.155302                       0.007290   \n",
+       "\n",
+       "   pitch_std (mix)  tempo_bpm_corrected (mix)  tempo_bpm_original (mix)  \\\n",
+       "0        69.616264                 152.006775                152.006775   \n",
+       "1       156.358124                 117.943665                117.943665   \n",
+       "2        69.311798                 124.344086                124.344086   \n",
+       "3        65.514191                 109.909401                109.909401   \n",
+       "4       127.873909                 151.955536                151.955536   \n",
+       "\n",
+       "   mfcc_mean_1 (mix)  mfcc_mean_2 (mix)  loudness_range_lu custom (mix)  \\\n",
+       "0         153.058243          -7.368185                        1.143722   \n",
+       "1         148.055832         -49.435444                        4.708925   \n",
+       "2         166.170502          -5.852028                        1.948263   \n",
+       "3         134.211624         -27.524824                        1.244652   \n",
+       "4         129.783264         -11.940094                        3.452597   \n",
+       "\n",
+       "   zero_crossing_rate (mix)  vocab_richness  \\\n",
+       "0                  0.045159           0.305   \n",
+       "1                  0.070352           0.352   \n",
+       "2                  0.036463           0.400   \n",
+       "3                  0.064695           0.365   \n",
+       "4                  0.056507           0.281   \n",
+       "\n",
+       "   loudness_integrated_lufs custom (mix)  readability_score  \\\n",
+       "0                             -12.104649               20.9   \n",
+       "1                             -11.814197               89.4   \n",
+       "2                             -12.000179              107.0   \n",
+       "3                             -12.435019              105.8   \n",
+       "4                             -16.217091              212.3   \n",
+       "\n",
+       "   energy_essentia (mix)  energy_librosa (mix)  rms_energy_mean (mix)  \\\n",
+       "0               0.349783              0.576250               0.578454   \n",
+       "1               0.429792              0.619931               0.632507   \n",
+       "2               0.325114              0.520390               0.579053   \n",
+       "3               0.407647              0.617191               0.617197   \n",
+       "4               0.255311              0.446953               0.454712   \n",
+       "\n",
+       "   melody_complexity (vocals)  \n",
+       "0                    2.250000  \n",
+       "1                    1.916667  \n",
+       "2                    1.833333  \n",
+       "3                    1.916667  \n",
+       "4                    3.500000  "
+      ],
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>genre_subgenre</th>\n",
+       "      <th>spectral_contrast_mean (mix)</th>\n",
+       "      <th>chroma_mean (mix)</th>\n",
+       "      <th>melody_variability (vocals)</th>\n",
+       "      <th>rhythm_onset_rate (mix)</th>\n",
+       "      <th>spectral_centroid_mean custom (mix)</th>\n",
+       "      <th>repetition_score custom (mix)</th>\n",
+       "      <th>pitch_std (mix)</th>\n",
+       "      <th>tempo_bpm_corrected (mix)</th>\n",
+       "      <th>tempo_bpm_original (mix)</th>\n",
+       "      <th>mfcc_mean_1 (mix)</th>\n",
+       "      <th>mfcc_mean_2 (mix)</th>\n",
+       "      <th>loudness_range_lu custom (mix)</th>\n",
+       "      <th>zero_crossing_rate (mix)</th>\n",
+       "      <th>vocab_richness</th>\n",
+       "      <th>loudness_integrated_lufs custom (mix)</th>\n",
+       "      <th>readability_score</th>\n",
+       "      <th>energy_essentia (mix)</th>\n",
+       "      <th>energy_librosa (mix)</th>\n",
+       "      <th>rms_energy_mean (mix)</th>\n",
+       "      <th>melody_complexity (vocals)</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>[Alternative Hip Hop, Boom Bap, Contemporary H...</td>\n",
+       "      <td>20.440557</td>\n",
+       "      <td>0.464204</td>\n",
+       "      <td>0.434021</td>\n",
+       "      <td>3.250000</td>\n",
+       "      <td>0.132002</td>\n",
+       "      <td>0.006988</td>\n",
+       "      <td>69.616264</td>\n",
+       "      <td>152.006775</td>\n",
+       "      <td>152.006775</td>\n",
+       "      <td>153.058243</td>\n",
+       "      <td>-7.368185</td>\n",
+       "      <td>1.143722</td>\n",
+       "      <td>0.045159</td>\n",
+       "      <td>0.305</td>\n",
+       "      <td>-12.104649</td>\n",
+       "      <td>20.9</td>\n",
+       "      <td>0.349783</td>\n",
+       "      <td>0.576250</td>\n",
+       "      <td>0.578454</td>\n",
+       "      <td>2.250000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>[Americana, Contemporary Country, Bro Country,...</td>\n",
+       "      <td>19.657228</td>\n",
+       "      <td>0.424237</td>\n",
+       "      <td>0.351001</td>\n",
+       "      <td>2.425576</td>\n",
+       "      <td>0.139757</td>\n",
+       "      <td>0.007367</td>\n",
+       "      <td>156.358124</td>\n",
+       "      <td>117.943665</td>\n",
+       "      <td>117.943665</td>\n",
+       "      <td>148.055832</td>\n",
+       "      <td>-49.435444</td>\n",
+       "      <td>4.708925</td>\n",
+       "      <td>0.070352</td>\n",
+       "      <td>0.352</td>\n",
+       "      <td>-11.814197</td>\n",
+       "      <td>89.4</td>\n",
+       "      <td>0.429792</td>\n",
+       "      <td>0.619931</td>\n",
+       "      <td>0.632507</td>\n",
+       "      <td>1.916667</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>[Alternative Hip Hop, Boom Bap, Contemporary H...</td>\n",
+       "      <td>20.115561</td>\n",
+       "      <td>0.454093</td>\n",
+       "      <td>0.327000</td>\n",
+       "      <td>2.738889</td>\n",
+       "      <td>0.111282</td>\n",
+       "      <td>0.006671</td>\n",
+       "      <td>69.311798</td>\n",
+       "      <td>124.344086</td>\n",
+       "      <td>124.344086</td>\n",
+       "      <td>166.170502</td>\n",
+       "      <td>-5.852028</td>\n",
+       "      <td>1.948263</td>\n",
+       "      <td>0.036463</td>\n",
+       "      <td>0.400</td>\n",
+       "      <td>-12.000179</td>\n",
+       "      <td>107.0</td>\n",
+       "      <td>0.325114</td>\n",
+       "      <td>0.520390</td>\n",
+       "      <td>0.579053</td>\n",
+       "      <td>1.833333</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>[Alternative Hip Hop, Contemporary Hip Hop, Hi...</td>\n",
+       "      <td>19.659283</td>\n",
+       "      <td>0.443467</td>\n",
+       "      <td>0.360642</td>\n",
+       "      <td>3.500000</td>\n",
+       "      <td>0.159247</td>\n",
+       "      <td>0.007202</td>\n",
+       "      <td>65.514191</td>\n",
+       "      <td>109.909401</td>\n",
