{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Battle of the Bands: generic data analysis\n", "\n", "This does the analysis of the band data, once it's [been gathered](multi-artist-gather-data.ipynb). " ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "run_control": { "read_only": false } }, "outputs": [], "source": [ "import pandas as pd\n", "import numpy as np\n", "import matplotlib\n", "import matplotlib.pyplot as plt\n", "%matplotlib inline \n", "import urllib.request\n", "import urllib.parse\n", "import urllib.error\n", "import json\n", "import base64\n", "import configparser\n", "from bs4 import BeautifulSoup\n", "import re\n", "import pymongo\n", "from datetime import datetime\n", "import time\n", "import collections\n", "import editdistance\n", "import math\n", "from scipy.spatial import ConvexHull\n", "from pandas.plotting import scatter_matrix" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We'll use MongoDB to store the data, to save keeping it all in memory, and mean we don't have to recapture all the data to to a different analysis." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "# Open a connection to the Mongo server\n", "client = pymongo.MongoClient('mongodb://localhost:27017/')" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "# Create a database and a collections within it.\n", "songs_db = client.songs\n", "albums = songs_db.albums\n", "unfiltered_tracks = songs_db.tracks\n", "genius_tracks = songs_db.gtracks" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "unfiltered_tracks.update_many({'gloom': {'$exists': True}}, {'$unset': {'gloom': ''}})\n", "unfiltered_tracks.update_many({'complexity': {'$exists': True}}, {'$unset': {'complexity': ''}})" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['system.views',\n", " 'gtracks',\n", " 'system.indexes',\n", " 'sentiments',\n", " 'albums',\n", " 'tracks',\n", " 'interesting_tracks']" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "songs_db.collection_names()" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "interesting_pipe = [{'$match':{'ignore': {'$exists': False}}}]\n", "if 'interesting_tracks' not in songs_db.collection_names():\n", " songs_db.command(\"create\", \"interesting_tracks\",\n", " viewOn='tracks', \n", " pipeline=interesting_pipe)" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "tracks = songs_db.interesting_tracks" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "API keys and the like are kept in a configuration file, which is read here.\n", "\n", "You'll need to create a web API key for Spotify and Genius. " ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['app_name', 'client_id', 'client_secret', 'redirect_uri', 'token']" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "config = configparser.ConfigParser()\n", "config.read('secrets.ini')\n", "[k for k in config['genius']]" ] }, { "cell_type": "code", "execution_count": 57, "metadata": {}, "outputs": [], "source": [ "artist_ids = { 'The Rolling Stones': '22bE4uQ6baNwSHPVcDxLCe'\n", " , 'The Beatles': '3WrFJ7ztbogyGnTHbHJFl2'\n", " , 'Radiohead': '4Z8W4fKeB5YxbusRsdQVPb'\n", " , 'Spice Girls': '0uq5PttqEjj3IH1bzwcrXF'\n", " , 'Abba': '0LcJLqbBmaGUft1e9Mm8HV'\n", " , 'Foo Fighters': '7jy3rLJdDQY21OgRLCZ9sD'\n", " , 'Led Zeppelin': '36QJpDe2go2KgaRleHCDTp'\n", " , 'Queen' : '1dfeR4HaWDbWqFHLkxsg1d'\n", " }" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "# radiohead_id = albums.find_one({'artist_name': 'Radiohead'})['artist_id']\n", "# radiohead_id" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Which values to analyse?