feature_extraction_genres.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "feature_extraction_genres.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "yvBDi7ZvJiVQ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Second iteration of this function to build up the dataframe\n",
        "\n",
        "import requests\n",
        "import spotipy\n",
        "from spotipy.oauth2 import SpotifyClientCredentials\n",
        "import lyricsgenius as genius\n",
        "import pandas as pd\n",
        "import json\n",
        "import time\n",
        "\n",
        "# #################################################\n",
        "#   Spotify API user credentials, generate token\n",
        "CLIENT_ID = <CLIENT_ID>\n",
        "CLIENT_SECRET = <CLIENT_SECRET>\n",
        "credentials = SpotifyClientCredentials(\n",
        "        client_id=CLIENT_ID,\n",
        "        client_secret=CLIENT_SECRET)\n",
        "\n",
        "token = credentials.get_access_token(as_dict=False)\n",
        "spotify = spotipy.Spotify(auth=token)\n",
        "spotify.trace = False\n",
        "# #################################################\n",
        "\n",
        "# #################################################\n",
        "#   Genius API settings and user credentials\n",
        "genius_api = genius.Genius(<GENIUS_CREDENTIALS>)\n",
        "genius.verbose = True                         # Keep status messages on\n",
        "genius.remove_section_headers = False         # Remove section headers (e.g. [Chorus]) from lyrics when searching\n",
        "genius.excluded_terms = [\"(Remix)\", \"(Live)\"] # Exclude songs with these words in their title\n",
        "# #################################################\n",
        "\n",
        "cols =['track_name',\n",
        "       'artist_name',\n",
        "       'track_id',\n",
        "       'song_spotify_link',\n",
        "       'artist_id',\n",
        "       'release_date',\n",
        "       'release_date_precision',\n",
        "       'duration',\n",
        "       'popularity',\n",
        "       'explicit',\n",
        "       'audio_preview',\n",
        "       'cover_art',\n",
        "       'danceability',\n",
        "       'energy',\n",
        "       'key',\n",
        "       'loudness',\n",
        "       'mode',\n",
        "       'speechiness',\n",
        "       'acousticness',\n",
        "       'instrumentalness',\n",
        "       'liveness',\n",
        "       'valence',\n",
        "       'tempo',\n",
        "       'time_sig',\n",
        "       'num_artist_followers',\n",
        "       'artist_genres',\n",
        "       'artist_popularity',\n",
        "       'artist_spotify_avatar',\n",
        "       'artist_spotify_link']\n",
        "\n",
        "df = pd.DataFrame(columns=cols)\n",
        "\n",
        "genre_list = pd.read_csv('max_genres.csv')\n",
        "\n",
        "genre_list = genre_list[1368:]\n",
        "\n",
        "start = time.time()\n",
        "\n",
        "# #################################################\n",
        "#      Initial user query for a track name\n",
        "# #################################################\n",
        "\n",
        "#results = spotify.search(q='track:' + \"Thriller\", type='track', limit=1)\n",
        "#results = spotify.search(\"Californication\", type=\"track\", limit=1)\n",
        "\n",
        "for genre in genre_list['genre']:\n",
        "  print(genre)\n",
        "  api_query = 'genre:' + genre\n",
        "  results = spotify.search(q=api_query, limit=50, market='US')\n",
        "  time.sleep(1)\n",
        "  data = {}\n",
        "  if len(results['tracks']['items']) > 0:\n",
        "    for i in range(0,len(results['tracks']['items'])):\n",
        "      for j in cols:\n",
        "        data[j] = []\n",
        "      # API calls\n",
        "      track = results['tracks']['items'][i]\n",
        "      audio_features = spotify.audio_features(str(track['uri']))[0]\n",
        "      primary_artist = spotify.artist(str(track['album']['artists'][0]['uri']))\n",
        "      # Build features\n",
        "      data['track_name'] = track['name']\n",
        "      data['artist_name'] = track['artists'][0]['name']\n",
        "      data['track_id'] = track['id']\n",
        "      data['song_spotify_link'] = track['external_urls']['spotify']\n",
        "      data['artist_id'] = track['artists'][0]['id']\n",
        "      data['release_date'] = track['album']['release_date']\n",
        "      data['release_date_precision'] = track['album']['release_date_precision']\n",
        "      data['duration'] = track['duration_ms']\n",
        "      data['popularity'] = track['popularity']\n",
        "      data['explicit'] = track['explicit']\n",
        "      data['audio_preview'] = track['preview_url']\n",
        "      data['cover_art'] = track['album']['images'][1]['url']\n",
        "      data['danceability'] = audio_features['danceability']\n",
        "      data['energy'] = audio_features['energy']\n",
        "      data['key'] = audio_features['key']\n",
        "      data['loudness'] = audio_features['loudness']\n",
        "      data['mode'] = audio_features['mode']\n",
        "      data['speechiness'] = audio_features['speechiness']\n",
        "      data['acousticness'] = audio_features['acousticness']\n",
        "      data['instrumentalness'] = audio_features['instrumentalness']\n",
        "      data['liveness'] = audio_features['liveness']\n",
        "      data['valence'] = audio_features['valence']\n",
        "      data['tempo'] = audio_features['tempo']\n",
        "      data['time_sig'] = audio_features['time_signature']\n",
        "      data['num_artist_followers'] = primary_artist['followers']['total']\n",
        "      data['artist_genres'] = primary_artist['genres']\n",
        "      data['artist_popularity'] = primary_artist['popularity']\n",
        "      data['artist_spotify_avatar'] = primary_artist['images'][1]['url']\n",
        "      data['artist_spotify_link'] = primary_artist['external_urls']['spotify']\n",
        "      df = df.append(data, ignore_index=True)\n",
        "      time.sleep(0.2)\n",
        "  else:\n",
        "    print('Track not found')\n",
        "  time.sleep(0.5)"
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}