+       "      <td>109.909401</td>\n",
+       "      <td>134.211624</td>\n",
+       "      <td>-27.524824</td>\n",
+       "      <td>1.244652</td>\n",
+       "      <td>0.064695</td>\n",
+       "      <td>0.365</td>\n",
+       "      <td>-12.435019</td>\n",
+       "      <td>105.8</td>\n",
+       "      <td>0.407647</td>\n",
+       "      <td>0.617191</td>\n",
+       "      <td>0.617197</td>\n",
+       "      <td>1.916667</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>[Alternative Hip Hop, Contemporary Hip Hop, Tr...</td>\n",
+       "      <td>20.855880</td>\n",
+       "      <td>0.461875</td>\n",
+       "      <td>0.767493</td>\n",
+       "      <td>4.983333</td>\n",
+       "      <td>0.155302</td>\n",
+       "      <td>0.007290</td>\n",
+       "      <td>127.873909</td>\n",
+       "      <td>151.955536</td>\n",
+       "      <td>151.955536</td>\n",
+       "      <td>129.783264</td>\n",
+       "      <td>-11.940094</td>\n",
+       "      <td>3.452597</td>\n",
+       "      <td>0.056507</td>\n",
+       "      <td>0.281</td>\n",
+       "      <td>-16.217091</td>\n",
+       "      <td>212.3</td>\n",
+       "      <td>0.255311</td>\n",
+       "      <td>0.446953</td>\n",
+       "      <td>0.454712</td>\n",
+       "      <td>3.500000</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 6
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T03:55:06.547702800Z",
+     "start_time": "2026-03-31T03:55:04.612860200Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df.to_csv(\"final_clean_data.csv\", index=False)",
+   "id": "28af8793bd0eab61",
+   "outputs": [],
+   "execution_count": 7
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T03:57:01.018726Z",
+     "start_time": "2026-03-31T03:57:00.889301700Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df.isna().sum()",
+   "id": "8b904435409f0a49",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "genre_subgenre                           0\n",
+       "spectral_contrast_mean (mix)             0\n",
+       "chroma_mean (mix)                        0\n",
+       "melody_variability (vocals)              0\n",
+       "rhythm_onset_rate (mix)                  0\n",
+       "spectral_centroid_mean custom (mix)      0\n",
+       "repetition_score custom (mix)            0\n",
+       "pitch_std (mix)                          0\n",
+       "tempo_bpm_corrected (mix)                0\n",
+       "tempo_bpm_original (mix)                 0\n",
+       "mfcc_mean_1 (mix)                        0\n",
+       "mfcc_mean_2 (mix)                        0\n",
+       "loudness_range_lu custom (mix)           0\n",
+       "zero_crossing_rate (mix)                 0\n",
+       "vocab_richness                           0\n",
+       "loudness_integrated_lufs custom (mix)    0\n",
+       "readability_score                        0\n",
+       "energy_essentia (mix)                    0\n",
+       "energy_librosa (mix)                     0\n",
+       "rms_energy_mean (mix)                    0\n",
+       "melody_complexity (vocals)               0\n",
+       "dtype: int64"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 8
+  },
+  {
+   "metadata": {},
+   "cell_type": "markdown",
+   "source": "### 2) Load Dataset",
+   "id": "9168083c291938d"
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T03:57:05.398199900Z",
+     "start_time": "2026-03-31T03:57:04.817653900Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "df = pd.read_csv(\"final_clean_data.csv\", low_memory=False)\n",
+    "df.shape"
+   ],
+   "id": "61c97b293f69eec",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(125169, 21)"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 9
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T03:57:05.736114800Z",
+     "start_time": "2026-03-31T03:57:05.656718500Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df.head(10)",
+   "id": "1b84bca88fc20c5b",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "                                      genre_subgenre  \\\n",
+       "0  ['Alternative Hip Hop', 'Boom Bap', 'Contempor...   \n",
+       "1  ['Americana', 'Contemporary Country', 'Bro Cou...   \n",
+       "2  ['Alternative Hip Hop', 'Boom Bap', 'Contempor...   \n",
+       "3  ['Alternative Hip Hop', 'Contemporary Hip Hop'...   \n",
+       "4  ['Alternative Hip Hop', 'Contemporary Hip Hop'...   \n",
+       "5  ['Alternative Hip Hop', 'Contemporary Hip Hop'...   \n",
+       "6  ['Alternative Hip Hop', 'Boom Bap', 'Contempor...   \n",
+       "7  ['Alternative Hip Hop', 'Contemporary Hip Hop'...   \n",
+       "8  ['Alternative Hip Hop', 'Boom Bap', 'Contempor...   \n",
+       "9  ['Boom Bap', 'Contemporary Hip Hop', 'Trap', '...   \n",
+       "\n",
+       "   spectral_contrast_mean (mix)  chroma_mean (mix)  \\\n",
+       "0                     20.440557           0.464204   \n",
+       "1                     19.657228           0.424237   \n",
+       "2                     20.115561           0.454093   \n",
+       "3                     19.659283           0.443467   \n",
+       "4                     20.855880           0.461875   \n",
+       "5                     20.649601           0.582342   \n",
+       "6                     20.591029           0.531958   \n",
+       "7                     20.365473           0.527672   \n",
+       "8                     20.128469           0.526556   \n",
+       "9                     20.410229           0.432254   \n",
+       "\n",
+       "   melody_variability (vocals)  rhythm_onset_rate (mix)  \\\n",
+       "0                     0.434021                 3.250000   \n",
+       "1                     0.351001                 2.425576   \n",
+       "2                     0.327000                 2.738889   \n",
+       "3                     0.360642                 3.500000   \n",
+       "4                     0.767493                 4.983333   \n",
+       "5                     0.625612                 5.522135   \n",
+       "6                     0.700000                 4.647586   \n",
+       "7                     0.604867                 5.311232   \n",
+       "8                     0.802125                 4.563691   \n",
+       "9                     0.613747                 4.785156   \n",
+       "\n",
+       "   spectral_centroid_mean custom (mix)  repetition_score custom (mix)  \\\n",
+       "0                             0.132002                       0.006988   \n",
+       "1                             0.139757                       0.007367   \n",
+       "2                             0.111282                       0.006671   \n",
+       "3                             0.159247                       0.007202   \n",
+       "4                             0.155302                       0.007290   \n",
+       "5                             0.158728                       0.007571   \n",