\n", "\n", "Find all the possible scores and pull them into a dataframe." ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/plain": [ "dict_keys(['artist_name', 'original_lyrics', 'instrumentalness', 'explicit', 'mode', 'tempo', 'acousticness', 'duration_ms', 'liveness', 'preview_url', 'analysis_url', 'available_markets', 'href', 'album_id', 'track_number', 'lyrical_density', 'disc_number', 'name', 'album', 'artists', 'time_signature', 'energy', 'id', '_id', 'key', 'uri', 'speechiness', 'popularity', 'external_ids', 'artist_id', 'danceability', 'external_urls', 'track_href', 'type', 'loudness', 'ctitle', 'valence'])" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tracks.find_one().keys()" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'_id': 0,\n", " 'acousticness': '$acousticness',\n", " 'danceability': '$danceability',\n", " 'energy': '$energy',\n", " 'instrumentalness': '$instrumentalness',\n", " 'key': '$key',\n", " 'liveness': '$liveness',\n", " 'loudness': '$loudness',\n", " 'lyrical_density': '$lyrical_density',\n", " 'neg': '$sentiment.probability.neg',\n", " 'nnrc_anger': '$nnrc_sentiment.anger',\n", " 'nnrc_anticipation': '$nnrc_sentiment.anticipation',\n", " 'nnrc_disgust': '$nnrc_sentiment.disgust',\n", " 'nnrc_fear': '$nnrc_sentiment.fear',\n", " 'nnrc_joy': '$nnrc_sentiment.joy',\n", " 'nnrc_negative': '$nnrc_sentiment.negative',\n", " 'nnrc_positive': '$nnrc_sentiment.positive',\n", " 'nnrc_sadness': '$nnrc_sentiment.sadness',\n", " 'nnrc_surprise': '$nnrc_sentiment.surprise',\n", " 'nnrc_trust': '$nnrc_sentiment.trust',\n", " 'popularity': '$popularity',\n", " 'popularity0': {'$literal': 0},\n", " 'pos': '$sentiment.probability.pos',\n", " 'speechiness': '$speechiness',\n", " 'tempo': '$tempo',\n", " 'time_signature': '$time_signature',\n", " 'valence': '$valence'}" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "numeric_keys = ['popularity', 'instrumentalness', 'speechiness', 'tempo', \n", " 'danceability', 'acousticness', 'loudness', 'time_signature', \n", " 'lyrical_density', 'valence', 'liveness', 'energy', 'key']\n", "# 'explicit', # complexity, # gloom\n", "\n", "projection_dict = {k: '$' + k for k in numeric_keys}\n", "projection_dict.update({'neg': '$sentiment.probability.neg',\n", " 'pos': '$sentiment.probability.pos', \n", " 'nnrc_fear': '$nnrc_sentiment.fear',\n", " 'nnrc_trust': '$nnrc_sentiment.trust',\n", " 'nnrc_surprise': '$nnrc_sentiment.surprise',\n", " 'nnrc_anticipation': '$nnrc_sentiment.anticipation',\n", " 'nnrc_sadness': '$nnrc_sentiment.sadness',\n", " 'nnrc_joy': '$nnrc_sentiment.joy',\n", " 'nnrc_positive': '$nnrc_sentiment.positive',\n", " 'nnrc_disgust': '$nnrc_sentiment.disgust',\n", " 'nnrc_anger': '$nnrc_sentiment.anger',\n", " 'nnrc_negative': '$nnrc_sentiment.negative',\n", " 'popularity0': {'$literal': 0},\n", " '_id': 0})\n", "projection_dict" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [], "source": [ "pipeline = [\n", " {'$match': {'lyrics': {'$exists': True}, 'sentiment': {'$exists': True}, 'valence': {'$exists': True}}},\n", " {'$project': projection_dict}\n", "]\n", "all_pre_raw_df = pd.DataFrame(list(tracks.aggregate(pipeline)))" ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "scrolled": false }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
