{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "df958f0a",
   "metadata": {},
   "source": [
    "# Cleaning the \"Genre\" column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a28a15b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "import altair as alt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from pandas.api.types import is_numeric_dtype\n",
    "pd.set_option('display.float_format', lambda x: '%.3f' % x)\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "rng = np.random.default_rng()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "01c87f3d",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"../data/spotify_dataset.csv\", na_values = \" \")\n",
    "df[\"Streams\"] = pd.to_numeric(df[\"Streams\"].map(lambda s: s.replace(\",\",\"\")))\n",
    "df = df[df.notna().all(axis=1)].copy()\n",
    "numeric_cols = [c for c in df.columns if is_numeric_dtype(df[c])]\n",
    "\n",
    "scaler = StandardScaler()\n",
    "scaler.fit(df[numeric_cols])\n",
    "\n",
    "df[numeric_cols] = scaler.transform(df[numeric_cols])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "156bd759-1e39-4b33-ba60-e5bbb154b1db",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "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>Index</th>\n",
       "      <th>Highest Charting Position</th>\n",
       "      <th>Number of Times Charted</th>\n",
       "      <th>Week of Highest Charting</th>\n",
       "      <th>Song Name</th>\n",
       "      <th>Streams</th>\n",
       "      <th>Artist</th>\n",
       "      <th>Artist Followers</th>\n",
       "      <th>Song ID</th>\n",
       "      <th>Genre</th>\n",
       "      <th>...</th>\n",
       "      <th>Danceability</th>\n",
       "      <th>Energy</th>\n",
       "      <th>Loudness</th>\n",
       "      <th>Speechiness</th>\n",
       "      <th>Acousticness</th>\n",
       "      <th>Liveness</th>\n",
       "      <th>Tempo</th>\n",
       "      <th>Duration (ms)</th>\n",
       "      <th>Valence</th>\n",
       "      <th>Chord</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.731</td>\n",
       "      <td>-1.494</td>\n",
       "      <td>-0.164</td>\n",
       "      <td>2021-07-23--2021-07-30</td>\n",
       "      <td>Beggin'</td>\n",
       "      <td>12.535</td>\n",
       "      <td>Måneskin</td>\n",
       "      <td>-0.680</td>\n",
       "      <td>3Wrjm47oTz2sjIgck11l5e</td>\n",
       "      <td>['indie rock italiano', 'italian pop']</td>\n",
       "      <td>...</td>\n",
       "      <td>0.169</td>\n",
       "      <td>1.031</td>\n",
       "      <td>0.614</td>\n",
       "      <td>-0.664</td>\n",
       "      <td>-0.486</td>\n",
       "      <td>1.234</td>\n",
       "      <td>0.378</td>\n",
       "      <td>0.289</td>\n",
       "      <td>0.327</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-1.729</td>\n",
       "      <td>-1.477</td>\n",
       "      <td>-0.469</td>\n",
       "      <td>2021-07-23--2021-07-30</td>\n",
       "      <td>STAY (with Justin Bieber)</td>\n",
       "      <td>12.124</td>\n",
       "      <td>The Kid LAROI</td>\n",
       "      <td>-0.749</td>\n",
       "      <td>5HCyWlXZPP0y6Gqq8TgA20</td>\n",
       "      <td>['australian hip hop']</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.695</td>\n",
       "      <td>0.808</td>\n",
       "      <td>0.345</td>\n",
       "      <td>-0.683</td>\n",
       "      <td>-0.841</td>\n",
       "      <td>-0.543</td>\n",
       "      <td>1.593</td>\n",
       "      <td>-1.191</td>\n",
       "      <td>-0.162</td>\n",
       "      <td>C#/Db</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-1.727</td>\n",
       "      <td>-1.494</td>\n",
       "      <td>0.019</td>\n",
       "      <td>2021-06-25--2021-07-02</td>\n",
       "      <td>good 4 u</td>\n",
       "      <td>10.024</td>\n",
       "      <td>Olivia Rodrigo</td>\n",
       "      <td>-0.507</td>\n",
       "      <td>4ZtFanR9U6ndgddUvNcjcG</td>\n",
       "      <td>['pop']</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.892</td>\n",
       "      <td>0.189</td>\n",
       "      <td>0.520</td>\n",
       "      <td>0.275</td>\n",
       "      <td>0.345</td>\n",
       "      <td>-0.669</td>\n",
       "      <td>1.491</td>\n",
       "      <td>-0.420</td>\n",
       "      <td>0.763</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-1.725</td>\n",
       "      <td>-1.460</td>\n",
       "      <td>-0.347</td>\n",
       "      <td>2021-07-02--2021-07-09</td>\n",
       "      <td>Bad Habits</td>\n",
       "      <td>9.324</td>\n",
       "      <td>Ed Sheeran</td>\n",
       "      <td>4.114</td>\n",
       "      <td>6PQ88X9TkUIAUIZJHW2upE</td>\n",
       "      <td>['pop', 'uk pop']</td>\n",
       "      <td>...</td>\n",
       "      <td>0.829</td>\n",