+       "6                             0.144791                       0.008084   \n",
+       "7                             0.193616                       0.007258   \n",
+       "8                             0.158051                       0.007390   \n",
+       "9                             0.215118                       0.007961   \n",
+       "\n",
+       "   pitch_std (mix)  tempo_bpm_corrected (mix)  tempo_bpm_original (mix)  \\\n",
+       "0        69.616264                 152.006775                152.006775   \n",
+       "1       156.358124                 117.943665                117.943665   \n",
+       "2        69.311798                 124.344086                124.344086   \n",
+       "3        65.514191                 109.909401                109.909401   \n",
+       "4       127.873909                 151.955536                151.955536   \n",
+       "5        98.469620                 103.800598                103.800598   \n",
+       "6        96.689476                 133.976532                133.976532   \n",
+       "7       175.793594                 129.943954                129.943954   \n",
+       "8       143.052185                 150.016129                150.016129   \n",
+       "9       127.890625                 116.976936                116.976936   \n",
+       "\n",
+       "   mfcc_mean_1 (mix)  mfcc_mean_2 (mix)  loudness_range_lu custom (mix)  \\\n",
+       "0         153.058243          -7.368185                        1.143722   \n",
+       "1         148.055832         -49.435444                        4.708925   \n",
+       "2         166.170502          -5.852028                        1.948263   \n",
+       "3         134.211624         -27.524824                        1.244652   \n",
+       "4         129.783264         -11.940094                        3.452597   \n",
+       "5         115.804680         -15.190744                        4.185527   \n",
+       "6         128.920319          -6.285958                        5.836010   \n",
+       "7         110.702339          -5.517988                        8.970278   \n",
+       "8         122.700783         -31.691156                        4.626307   \n",
+       "9          94.988937         -20.226351                        4.125075   \n",
+       "\n",
+       "   zero_crossing_rate (mix)  vocab_richness  \\\n",
+       "0                  0.045159           0.305   \n",
+       "1                  0.070352           0.352   \n",
+       "2                  0.036463           0.400   \n",
+       "3                  0.064695           0.365   \n",
+       "4                  0.056507           0.281   \n",
+       "5                  0.049998           0.255   \n",
+       "6                  0.043094           0.308   \n",
+       "7                  0.078429           0.267   \n",
+       "8                  0.060048           0.261   \n",
+       "9                  0.111702           0.503   \n",
+       "\n",
+       "   loudness_integrated_lufs custom (mix)  readability_score  \\\n",
+       "0                             -12.104649               20.9   \n",
+       "1                             -11.814197               89.4   \n",
+       "2                             -12.000179              107.0   \n",
+       "3                             -12.435019              105.8   \n",
+       "4                             -16.217091              212.3   \n",
+       "5                             -16.153070               43.6   \n",
+       "6                             -17.065815               94.9   \n",
+       "7                             -18.897171               38.7   \n",
+       "8                             -15.986657                2.2   \n",
+       "9                             -13.382388               41.0   \n",
+       "\n",
+       "   energy_essentia (mix)  energy_librosa (mix)  rms_energy_mean (mix)  \\\n",
+       "0               0.349783              0.576250               0.578454   \n",
+       "1               0.429792              0.619931               0.632507   \n",
+       "2               0.325114              0.520390               0.579053   \n",
+       "3               0.407647              0.617191               0.617197   \n",
+       "4               0.255311              0.446953               0.454712   \n",
+       "5               0.310462              0.503279               0.503863   \n",
+       "6               0.299608              0.478562               0.484519   \n",
+       "7               0.221316              0.408594               0.419370   \n",
+       "8               0.252100              0.435974               0.464947   \n",
+       "9               0.335388              0.531125               0.549042   \n",
+       "\n",
+       "   melody_complexity (vocals)  \n",
+       "0                    2.250000  \n",
+       "1                    1.916667  \n",
+       "2                    1.833333  \n",
+       "3                    1.916667  \n",
+       "4                    3.500000  \n",
+       "5                    2.916667  \n",
+       "6                    3.250000  \n",
+       "7                    2.500000  \n",
+       "8                    3.416667  \n",
+       "9                    2.833333  "
+      ],
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>genre_subgenre</th>\n",
+       "      <th>spectral_contrast_mean (mix)</th>\n",
+       "      <th>chroma_mean (mix)</th>\n",
+       "      <th>melody_variability (vocals)</th>\n",
+       "      <th>rhythm_onset_rate (mix)</th>\n",
+       "      <th>spectral_centroid_mean custom (mix)</th>\n",
+       "      <th>repetition_score custom (mix)</th>\n",
+       "      <th>pitch_std (mix)</th>\n",
+       "      <th>tempo_bpm_corrected (mix)</th>\n",
+       "      <th>tempo_bpm_original (mix)</th>\n",
+       "      <th>mfcc_mean_1 (mix)</th>\n",
+       "      <th>mfcc_mean_2 (mix)</th>\n",
+       "      <th>loudness_range_lu custom (mix)</th>\n",
+       "      <th>zero_crossing_rate (mix)</th>\n",
+       "      <th>vocab_richness</th>\n",
+       "      <th>loudness_integrated_lufs custom (mix)</th>\n",
+       "      <th>readability_score</th>\n",
+       "      <th>energy_essentia (mix)</th>\n",
+       "      <th>energy_librosa (mix)</th>\n",
+       "      <th>rms_energy_mean (mix)</th>\n",
+       "      <th>melody_complexity (vocals)</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>['Alternative Hip Hop', 'Boom Bap', 'Contempor...</td>\n",
+       "      <td>20.440557</td>\n",
+       "      <td>0.464204</td>\n",
+       "      <td>0.434021</td>\n",
+       "      <td>3.250000</td>\n",
+       "      <td>0.132002</td>\n",
+       "      <td>0.006988</td>\n",
+       "      <td>69.616264</td>\n",
+       "      <td>152.006775</td>\n",