countmeanstdmin25%50%75%max
acousticness1274.00.2479550.2632120.0000020.0239000.1455000.3940000.968000
danceability1274.00.4746830.1567530.0000000.3592500.4720000.5830000.933000
energy1274.00.6922220.2185280.0165000.5330000.7300000.8787500.999000
instrumentalness1274.00.1044110.2318550.0000000.0000080.0005500.0441500.999000
key1274.04.9285713.4959860.0000002.0000005.0000008.00000011.000000
liveness1274.00.2842020.2592440.0000000.1020000.1720000.3590000.990000
loudness1274.0-7.9856592.905025-23.459000-9.788750-7.731500-5.871250-1.429000
lyrical_density1274.00.9151630.4187260.0038150.6274430.8695661.1401012.889722
neg1274.00.6189220.1718870.1019050.5068200.6249370.7683550.904388
nnrc_anger1005.00.3115440.2285230.0128210.1428570.2500000.4230771.000000
nnrc_anticipation1147.00.4800080.2787250.0322580.2500000.4444440.6666671.000000
nnrc_disgust886.00.2792930.2222650.0128210.1176470.2000000.3846151.000000
nnrc_fear1082.00.3729230.2671280.0128210.1500000.3076920.5000001.000000
nnrc_joy1186.00.5868790.2828030.0357140.3333330.6153850.8333331.000000
nnrc_negative1180.00.6195850.3321320.0128210.3333330.6220241.0000001.000000
nnrc_positive1238.00.8182180.2713210.0357140.6488101.0000001.0000001.000000
nnrc_sadness1082.00.4014610.2620950.0128210.2000000.3333330.5714291.000000
nnrc_surprise994.00.2924850.2086300.0128210.1333330.2380950.4086701.000000
nnrc_trust1174.00.4412660.2619090.0303030.2307690.4000000.6111111.000000
popularity1274.037.84693914.1510520.00000025.00000038.00000049.00000078.000000
popularity01274.00.0000000.0000000.0000000.0000000.0000000.0000000.000000
pos1274.00.3810780.1718870.0956120.2316450.3750630.4931800.898095
speechiness1274.00.0579980.0449480.0000000.0332000.0423000.0638000.475000
tempo1274.0122.95484230.4831330.000000101.329250121.364000140.994000211.099000
time_signature1274.03.8854000.4225280.0000004.0000004.0000004.0000005.000000
valence1274.00.5250840.2478210.0000000.3320000.5290000.7230000.976000
\n", "
" ], "text/plain": [ " count mean std min 25% \\\n", "acousticness 1274.0 0.247955 0.263212 0.000002 0.023900 \n", "danceability 1274.0 0.474683 0.156753 0.000000 0.359250 \n", "energy 1274.0 0.692222 0.218528 0.016500 0.533000 \n", "instrumentalness 1274.0 0.104411 0.231855 0.000000 0.000008 \n", "key 1274.0 4.928571 3.495986 0.000000 2.000000 \n", "liveness 1274.0 0.284202 0.259244 0.000000 0.102000 \n", "loudness 1274.0 -7.985659 2.905025 -23.459000 -9.788750 \n", "lyrical_density 1274.0 0.915163 0.418726 0.003815 0.627443 \n", "neg 1274.0 0.618922 0.171887 0.101905 0.506820 \n", "nnrc_anger 1005.0 0.311544 0.228523 0.012821 0.142857 \n", "nnrc_anticipation 1147.0 0.480008 0.278725 0.032258 0.250000 \n", "nnrc_disgust 886.0 0.279293 0.222265 0.012821 0.117647 \n", "nnrc_fear 1082.0 0.372923 0.267128 0.012821 0.150000 \n", "nnrc_joy 1186.0 0.586879 0.282803 0.035714 0.333333 \n", "nnrc_negative 