       "      <td>1.631</td>\n",
       "      <td>1.051</td>\n",
       "      <td>-0.805</td>\n",
       "      <td>-0.806</td>\n",
       "      <td>1.269</td>\n",
       "      <td>0.109</td>\n",
       "      <td>0.702</td>\n",
       "      <td>0.336</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-1.722</td>\n",
       "      <td>-1.425</td>\n",
       "      <td>-0.591</td>\n",
       "      <td>2021-07-23--2021-07-30</td>\n",
       "      <td>INDUSTRY BABY (feat. Jack Harlow)</td>\n",
       "      <td>8.183</td>\n",
       "      <td>Lil Nas X</td>\n",
       "      <td>-0.554</td>\n",
       "      <td>27NovPIUIRrOZoCHxABJwK</td>\n",
       "      <td>['lgbtq+ hip hop', 'pop rap']</td>\n",
       "      <td>...</td>\n",
       "      <td>0.323</td>\n",
       "      <td>0.436</td>\n",
       "      <td>-0.423</td>\n",
       "      <td>-0.563</td>\n",
       "      <td>-0.913</td>\n",
       "      <td>-0.910</td>\n",
       "      <td>0.919</td>\n",
       "      <td>0.298</td>\n",
       "      <td>1.669</td>\n",
       "      <td>D#/Eb</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1551</th>\n",
       "      <td>1.720</td>\n",
       "      <td>1.844</td>\n",
       "      <td>-0.591</td>\n",
       "      <td>2019-12-27--2020-01-03</td>\n",
       "      <td>New Rules</td>\n",
       "      <td>-0.506</td>\n",
       "      <td>Dua Lipa</td>\n",
       "      <td>0.747</td>\n",
       "      <td>2ekn2ttSfGqwhhate0LSR0</td>\n",
       "      <td>['dance pop', 'pop', 'uk pop']</td>\n",
       "      <td>...</td>\n",
       "      <td>0.506</td>\n",
       "      <td>0.412</td>\n",
       "      <td>0.131</td>\n",
       "      <td>-0.492</td>\n",
       "      <td>-0.983</td>\n",
       "      <td>-0.196</td>\n",
       "      <td>-0.228</td>\n",
       "      <td>0.241</td>\n",
       "      <td>0.411</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1552</th>\n",
       "      <td>1.722</td>\n",
       "      <td>1.861</td>\n",
       "      <td>-0.591</td>\n",
       "      <td>2019-12-27--2020-01-03</td>\n",
       "      <td>Cheirosa - Ao Vivo</td>\n",
       "      <td>-0.508</td>\n",
       "      <td>Jorge &amp; Mateus</td>\n",
       "      <td>0.018</td>\n",
       "      <td>2PWjKmjyTZeDpmOUa3a5da</td>\n",
       "      <td>['sertanejo', 'sertanejo universitario']</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.138</td>\n",
       "      <td>1.464</td>\n",
       "      <td>1.286</td>\n",
       "      <td>-0.349</td>\n",
       "      <td>-0.035</td>\n",
       "      <td>1.054</td>\n",
       "      <td>0.999</td>\n",
       "      <td>-0.340</td>\n",
       "      <td>0.877</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1553</th>\n",
       "      <td>1.724</td>\n",
       "      <td>1.879</td>\n",
       "      <td>-0.591</td>\n",
       "      <td>2019-12-27--2020-01-03</td>\n",
       "      <td>Havana (feat. Young Thug)</td>\n",
       "      <td>-0.509</td>\n",
       "      <td>Camila Cabello</td>\n",
       "      <td>0.479</td>\n",
       "      <td>1rfofaqEpACxVEHIZBJe6W</td>\n",
       "      <td>['dance pop', 'electropop', 'pop', 'post-teen ...</td>\n",
       "      <td>...</td>\n",
       "      <td>0.527</td>\n",
       "      <td>-0.684</td>\n",
       "      <td>0.803</td>\n",
       "      <td>-0.849</td>\n",
       "      <td>-0.259</td>\n",
       "      <td>-0.342</td>\n",
       "      <td>-0.603</td>\n",
       "      <td>0.411</td>\n",
       "      <td>-0.531</td>\n",
       "      <td>D</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1554</th>\n",
       "      <td>1.726</td>\n",
       "      <td>1.896</td>\n",
       "      <td>-0.591</td>\n",
       "      <td>2019-12-27--2020-01-03</td>\n",
       "      <td>Surtada - Remix Brega Funk</td>\n",
       "      <td>-0.513</td>\n",
       "      <td>Dadá Boladão, Tati Zaqui, OIK</td>\n",
       "      <td>-0.870</td>\n",
       "      <td>5F8ffc8KWKNawllr5WsW0r</td>\n",
       "      <td>['brega funk', 'funk carioca']</td>\n",
       "      <td>...</td>\n",
       "      <td>0.997</td>\n",
       "      <td>-0.517</td>\n",
       "      <td>-0.270</td>\n",
       "      <td>-0.589</td>\n",
       "      <td>0.001</td>\n",
       "      <td>0.006</td>\n",
       "      <td>1.057</td>\n",
       "      <td>-0.958</td>\n",
       "      <td>1.612</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1555</th>\n",
       "      <td>1.728</td>\n",
       "      <td>1.913</td>\n",
       "      <td>-0.591</td>\n",
       "      <td>2019-12-27--2020-01-03</td>\n",
       "      <td>Lover (Remix) [feat. Shawn Mendes]</td>\n",
       "      <td>-0.516</td>\n",
       "      <td>Taylor Swift</td>\n",
       "      <td>1.650</td>\n",
       "      <td>3i9UVldZOE0aD0JnyfAZZ0</td>\n",