+       "      <td>152.006775</td>\n",
+       "      <td>153.058243</td>\n",
+       "      <td>-7.368185</td>\n",
+       "      <td>1.143722</td>\n",
+       "      <td>0.045159</td>\n",
+       "      <td>0.305</td>\n",
+       "      <td>-12.104649</td>\n",
+       "      <td>20.9</td>\n",
+       "      <td>0.349783</td>\n",
+       "      <td>0.576250</td>\n",
+       "      <td>0.578454</td>\n",
+       "      <td>2.250000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>['Americana', 'Contemporary Country', 'Bro Cou...</td>\n",
+       "      <td>19.657228</td>\n",
+       "      <td>0.424237</td>\n",
+       "      <td>0.351001</td>\n",
+       "      <td>2.425576</td>\n",
+       "      <td>0.139757</td>\n",
+       "      <td>0.007367</td>\n",
+       "      <td>156.358124</td>\n",
+       "      <td>117.943665</td>\n",
+       "      <td>117.943665</td>\n",
+       "      <td>148.055832</td>\n",
+       "      <td>-49.435444</td>\n",
+       "      <td>4.708925</td>\n",
+       "      <td>0.070352</td>\n",
+       "      <td>0.352</td>\n",
+       "      <td>-11.814197</td>\n",
+       "      <td>89.4</td>\n",
+       "      <td>0.429792</td>\n",
+       "      <td>0.619931</td>\n",
+       "      <td>0.632507</td>\n",
+       "      <td>1.916667</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>['Alternative Hip Hop', 'Boom Bap', 'Contempor...</td>\n",
+       "      <td>20.115561</td>\n",
+       "      <td>0.454093</td>\n",
+       "      <td>0.327000</td>\n",
+       "      <td>2.738889</td>\n",
+       "      <td>0.111282</td>\n",
+       "      <td>0.006671</td>\n",
+       "      <td>69.311798</td>\n",
+       "      <td>124.344086</td>\n",
+       "      <td>124.344086</td>\n",
+       "      <td>166.170502</td>\n",
+       "      <td>-5.852028</td>\n",
+       "      <td>1.948263</td>\n",
+       "      <td>0.036463</td>\n",
+       "      <td>0.400</td>\n",
+       "      <td>-12.000179</td>\n",
+       "      <td>107.0</td>\n",
+       "      <td>0.325114</td>\n",
+       "      <td>0.520390</td>\n",
+       "      <td>0.579053</td>\n",
+       "      <td>1.833333</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>['Alternative Hip Hop', 'Contemporary Hip Hop'...</td>\n",
+       "      <td>19.659283</td>\n",
+       "      <td>0.443467</td>\n",
+       "      <td>0.360642</td>\n",
+       "      <td>3.500000</td>\n",
+       "      <td>0.159247</td>\n",
+       "      <td>0.007202</td>\n",
+       "      <td>65.514191</td>\n",
+       "      <td>109.909401</td>\n",
+       "      <td>109.909401</td>\n",
+       "      <td>134.211624</td>\n",
+       "      <td>-27.524824</td>\n",
+       "      <td>1.244652</td>\n",
+       "      <td>0.064695</td>\n",
+       "      <td>0.365</td>\n",
+       "      <td>-12.435019</td>\n",
+       "      <td>105.8</td>\n",
+       "      <td>0.407647</td>\n",
+       "      <td>0.617191</td>\n",
+       "      <td>0.617197</td>\n",
+       "      <td>1.916667</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>['Alternative Hip Hop', 'Contemporary Hip Hop'...</td>\n",
+       "      <td>20.855880</td>\n",
+       "      <td>0.461875</td>\n",
+       "      <td>0.767493</td>\n",
+       "      <td>4.983333</td>\n",
+       "      <td>0.155302</td>\n",
+       "      <td>0.007290</td>\n",
+       "      <td>127.873909</td>\n",
+       "      <td>151.955536</td>\n",
+       "      <td>151.955536</td>\n",
+       "      <td>129.783264</td>\n",
+       "      <td>-11.940094</td>\n",
+       "      <td>3.452597</td>\n",
+       "      <td>0.056507</td>\n",
+       "      <td>0.281</td>\n",
+       "      <td>-16.217091</td>\n",
+       "      <td>212.3</td>\n",
+       "      <td>0.255311</td>\n",
+       "      <td>0.446953</td>\n",
+       "      <td>0.454712</td>\n",
+       "      <td>3.500000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>['Alternative Hip Hop', 'Contemporary Hip Hop'...</td>\n",
+       "      <td>20.649601</td>\n",
+       "      <td>0.582342</td>\n",
+       "      <td>0.625612</td>\n",
+       "      <td>5.522135</td>\n",
+       "      <td>0.158728</td>\n",
+       "      <td>0.007571</td>\n",
+       "      <td>98.469620</td>\n",
+       "      <td>103.800598</td>\n",
+       "      <td>103.800598</td>\n",
+       "      <td>115.804680</td>\n",
+       "      <td>-15.190744</td>\n",
+       "      <td>4.185527</td>\n",
+       "      <td>0.049998</td>\n",
+       "      <td>0.255</td>\n",
+       "      <td>-16.153070</td>\n",
+       "      <td>43.6</td>\n",
+       "      <td>0.310462</td>\n",
+       "      <td>0.503279</td>\n",
+       "      <td>0.503863</td>\n",
+       "      <td>2.916667</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>['Alternative Hip Hop', 'Boom Bap', 'Contempor...</td>\n",
+       "      <td>20.591029</td>\n",
+       "      <td>0.531958</td>\n",
+       "      <td>0.700000</td>\n",
+       "      <td>4.647586</td>\n",
+       "      <td>0.144791</td>\n",
+       "      <td>0.008084</td>\n",
+       "      <td>96.689476</td>\n",
+       "      <td>133.976532</td>\n",
+       "      <td>133.976532</td>\n",
+       "      <td>128.920319</td>\n",
+       "      <td>-6.285958</td>\n",
+       "      <td>5.836010</td>\n",
+       "      <td>0.043094</td>\n",
+       "      <td>0.308</td>\n",
+       "      <td>-17.065815</td>\n",
+       "      <td>94.9</td>\n",
+       "      <td>0.299608</td>\n",
+       "      <td>0.478562</td>\n",
+       "      <td>0.484519</td>\n",
+       "      <td>3.250000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td>['Alternative Hip Hop', 'Contemporary Hip Hop'...</td>\n",
+       "      <td>20.365473</td>\n",
+       "      <td>0.527672</td>\n",
+       "      <td>0.604867</td>\n",
+       "      <td>5.311232</td>\n",
+       "      <td>0.193616</td>\n",
+       "      <td>0.007258</td>\n",
+       "      <td>175.793594</td>\n",
+       "      <td>129.943954</td>\n",
+       "      <td>129.943954</td>\n",
+       "      <td>110.702339</td>\n",
+       "      <td>-5.517988</td>\n",
+       "      <td>8.970278</td>\n",
+       "      <td>0.078429</td>\n",
+       "      <td>0.267</td>\n",
+       "      <td>-18.897171</td>\n",
+       "      <td>38.7</td>\n",
+       "      <td>0.221316</td>\n",
+       "      <td>0.408594</td>\n",
+       "      <td>0.419370</td>\n",
+       "      <td>2.500000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8</th>\n",
+       "      <td>['Alternative Hip Hop', 'Boom Bap', 'Contempor...</td>\n",
+       "      <td>20.128469</td>\n",
+       "      <td>0.526556</td>\n",
+       "      <td>0.802125</td>\n",
+       "      <td>4.563691</td>\n",
+       "      <td>0.158051</td>\n",
+       "      <td>0.007390</td>\n",
+       "      <td>143.052185</td>\n",
+       "      <td>150.016129</td>\n",
+       "      <td>150.016129</td>\n",