1180.0 0.619585 0.332132 0.012821 0.333333 \n", "nnrc_positive 1238.0 0.818218 0.271321 0.035714 0.648810 \n", "nnrc_sadness 1082.0 0.401461 0.262095 0.012821 0.200000 \n", "nnrc_surprise 994.0 0.292485 0.208630 0.012821 0.133333 \n", "nnrc_trust 1174.0 0.441266 0.261909 0.030303 0.230769 \n", "popularity 1274.0 37.846939 14.151052 0.000000 25.000000 \n", "popularity0 1274.0 0.000000 0.000000 0.000000 0.000000 \n", "pos 1274.0 0.381078 0.171887 0.095612 0.231645 \n", "speechiness 1274.0 0.057998 0.044948 0.000000 0.033200 \n", "tempo 1274.0 122.954842 30.483133 0.000000 101.329250 \n", "time_signature 1274.0 3.885400 0.422528 0.000000 4.000000 \n", "valence 1274.0 0.525084 0.247821 0.000000 0.332000 \n", "\n", " 50% 75% max \n", "acousticness 0.145500 0.394000 0.968000 \n", "danceability 0.472000 0.583000 0.933000 \n", "energy 0.730000 0.878750 0.999000 \n", "instrumentalness 0.000550 0.044150 0.999000 \n", "key 5.000000 8.000000 11.000000 \n", "liveness 0.172000 0.359000 0.990000 \n", "loudness -7.731500 -5.871250 -1.429000 \n", "lyrical_density 0.869566 1.140101 2.889722 \n", "neg 0.624937 0.768355 0.904388 \n", "nnrc_anger 0.250000 0.423077 1.000000 \n", "nnrc_anticipation 0.444444 0.666667 1.000000 \n", "nnrc_disgust 0.200000 0.384615 1.000000 \n", "nnrc_fear 0.307692 0.500000 1.000000 \n", "nnrc_joy 0.615385 0.833333 1.000000 \n", "nnrc_negative 0.622024 1.000000 1.000000 \n", "nnrc_positive 1.000000 1.000000 1.000000 \n", "nnrc_sadness 0.333333 0.571429 1.000000 \n", "nnrc_surprise 0.238095 0.408670 1.000000 \n", "nnrc_trust 0.400000 0.611111 1.000000 \n", "popularity 38.000000 49.000000 78.000000 \n", "popularity0 0.000000 0.000000 0.000000 \n", "pos 0.375063 0.493180 0.898095 \n", "speechiness 0.042300 0.063800 0.475000 \n", "tempo 121.364000 140.994000 211.099000 \n", "time_signature 4.000000 4.000000 5.000000 \n", "valence 0.529000 0.723000 0.976000 " ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "all_pre_raw_df.describe().T" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now we have the ranges, move them all into the range 0-1, remembering the scaling for future use." ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
countmeanstdmin25%50%75%max
acousticness1274.00.2561500.2719140.00.0246880.1503080.4070231.0
danceability1274.00.5087710.1680090.00.3850480.5058950.6248661.0
energy1274.00.6877570.2224200.00.5257000.7262090.8776081.0
instrumentalness1274.00.1045160.2320870.00.0000080.0005510.0441941.0
key1274.00.4480520.3178170.00.1818180.4545450.7272731.0
liveness1274.00.2870730.2618620.00.1030300.1737370.3626261.0
loudness1274.00.7023760.1318670.00.6205290.7139130.7983551.0
lyrical_density1274.00.3157920.1450930.00.2160940.2999930.3937361.0
neg1274.00.6442720.2141940.00.5045780.6517670.8304851.0
nnrc_anger1005.00.3026030.2314910.00.1317250.2402600.4155841.0
nnrc_anticipation1147.00.4626750.2880160.00.2250000.4259260.6555561.0
nnrc_disgust886.00.2699330.2251520.00.1061880.1896100.3766231.0
nnrc_fear1082.00.3647790.2705970.00.1389610.2987010.4935061.0
nnrc_joy1186.00.5715780.2932770.00.3086420.6011400.8271601.0
nnrc_negative1180.00.6146440.3364450.00.3246750.6171151.0000001.0
nnrc_positive1238.00.8114850.2813700.00.6358021.0000001.0000001.0