       "      <td>['pop', 'post-teen pop']</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.699</td>\n",
       "      <td>-0.189</td>\n",
       "      <td>-0.330</td>\n",
       "      <td>-0.541</td>\n",
       "      <td>0.737</td>\n",
       "      <td>-0.660</td>\n",
       "      <td>2.788</td>\n",
       "      <td>0.496</td>\n",
       "      <td>-0.408</td>\n",
       "      <td>G</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1545 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Index  Highest Charting Position  Number of Times Charted  \\\n",
       "0    -1.731                     -1.494                   -0.164   \n",
       "1    -1.729                     -1.477                   -0.469   \n",
       "2    -1.727                     -1.494                    0.019   \n",
       "3    -1.725                     -1.460                   -0.347   \n",
       "4    -1.722                     -1.425                   -0.591   \n",
       "...     ...                        ...                      ...   \n",
       "1551  1.720                      1.844                   -0.591   \n",
       "1552  1.722                      1.861                   -0.591   \n",
       "1553  1.724                      1.879                   -0.591   \n",
       "1554  1.726                      1.896                   -0.591   \n",
       "1555  1.728                      1.913                   -0.591   \n",
       "\n",
       "     Week of Highest Charting                           Song Name  Streams  \\\n",
       "0      2021-07-23--2021-07-30                             Beggin'   12.535   \n",
       "1      2021-07-23--2021-07-30           STAY (with Justin Bieber)   12.124   \n",
       "2      2021-06-25--2021-07-02                            good 4 u   10.024   \n",
       "3      2021-07-02--2021-07-09                          Bad Habits    9.324   \n",
       "4      2021-07-23--2021-07-30   INDUSTRY BABY (feat. Jack Harlow)    8.183   \n",
       "...                       ...                                 ...      ...   \n",
       "1551   2019-12-27--2020-01-03                           New Rules   -0.506   \n",
       "1552   2019-12-27--2020-01-03                  Cheirosa - Ao Vivo   -0.508   \n",
       "1553   2019-12-27--2020-01-03           Havana (feat. Young Thug)   -0.509   \n",
       "1554   2019-12-27--2020-01-03          Surtada - Remix Brega Funk   -0.513   \n",
       "1555   2019-12-27--2020-01-03  Lover (Remix) [feat. Shawn Mendes]   -0.516   \n",
       "\n",
       "                             Artist  Artist Followers                 Song ID  \\\n",
       "0                          Måneskin            -0.680  3Wrjm47oTz2sjIgck11l5e   \n",
       "1                     The Kid LAROI            -0.749  5HCyWlXZPP0y6Gqq8TgA20   \n",
       "2                    Olivia Rodrigo            -0.507  4ZtFanR9U6ndgddUvNcjcG   \n",
       "3                        Ed Sheeran             4.114  6PQ88X9TkUIAUIZJHW2upE   \n",
       "4                         Lil Nas X            -0.554  27NovPIUIRrOZoCHxABJwK   \n",
       "...                             ...               ...                     ...   \n",
       "1551                       Dua Lipa             0.747  2ekn2ttSfGqwhhate0LSR0   \n",
       "1552                 Jorge & Mateus             0.018  2PWjKmjyTZeDpmOUa3a5da   \n",
       "1553                 Camila Cabello             0.479  1rfofaqEpACxVEHIZBJe6W   \n",
       "1554  Dadá Boladão, Tati Zaqui, OIK            -0.870  5F8ffc8KWKNawllr5WsW0r   \n",
       "1555                   Taylor Swift             1.650  3i9UVldZOE0aD0JnyfAZZ0   \n",
       "\n",
       "                                                  Genre  ... Danceability  \\\n",
       "0                ['indie rock italiano', 'italian pop']  ...        0.169   \n",
       "1                                ['australian hip hop']  ...       -0.695   \n",
       "2                                               ['pop']  ...       -0.892   \n",
       "3                                     ['pop', 'uk pop']  ...        0.829   \n",
       "4                         ['lgbtq+ hip hop', 'pop rap']  ...        0.323   \n",
       "...                                                 ...  ...          ...   \n",
       "1551                     ['dance pop', 'pop', 'uk pop']  ...        0.506   \n",
       "1552           ['sertanejo', 'sertanejo universitario']  ...       -1.138   \n",
       "1553  ['dance pop', 'electropop', 'pop', 'post-teen ...  ...        0.527   \n",
       "1554                     ['brega funk', 'funk carioca']  ...        0.997   \n",
       "1555                           ['pop', 'post-teen pop']  ...       -1.699   \n",
       "\n",
       "     Energy  Loudness  Speechiness  Acousticness  Liveness  Tempo  \\\n",
       "0     1.031     0.614       -0.664        -0.486     1.234  0.378   \n",