+       "      <td>122.700783</td>\n",
+       "      <td>-31.691156</td>\n",
+       "      <td>4.626307</td>\n",
+       "      <td>0.060048</td>\n",
+       "      <td>0.261</td>\n",
+       "      <td>-15.986657</td>\n",
+       "      <td>2.2</td>\n",
+       "      <td>0.252100</td>\n",
+       "      <td>0.435974</td>\n",
+       "      <td>0.464947</td>\n",
+       "      <td>3.416667</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td>['Boom Bap', 'Contemporary Hip Hop', 'Trap', '...</td>\n",
+       "      <td>20.410229</td>\n",
+       "      <td>0.432254</td>\n",
+       "      <td>0.613747</td>\n",
+       "      <td>4.785156</td>\n",
+       "      <td>0.215118</td>\n",
+       "      <td>0.007961</td>\n",
+       "      <td>127.890625</td>\n",
+       "      <td>116.976936</td>\n",
+       "      <td>116.976936</td>\n",
+       "      <td>94.988937</td>\n",
+       "      <td>-20.226351</td>\n",
+       "      <td>4.125075</td>\n",
+       "      <td>0.111702</td>\n",
+       "      <td>0.503</td>\n",
+       "      <td>-13.382388</td>\n",
+       "      <td>41.0</td>\n",
+       "      <td>0.335388</td>\n",
+       "      <td>0.531125</td>\n",
+       "      <td>0.549042</td>\n",
+       "      <td>2.833333</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 10
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T03:57:09.131100100Z",
+     "start_time": "2026-03-31T03:57:07.403030400Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "df['genre_subgenre'] = df['genre_subgenre'].apply(lambda x: ast.literal_eval(x) if isinstance(x, str) else x)\n",
+   "id": "482a32be03dbe231",
+   "outputs": [],
+   "execution_count": 11
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T03:57:09.843702500Z",
+     "start_time": "2026-03-31T03:57:09.739835200Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "type(df['genre_subgenre'][0])",
+   "id": "d6985abd006fac55",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "list"
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 12
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T04:27:54.232241400Z",
+     "start_time": "2026-03-31T04:27:51.718031900Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "import pandas as pd\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from sklearn.preprocessing import LabelEncoder\n",
+    "from sklearn.ensemble import RandomForestClassifier\n",
+    "from sklearn.metrics import classification_report, f1_score\n",
+    "data = pd.read_csv(\"final_clean_data.csv\")\n",
+    "\n",
+    "data['genre_subgenre'] = data['genre_subgenre'].apply(lambda x: ast.literal_eval(x) if isinstance(x, str) else x)\n",
+    "\n",
+    "X = data.drop(columns=[\"genre_subgenre\"])\n",
+    "\n",
+    "\n"
+   ],
+   "id": "735e631b1046e1c9",
+   "outputs": [],
+   "execution_count": 31
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T04:28:25.805903600Z",
+     "start_time": "2026-03-31T04:28:25.687964300Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "data[\"genre_subgenre\"].iloc[0:2,]",
+   "id": "6cc878bce2e5d14",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0    [Alternative Hip Hop, Boom Bap, Contemporary H...\n",
+       "1    [Americana, Contemporary Country, Bro Country,...\n",
+       "Name: genre_subgenre, dtype: object"
+      ]
+     },
+     "execution_count": 33,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 33
+  },
+  {
+   "metadata": {},
+   "cell_type": "code",
+   "outputs": [],
+   "execution_count": null,
+   "source": "",
+   "id": "702f5d9bc5c7bca7"
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T04:25:17.808781400Z",
+     "start_time": "2026-03-31T04:25:17.719637100Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "X.head(2)",
+   "id": "4d9d2e079f5dbd52",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "   spectral_contrast_mean (mix)  chroma_mean (mix)  \\\n",
+       "0                     20.440557           0.464204   \n",
+       "1                     19.657228           0.424237   \n",
+       "\n",
+       "   melody_variability (vocals)  rhythm_onset_rate (mix)  \\\n",
+       "0                     0.434021                 3.250000   \n",
+       "1                     0.351001                 2.425576   \n",
+       "\n",
+       "   spectral_centroid_mean custom (mix)  repetition_score custom (mix)  \\\n",
+       "0                             0.132002                       0.006988   \n",
+       "1                             0.139757                       0.007367   \n",
+       "\n",
+       "   pitch_std (mix)  tempo_bpm_corrected (mix)  tempo_bpm_original (mix)  \\\n",
+       "0        69.616264                 152.006775                152.006775   \n",
+       "1       156.358124                 117.943665                117.943665   \n",
+       "\n",
+       "   mfcc_mean_1 (mix)  mfcc_mean_2 (mix)  loudness_range_lu custom (mix)  \\\n",
+       "0         153.058243          -7.368185                        1.143722   \n",
+       "1         148.055832         -49.435444                        4.708925   \n",
+       "\n",
+       "   zero_crossing_rate (mix)  vocab_richness  \\\n",
+       "0                  0.045159           0.305   \n",
+       "1                  0.070352           0.352   \n",
+       "\n",
+       "   loudness_integrated_lufs custom (mix)  readability_score  \\\n",
+       "0                             -12.104649               20.9   \n",
+       "1                             -11.814197               89.4   \n",
+       "\n",
+       "   energy_essentia (mix)  energy_librosa (mix)  rms_energy_mean (mix)  \\\n",
+       "0               0.349783              0.576250               0.578454   \n",
+       "1               0.429792              0.619931               0.632507   \n",
+       "\n",
+       "   melody_complexity (vocals)  \n",
+       "0                    2.250000  \n",
+       "1                    1.916667  "
+      ],
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>spectral_contrast_mean (mix)</th>\n",
+       "      <th>chroma_mean (mix)</th>\n",
+       "      <th>melody_variability (vocals)</th>\n",
+       "      <th>rhythm_onset_rate (mix)</th>\n",
+       "      <th>spectral_centroid_mean custom (mix)</th>\n",
+       "      <th>repetition_score custom (mix)</th>\n",
+       "      <th>pitch_std (mix)</th>\n",
+       "      <th>tempo_bpm_corrected (mix)</th>\n",
+       "      <th>tempo_bpm_original (mix)</th>\n",
+       "      <th>mfcc_mean_1 (mix)</th>\n",
+       "      <th>mfcc_mean_2 (mix)</th>\n",
+       "      <th>loudness_range_lu custom (mix)</th>\n",
+       "      <th>zero_crossing_rate (mix)</th>\n",
+       "      <th>vocab_richness</th>\n",
+       "      <th>loudness_integrated_lufs custom (mix)</th>\n",
+       "      <th>readability_score</th>\n",
+       "      <th>energy_essentia (mix)</th>\n",