nnrc_sadness1082.00.3936880.2654990.00.1896100.3246750.5658631.0
nnrc_surprise994.00.2832970.2113390.00.1220780.2282000.4009901.0
nnrc_trust1174.00.4238060.2700940.00.2067310.3812500.5989581.0
popularity1274.00.4852170.1814240.00.3205130.4871790.6282051.0
popularity01274.00.0000000.0000000.00.0000000.0000000.0000000.0
pos1274.00.3557280.2141940.00.1695150.3482330.4954221.0
speechiness1274.00.1221010.0946270.00.0698950.0890530.1343161.0
tempo1274.00.5824510.1444020.00.4800080.5749150.6679051.0
time_signature1274.00.7770800.0845060.00.8000000.8000000.8000001.0
valence1274.00.5379960.2539150.00.3401640.5420080.7407791.0
\n", "
" ], "text/plain": [ " count mean std min 25% 50% \\\n", "acousticness 1274.0 0.256150 0.271914 0.0 0.024688 0.150308 \n", "danceability 1274.0 0.508771 0.168009 0.0 0.385048 0.505895 \n", "energy 1274.0 0.687757 0.222420 0.0 0.525700 0.726209 \n", "instrumentalness 1274.0 0.104516 0.232087 0.0 0.000008 0.000551 \n", "key 1274.0 0.448052 0.317817 0.0 0.181818 0.454545 \n", "liveness 1274.0 0.287073 0.261862 0.0 0.103030 0.173737 \n", "loudness 1274.0 0.702376 0.131867 0.0 0.620529 0.713913 \n", "lyrical_density 1274.0 0.315792 0.145093 0.0 0.216094 0.299993 \n", "neg 1274.0 0.644272 0.214194 0.0 0.504578 0.651767 \n", "nnrc_anger 1005.0 0.302603 0.231491 0.0 0.131725 0.240260 \n", "nnrc_anticipation 1147.0 0.462675 0.288016 0.0 0.225000 0.425926 \n", "nnrc_disgust 886.0 0.269933 0.225152 0.0 0.106188 0.189610 \n", "nnrc_fear 1082.0 0.364779 0.270597 0.0 0.138961 0.298701 \n", "nnrc_joy 1186.0 0.571578 0.293277 0.0 0.308642 0.601140 \n", "nnrc_negative 1180.0 0.614644 0.336445 0.0 0.324675 0.617115 \n", "nnrc_positive 1238.0 0.811485 0.281370 0.0 0.635802 1.000000 \n", "nnrc_sadness 1082.0 0.393688 0.265499 0.0 0.189610 0.324675 \n", "nnrc_surprise 994.0 0.283297 0.211339 0.0 0.122078 0.228200 \n", "nnrc_trust 1174.0 0.423806 0.270094 0.0 0.206731 0.381250 \n", "popularity 1274.0 0.485217 0.181424 0.0 0.320513 0.487179 \n", "popularity0 1274.0 0.000000 0.000000 0.0 0.000000 0.000000 \n", "pos 1274.0 0.355728 0.214194 0.0 0.169515 0.348233 \n", "speechiness 1274.0 0.122101 0.094627 0.0 0.069895 0.089053 \n", "tempo 1274.0 0.582451 0.144402 0.0 0.480008 0.574915 \n", "time_signature 1274.0 0.777080 0.084506 0.0 0.800000 0.800000 \n", "valence 1274.0 0.537996 0.253915 0.0 0.340164 0.542008 \n", "\n", " 75% max \n", "acousticness 0.407023 1.0 \n", "danceability 0.624866 1.0 \n", "energy 0.877608 1.0 \n", "instrumentalness 0.044194 1.0 \n", "key 0.727273 1.0 \n", "liveness 0.362626 1.0 \n", "loudness 0.798355 1.0 \n", "lyrical_density 0.393736 1.0 \n", "neg 0.830485 1.0 \n", "nnrc_anger 0.415584 1.0 \n", "nnrc_anticipation 0.655556 1.0 \n", "nnrc_disgust 0.376623 1.0 \n", "nnrc_fear 0.493506 1.0 \n", "nnrc_joy 0.827160 1.0 \n", "nnrc_negative 1.000000 1.0 \n", "nnrc_positive 1.000000 1.0 \n", "nnrc_sadness 0.565863 1.0 \n", "nnrc_surprise 0.400990 1.0 \n", "nnrc_trust 0.598958 1.0 \n", "popularity 0.628205 1.0 \n", "popularity0 0.000000 0.0 \n", "pos 0.495422 1.0 \n", "speechiness 0.134316 1.0 \n", "tempo 0.667905 1.0 \n", "time_signature 0.800000 1.0 \n", "valence 0.740779 1.0 " ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "all_pre_df=(all_pre_raw_df-all_pre_raw_df.min())/(all_pre_raw_df.max()-all_pre_raw_df.min())\n", "all_pre_df.popularity0 = 0\n", "all_pre_df.describe().T" ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