       "1     0.808     0.345       -0.683        -0.841    -0.543  1.593   \n",
       "2     0.189     0.520        0.275         0.345    -0.669  1.491   \n",
       "3     1.631     1.051       -0.805        -0.806     1.269  0.109   \n",
       "4     0.436    -0.423       -0.563        -0.913    -0.910  0.919   \n",
       "...     ...       ...          ...           ...       ...    ...   \n",
       "1551  0.412     0.131       -0.492        -0.983    -0.196 -0.228   \n",
       "1552  1.464     1.286       -0.349        -0.035     1.054  0.999   \n",
       "1553 -0.684     0.803       -0.849        -0.259    -0.342 -0.603   \n",
       "1554 -0.517    -0.270       -0.589         0.001     0.006  1.057   \n",
       "1555 -0.189    -0.330       -0.541         0.737    -0.660  2.788   \n",
       "\n",
       "      Duration (ms)  Valence  Chord  \n",
       "0             0.289    0.327      B  \n",
       "1            -1.191   -0.162  C#/Db  \n",
       "2            -0.420    0.763      A  \n",
       "3             0.702    0.336      B  \n",
       "4             0.298    1.669  D#/Eb  \n",
       "...             ...      ...    ...  \n",
       "1551          0.241    0.411      A  \n",
       "1552         -0.340    0.877      B  \n",
       "1553          0.411   -0.531      D  \n",
       "1554         -0.958    1.612      F  \n",
       "1555          0.496   -0.408      G  \n",
       "\n",
       "[1545 rows x 23 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a1663752",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       [\n",
       "1       [\n",
       "2       [\n",
       "3       [\n",
       "4       [\n",
       "       ..\n",
       "1551    [\n",
       "1552    [\n",
       "1553    [\n",
       "1554    [\n",
       "1555    [\n",
       "Name: Genre, Length: 1545, dtype: object"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"Genre\"].map(lambda s: s[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "3f609658",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"['australian hip hop']\""
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s = df.loc[1,\"Genre\"]\n",
    "s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "0fed6d0a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['[',\n",
       " \"'\",\n",
       " 'a',\n",
       " 'u',\n",
       " 's',\n",
       " 't',\n",
       " 'r',\n",
       " 'a',\n",
       " 'l',\n",
       " 'i',\n",
       " 'a',\n",
       " 'n',\n",
       " ' ',\n",
       " 'h',\n",
       " 'i',\n",
       " 'p',\n",
       " ' ',\n",
       " 'h',\n",
       " 'o',\n",
       " 'p',\n",
       " \"'\",\n",
       " ']']"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(s)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "5bfe03dc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['australian hip hop']"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eval(s)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "433d4390",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "eval(\"3+5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c7ac025b",
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "eval() arg 1 must be a string, bytes or code object",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m/var/folders/8j/gshrlmtn7dg4qtztj4d4t_w40000gn/T/ipykernel_38831/1095583267.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0meval\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m3\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m5\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m: eval() arg 1 must be a string, bytes or code object"
     ]
    }
   ],
   "source": [
    "eval(3+5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c8cdad35",
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"Genre\"].map(lambda s: eval(s))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "51f62987",
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"Genre\"].map(type)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "095a5026",
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"Genre_list\"] = df[\"Genre\"].map(eval)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e14cf428",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "df[\"Genre_list\"].map(type).value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "af86e0e0",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "df[\"Chord\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cbc2dda8",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.head(4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1699f9ad",
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"Genre_list\"].map(lambda x: \"pop\" in x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1c79bf00",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}