+       "      <th>energy_librosa (mix)</th>\n",
+       "      <th>rms_energy_mean (mix)</th>\n",
+       "      <th>melody_complexity (vocals)</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>20.440557</td>\n",
+       "      <td>0.464204</td>\n",
+       "      <td>0.434021</td>\n",
+       "      <td>3.250000</td>\n",
+       "      <td>0.132002</td>\n",
+       "      <td>0.006988</td>\n",
+       "      <td>69.616264</td>\n",
+       "      <td>152.006775</td>\n",
+       "      <td>152.006775</td>\n",
+       "      <td>153.058243</td>\n",
+       "      <td>-7.368185</td>\n",
+       "      <td>1.143722</td>\n",
+       "      <td>0.045159</td>\n",
+       "      <td>0.305</td>\n",
+       "      <td>-12.104649</td>\n",
+       "      <td>20.9</td>\n",
+       "      <td>0.349783</td>\n",
+       "      <td>0.576250</td>\n",
+       "      <td>0.578454</td>\n",
+       "      <td>2.250000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>19.657228</td>\n",
+       "      <td>0.424237</td>\n",
+       "      <td>0.351001</td>\n",
+       "      <td>2.425576</td>\n",
+       "      <td>0.139757</td>\n",
+       "      <td>0.007367</td>\n",
+       "      <td>156.358124</td>\n",
+       "      <td>117.943665</td>\n",
+       "      <td>117.943665</td>\n",
+       "      <td>148.055832</td>\n",
+       "      <td>-49.435444</td>\n",
+       "      <td>4.708925</td>\n",
+       "      <td>0.070352</td>\n",
+       "      <td>0.352</td>\n",
+       "      <td>-11.814197</td>\n",
+       "      <td>89.4</td>\n",
+       "      <td>0.429792</td>\n",
+       "      <td>0.619931</td>\n",
+       "      <td>0.632507</td>\n",
+       "      <td>1.916667</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ]
+     },
+     "execution_count": 28,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 28
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T04:25:36.595882500Z",
+     "start_time": "2026-03-31T04:25:36.530983400Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "y.iloc[0,]",
+   "id": "926df9dd64132e50",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['Alternative Hip Hop',\n",
+       " 'Boom Bap',\n",
+       " 'Contemporary Hip Hop',\n",
+       " 'Trap',\n",
+       " 'Hip Hop Rap']"
+      ]
+     },
+     "execution_count": 29,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 29
+  },
+  {
+   "metadata": {},
+   "cell_type": "code",
+   "outputs": [],
+   "execution_count": null,
+   "source": "",
+   "id": "64bb9319aa96eb18"
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

genre_finder/test_some_codes/sample_test.py

@@ -0,0 +1,59 @@
+import numpy as np
+import joblib
+nn_model = joblib.load("models/nn_model.pkl")
+pca = joblib.load("models/pca.pkl")
+scaler = joblib.load("models/scaler.pkl")
+feature_cols = joblib.load("models/feature_cols.pkl")
+meta = joblib.load("models/meta.pkl")
+embeddings = np.load("models/embeddings.npy")
+
+
+def recommend_song(input_song_dict, top_k=10):
+    input_vector = np.array([[input_song_dict[col] for col in feature_cols]])
+    input_vector = np.nan_to_num(input_vector)
+    input_scaled = scaler.transform(input_vector)
+    input_embedded = pca.transform(input_scaled)
+    distances, indices = nn_model.kneighbors(
+        input_embedded,
+        n_neighbors=top_k + 1
+    )
+    results = meta.iloc[indices[0]].copy()
+    results["similarity"] = 1 - distances[0]
+    results = results.sort_values("similarity", ascending=False)
+    output = [
+        {
+            "genre_and_sub_genre": row["genre_subgenre"],
+            "Score": float(row["similarity"])
+        }
+        for _, row in results.iterrows()
+    ]
+
+    return output
+
+
+input_song = {'spectral_contrast_mean (mix)': 18.775835817338336,
+ 'chroma_mean (mix)': 0.4393191337585449,
+ 'melody_variability (vocals)': 0.7237541064882048,
+ 'rhythm_onset_rate (mix)': 4.769279513338826,
+ 'spectral_centroid_mean custom (mix)': 0.1423719443498439,
+ 'repetition_score custom (mix)': 0.0086769169570016,
+ 'pitch_std (mix)': 70.28816223144531,
+ 'tempo_bpm_corrected (mix)': 120.0334014892578,
+ 'tempo_bpm_original (mix)': 120.0334014892578,
+ 'mfcc_mean_1 (mix)': 139.3588104248047,
+ 'mfcc_mean_2 (mix)': -10.140735626220703,
+ 'loudness_range_lu custom (mix)': 11.643880844116213,
+ 'zero_crossing_rate (mix)': 0.0576565340161323,
+ 'vocab_richness': 0.47,
+ 'loudness_integrated_lufs custom (mix)': -17.81583023071289,
+ 'readability_score': 159.6,
+ 'energy_essentia (mix)': 0.1938344933977563,
+ 'energy_librosa (mix)': 0.3731466044114756,
+ 'rms_energy_mean (mix)': 0.3730348890104648,
+ 'melody_complexity (vocals)': 3.1666666666666665}
+
+
+results = recommend_song(input_song, top_k=10)
+
+print("\n🎧 Top Recommendations:\n")
+print(results)

genre_finder/test_some_codes/test_file.ipynb

@@ -0,0 +1,481 @@
+{
+ "cells": [
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T08:26:04.824074400Z",
+     "start_time": "2026-03-31T08:26:04.141901500Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import joblib\n",
+    "\n",
+    "from sklearn.preprocessing import StandardScaler\n",
+    "from sklearn.decomposition import PCA\n",
+    "from sklearn.neighbors import NearestNeighbors\n",
+    "\n",
+    "df = pd.read_csv(\"final_clean_data.csv\")\n",
+    "\n",
+    "feature_cols = [\n",
+    "    'spectral_contrast_mean (mix)',\n",
+    "    'chroma_mean (mix)',\n",
+    "    'melody_variability (vocals)',\n",
+    "    'rhythm_onset_rate (mix)',\n",
+    "    'spectral_centroid_mean custom (mix)',\n",
+    "    'repetition_score custom (mix)',\n",
+    "    'pitch_std (mix)',\n",
+    "    'tempo_bpm_corrected (mix)',\n",
+    "    'mfcc_mean_1 (mix)',\n",
+    "    'mfcc_mean_2 (mix)',\n",
+    "    'loudness_range_lu custom (mix)',\n",
+    "    'zero_crossing_rate (mix)',\n",
+    "    'vocab_richness',\n",
+    "    'loudness_integrated_lufs custom (mix)',\n",
+    "    'readability_score',\n",
+    "    'energy_essentia (mix)',\n",
+    "    'energy_librosa (mix)',\n",
+    "    'rms_energy_mean (mix)',\n",
+    "    'melody_complexity (vocals)'\n",
+    "]\n",
+    "\n",
+    "\n",
+    "X = df[feature_cols].copy()\n",
+    "\n",
+    "X = X.fillna(X.mean())\n",
+    "\n",
+    "\n",
+    "scaler = StandardScaler()\n",
+    "X_scaled = scaler.fit_transform(X)\n",
+    "\n",
+    "pca = PCA(n_components=10, random_state=42)\n",
+    "embeddings = pca.fit_transform(X_scaled)\n",
+    "\n",
+    "nn_model = NearestNeighbors(\n",
+    "    n_neighbors=11,\n",
+    "    metric='cosine'\n",
+    ")\n",
+    "\n",
+    "nn_model.fit(embeddings)\n",
+    "\n",
+    "\n",
+    "joblib.dump(nn_model, \"nn_model.pkl\")\n",
+    "joblib.dump(pca, \"pca.pkl\")\n",
+    "joblib.dump(scaler, \"scaler.pkl\")\n",
+    "joblib.dump(feature_cols, \"feature_cols.pkl\")\n",
+    "np.save(\"embeddings.npy\", embeddings)\n",
+    "print(\"✅ Training complete. Model saved.\")"
+   ],
+   "id": "3cd6a6e15e46952d",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "✅ Training complete. Model saved.\n"
+     ]
+    }
+   ],
+   "execution_count": 6
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T08:26:06.210872Z",
+     "start_time": "2026-03-31T08:26:06.056664500Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import joblib\n",
+    "\n",
+    "nn_model = joblib.load(\"nn_model.pkl\")\n",
+    "pca = joblib.load(\"pca.pkl\")\n",
+    "scaler = joblib.load(\"scaler.pkl\")\n",
+    "feature_cols = joblib.load(\"feature_cols.pkl\")\n",
+    "embeddings = np.load(\"embeddings.npy\")\n",
+    "\n",
+    "def recommend_song(input_song_dict, top_k=10):\n",
+    "    input_vector = np.array([[input_song_dict[col] for col in feature_cols]])\n",
+    "    input_vector = np.nan_to_num(input_vector)\n",
+    "    input_scaled = scaler.transform(input_vector)\n",
+    "    input_embedded = pca.transform(input_scaled)\n",
+    "    distances, indices = nn_model.kneighbors(\n",
+    "        input_embedded,\n",
+    "        n_neighbors=top_k + 1\n",
+    "    )\n",
+    "    results = df.iloc[indices[0]].copy()\n",
+    "    results[\"similarity\"] = 1 - distances[0]\n",
+    "    results = results.sort_values(\"similarity\", ascending=False)\n",
+    "    return results\n",
+    "\n",
+    "\n",
+    "input_song = {'spectral_contrast_mean (mix)': 21.826375772089914, 'chroma_mean (mix)': 0.3961135149002075, 'melody_variability (vocals)': 0.38250987027377, 'rhythm_onset_rate (mix)': 3.1277777777777778, 'spectral_centroid_mean custom (mix)': 0.15499075010897467, 'repetition_score custom (mix)': 0.0074139935621249506, 'pitch_std (mix)': 74.42411041259766, 'tempo_bpm_corrected (mix)': 128.70570373535156, 'mfcc_mean_1 (mix)': 131.99278259277344, 'mfcc_mean_2 (mix)': -21.035446166992188, 'loudness_range_lu custom (mix)': 5.404609680175781, 'zero_crossing_rate (mix)': 0.06825674325227737, 'vocab_richness': 0.455, 'loudness_integrated_lufs custom (mix)': -13.08819580078125, 'readability_score': 32.8, 'energy_essentia (mix)': 0.32994533436041884, 'energy_librosa (mix)': 0.5203621226467141, 'rms_energy_mean (mix)': 0.5467776892768835, 'melody_complexity (vocals)': 2.25}\n",
+    "\n",
+    "\n",
+    "results = recommend_song(input_song, top_k=10)\n",
+    "\n",
+    "print(\"\\n🎧 Top Recommendations:\\n\")\n",
+    "print(results[['genre_subgenre', 'similarity']])"
+   ],
+   "id": "ae2f751d6ce2b7b3",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "🎧 Top Recommendations:\n",
+      "\n",
+      "                                           genre_subgenre  similarity\n",
+      "101296  ['Dance Pop', 'Electro Pop', 'Mainstream Pop',...    0.951255\n",
+      "120870  ['Alternative RnB', 'Contemporary RnB', 'Soul'...    0.931561\n",
+      "42886                               ['Americana', 'Rock']    0.931080\n",
+      "61391   ['Contemporary Singer Songwriter', 'Indie Folk...    0.929570\n",
+      "61991           ['Indie Electronic', 'Indie Pop', 'Rock']    0.928316\n",
+      "124708  ['Dance Pop', 'Electropop', 'Synth Pop', 'RnB ...    0.924446\n",
+      "86550                                             ['EDM']    0.922230\n",
+      "101281  ['Dance Pop', 'Electro Pop', 'Indie Pop', 'Mai...    0.922185\n",
+      "120767  ['Alternative RnB', 'Contemporary RnB', 'Soul'...    0.919144\n",
+      "62959           ['Indie Electronic', 'Indie Pop', 'Rock']    0.918303\n",
+      "64296   ['Pop Country', 'Contemporary Country', 'Count...    0.918228\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "C:\\Users\\VigneshSubramani\\PycharmProjects\\recommedation-system\\.venv\\Lib\\site-packages\\sklearn\\utils\\validation.py:2691: UserWarning: X does not have valid feature names, but StandardScaler was fitted with feature names\n",
+      "  warnings.warn(\n"
+     ]
+    }
+   ],
+   "execution_count": 7
+  },
+  {
+   "metadata": {},
+   "cell_type": "markdown",
+   "source": "2) Methods TWO",
+   "id": "e891fa93b34e27ff"
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T07:40:19.349193500Z",
+     "start_time": "2026-03-31T07:40:19.038193100Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import joblib\n",
+    "\n",
+    "from sklearn.preprocessing import StandardScaler\n",
+    "from sklearn.decomposition import PCA\n",
+    "from sklearn.neighbors import NearestNeighbors\n",
+    "\n",
+    "# Load dataset ONLY for training\n",
+    "df = pd.read_csv(\"final_clean_data.csv\")\n",
+    "\n",
+    "feature_cols = [\n",
+    "    'spectral_contrast_mean (mix)',\n",
+    "    'chroma_mean (mix)',\n",
+    "    'melody_variability (vocals)',\n",
+    "    'rhythm_onset_rate (mix)',\n",
+    "    'spectral_centroid_mean custom (mix)',\n",
+    "    'repetition_score custom (mix)',\n",
+    "    'pitch_std (mix)',\n",
+    "    'tempo_bpm_corrected (mix)',\n",
+    "    'mfcc_mean_1 (mix)',\n",
+    "    'mfcc_mean_2 (mix)',\n",
+    "    'loudness_range_lu custom (mix)',\n",
+    "    'zero_crossing_rate (mix)',\n",
+    "    'vocab_richness',\n",
+    "    'loudness_integrated_lufs custom (mix)',\n",
+    "    'readability_score',\n",
+    "    'energy_essentia (mix)',\n",
+    "    'energy_librosa (mix)',\n",
+    "    'rms_energy_mean (mix)',\n",
+    "    'melody_complexity (vocals)'\n",
+    "]\n",
+    "X = df[feature_cols].copy()\n",
+    "X.shape\n"
+   ],
+   "id": "ca229e8117826ebf",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(125169, 19)"
+      ]
+     },
+     "execution_count": 30,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 30
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T07:40:28.563021800Z",
+     "start_time": "2026-03-31T07:40:28.521940600Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "X.columns",
+   "id": "fd7fe702c6c3da52",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Index(['spectral_contrast_mean (mix)', 'chroma_mean (mix)',\n",
+       "       'melody_variability (vocals)', 'rhythm_onset_rate (mix)',\n",
+       "       'spectral_centroid_mean custom (mix)', 'repetition_score custom (mix)',\n",
+       "       'pitch_std (mix)', 'tempo_bpm_corrected (mix)', 'mfcc_mean_1 (mix)',\n",
+       "       'mfcc_mean_2 (mix)', 'loudness_range_lu custom (mix)',\n",
+       "       'zero_crossing_rate (mix)', 'vocab_richness',\n",
+       "       'loudness_integrated_lufs custom (mix)', 'readability_score',\n",
+       "       'energy_essentia (mix)', 'energy_librosa (mix)',\n",
+       "       'rms_energy_mean (mix)', 'melody_complexity (vocals)'],\n",
+       "      dtype='object')"
+      ]
+     },
+     "execution_count": 32,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 32
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T07:43:11.275221600Z",
+     "start_time": "2026-03-31T07:43:11.182964600Z"
+    }
+   },
+   "cell_type": "code",
+   "source": "X.isna().sum()",
+   "id": "d849dd3ee57a5c85",
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "spectral_contrast_mean (mix)             0\n",
+       "chroma_mean (mix)                        0\n",
+       "melody_variability (vocals)              0\n",
+       "rhythm_onset_rate (mix)                  0\n",
+       "spectral_centroid_mean custom (mix)      0\n",
+       "repetition_score custom (mix)            0\n",
+       "pitch_std (mix)                          0\n",
+       "tempo_bpm_corrected (mix)                0\n",
+       "mfcc_mean_1 (mix)                        0\n",
+       "mfcc_mean_2 (mix)                        0\n",
+       "loudness_range_lu custom (mix)           0\n",
+       "zero_crossing_rate (mix)                 0\n",
+       "vocab_richness                           0\n",
+       "loudness_integrated_lufs custom (mix)    0\n",
+       "readability_score                        0\n",
+       "energy_essentia (mix)                    0\n",
+       "energy_librosa (mix)                     0\n",
+       "rms_energy_mean (mix)                    0\n",
+       "melody_complexity (vocals)               0\n",
+       "dtype: int64"
+      ]
+     },
+     "execution_count": 33,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "execution_count": 33
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T07:43:52.008131400Z",
+     "start_time": "2026-03-31T07:43:51.856052100Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "#X = X.fillna(X.mean())\n",
+    "\n",
+    "# Scale\n",
+    "scaler = StandardScaler()\n",
+    "X_scaled = scaler.fit_transform(X)\n",
+    "\n",
+    "# Reduce dimensions\n",
+    "pca = PCA(n_components=10, random_state=42)\n",
+    "embeddings = pca.fit_transform(X_scaled)\n",
+    "\n",
+    "# Nearest Neighbors model\n",
+    "nn_model = NearestNeighbors(\n",
+    "    n_neighbors=11,\n",
+    "    metric='cosine'\n",
+    ")\n",
+    "nn_model.fit(embeddings)\n",
+    "\n",
+    "# 🔥 ONLY SAVE WHAT IS NEEDED (NO CSV)\n",
+    "meta = df[['genre_subgenre']].copy()   # add more columns if needed\n",
+    "\n",
+    "joblib.dump(nn_model, \"models/nn_model.pkl\")\n",
+    "joblib.dump(pca, \"models/pca.pkl\")\n",
+    "joblib.dump(scaler, \"models/scaler.pkl\")\n",
+    "joblib.dump(feature_cols, \"models/feature_cols.pkl\")\n",
+    "joblib.dump(meta, \"models/meta.pkl\")\n",
+    "\n",
+    "np.save(\"models/embeddings.npy\", embeddings)\n",
+    "\n",
+    "print(\"✅ Training complete. All artifacts saved. CSV no longer needed.\")"
+   ],
+   "id": "5c67fbababb8258a",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "✅ Training complete. All artifacts saved. CSV no longer needed.\n"
+     ]
+    }
+   ],
+   "execution_count": 34
+  },
+  {
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2026-03-31T07:44:09.070154500Z",
+     "start_time": "2026-03-31T07:44:08.952731900Z"
+    }
+   },
+   "cell_type": "code",
+   "source": [
+    "import numpy as np\n",
+    "import joblib\n",
+    "\n",
+    "# Load saved artifacts\n",
+    "nn_model = joblib.load(\"models/nn_model.pkl\")\n",
+    "pca = joblib.load(\"models/pca.pkl\")\n",
+    "scaler = joblib.load(\"models/scaler.pkl\")\n",
+    "feature_cols = joblib.load(\"models/feature_cols.pkl\")\n",
+    "meta = joblib.load(\"models/meta.pkl\")\n",
+    "embeddings = np.load(\"models/embeddings.npy\")\n",
+    "\n",
+    "\n",
+    "def recommend_song(input_song_dict, top_k=10):\n",
+    "\n",
+    "    # Convert input to vector in correct order\n",
+    "    input_vector = np.array([[input_song_dict[col] for col in feature_cols]])\n",
+    "    input_vector = np.nan_to_num(input_vector)\n",
+    "\n",
+    "    # Scale + PCA\n",
+    "    input_scaled = scaler.transform(input_vector)\n",
+    "    input_embedded = pca.transform(input_scaled)\n",
+    "\n",
+    "    # Nearest neighbors search\n",
+    "    distances, indices = nn_model.kneighbors(\n",
+    "        input_embedded,\n",
+    "        n_neighbors=top_k + 1\n",
+    "    )\n",
+    "\n",
+    "    # Get metadata instead of CSV\n",
+    "    results = meta.iloc[indices[0]].copy()\n",
+    "    results[\"similarity\"] = 1 - distances[0]\n",
+    "\n",
+    "    results = results.sort_values(\"similarity\", ascending=False)\n",
+    "\n",
+    "    return results\n",
+    "\n",
+    "\n",
+    "# Example input\n",
+    "input_song = {\n",
+    "    'spectral_contrast_mean (mix)': 21.82,\n",
+    "    'chroma_mean (mix)': 0.39,\n",
+    "    'melody_variability (vocals)': 0.38,\n",
+    "    'rhythm_onset_rate (mix)': 3.12,\n",
+    "    'spectral_centroid_mean custom (mix)': 0.15,\n",
+    "    'repetition_score custom (mix)': 0.007,\n",
+    "    'pitch_std (mix)': 74.42,\n",
+    "    'tempo_bpm_corrected (mix)': 128.70,\n",
+    "    'mfcc_mean_1 (mix)': 131.99,\n",
+    "    'mfcc_mean_2 (mix)': -21.03,\n",
+    "    'loudness_range_lu custom (mix)': 5.40,\n",
+    "    'zero_crossing_rate (mix)': 0.068,\n",
+    "    'vocab_richness': 0.455,\n",
+    "    'loudness_integrated_lufs custom (mix)': -13.08,\n",
+    "    'readability_score': 32.8,\n",
+    "    'energy_essentia (mix)': 0.329,\n",
+    "    'energy_librosa (mix)': 0.520,\n",
+    "    'rms_energy_mean (mix)': 0.546,\n",
+    "    'melody_complexity (vocals)': 2.25\n",
+    "}\n",
+    "\n",
+    "results = recommend_song(input_song, top_k=10)\n",
+    "\n",
+    "print(\"\\n🎧 Top Recommendations:\\n\")\n",
+    "print(results)"
+   ],
+   "id": "b00efe2d8a5714b4",
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "🎧 Top Recommendations:\n",
+      "\n",
+      "                                           genre_subgenre  similarity\n",
+      "101296  ['Dance Pop', 'Electro Pop', 'Mainstream Pop',...    0.947162\n",
+      "42886                               ['Americana', 'Rock']    0.936663\n",
+      "120870  ['Alternative RnB', 'Contemporary RnB', 'Soul'...    0.928777\n",
+      "61391   ['Contemporary Singer Songwriter', 'Indie Folk...    0.922589\n",
+      "64296   ['Pop Country', 'Contemporary Country', 'Count...    0.921481\n",
+      "62959           ['Indie Electronic', 'Indie Pop', 'Rock']    0.921080\n",
+      "101281  ['Dance Pop', 'Electro Pop', 'Indie Pop', 'Mai...    0.919901\n",
+      "123420  ['Contemporary RnB', 'Funk', 'Motown', 'Soul',...    0.918827\n",
+      "124708  ['Dance Pop', 'Electropop', 'Synth Pop', 'RnB ...    0.918593\n",
+      "61991           ['Indie Electronic', 'Indie Pop', 'Rock']    0.918424\n",
+      "120767  ['Alternative RnB', 'Contemporary RnB', 'Soul'...    0.917478\n"
+     ]
+    }
+   ],
+   "execution_count": 35
+  },
+  {
+   "metadata": {},
+   "cell_type": "code",
+   "outputs": [],
+   "execution_count": null,
+   "source": "",
+   "id": "b095ca2a85128d25"
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}