acousticnessdanceabilityenergyinstrumentalnesskeylivenessloudnesslyrical_densitynegnnrc_anger...nnrc_sadnessnnrc_surprisennrc_trustpopularitypopularity0posspeechinesstempotime_signaturevalence
00.4390480.8488750.7597960.6276280.1818180.7969700.5571490.0575500.360472NaN...NaNNaNNaN0.56410300.6395280.1065260.4258670.80.155738
10.1404940.5541260.5949110.8348350.1818180.1202020.6084880.0078070.622129NaN...NaNNaNNaN0.39743600.3778710.0869470.3353310.80.760246
20.7117760.2486600.4829520.9389390.0909090.0668690.7322290.0063600.537797NaN...NaNNaNNaN0.43589700.4622030.0692630.3297600.60.156762
30.0659070.4351550.6966920.8968970.3636360.2313130.7808440.0002070.537797NaN...NaNNaNNaN0.34615400.4622030.0633680.7399180.80.478484
40.8698340.3644160.3201020.8338340.0000000.1131310.2542900.0077840.687087NaN...NaNNaNNaN0.57692300.3129130.0684210.6710740.80.682377
50.2200390.3118970.9755730.9089090.5454550.3353540.5767140.0000000.537797NaN...NaNNaNNaN0.43589700.4622030.1172630.8540500.80.130123
60.2200390.3118970.9755730.9089090.5454550.3353540.5767140.0000000.537797NaN...NaNNaNNaN0.41025600.4622030.1172630.8540500.80.130123
70.1466920.0000000.0000001.0000000.9090910.0000000.3686340.0838790.504531NaN...NaNNaNNaN0.00000000.4954690.0000000.0000000.00.000000
80.5237590.6227220.8188300.0000000.0000000.1090910.8657290.3655940.532734NaN...NaN0.1558440.1406250.69230800.4672660.0669470.6449340.80.991803
90.2685930.4040730.9155220.0000000.3636360.0747470.9196550.5568540.5369180.366883...0.620130NaN0.0976560.74359000.4630820.1012630.3578080.80.934426
100.3987590.5251880.7109410.0000000.6363640.3141410.8129820.4432680.520248NaN...NaNNaN0.3125000.82051300.4797520.1002110.6192640.80.887295
110.0933860.6045020.8249360.0000040.6363640.1282830.7430780.4089800.369287NaN...NaN0.0466000.0900740.69230800.6307130.0595790.4256150.80.934426
120.1239650.7127550.7791350.0000040.5454550.1262630.6818430.2898300.726211NaN...NaN0.2764380.7053570.75641000.2737890.0646320.6511310.80.748975
130.0655970.4051450.6722650.0000000.1818180.1848480.6731280.6089990.5968230.392208...0.1896100.2909090.2781250.74359000.4031770.0627370.5024700.80.663934
140.1518570.5058950.8544530.0000030.6363640.6010100.7911480.4825910.605477NaN...NaN0.0649350.0480770.67948700.3945230.1183160.3727920.80.861680
150.1456590.6966770.7740460.3433430.1818180.2171720.8386290.3329650.6084550.263282...0.0791030.1711920.6250000.67948700.3915450.0618950.5192260.80.588115
160.0115680.4137190.6010180.0000140.9090910.0888890.7153430.1810300.5957860.392208...0.594805NaN0.3812500.87179500.4042140.0549470.6973360.80.545082
170.0048010.6977490.8910940.0000030.3636360.1282830.7674990.6145530.6296000.240260...0.1136360.7467530.4843750.66666700.3704000.0658950.6409740.80.985656
180.1006180.7781350.6977100.0000350.6363640.2414140.5765770.3055330.759113NaN...NaN0.2764380.4107140.62820500.2408870.0673680.5168850.80.953893
190.4876020.5766350.4697200.0000001.0000000.1717170.6582840.2714640.5740130.662338...NaNNaNNaN0.60256400.4259870.0816840.2795090.80.539959
200.3729320.2754560.4025450.0000870.0909090.0709090.5297780.3237230.2412760.189610...0.3246750.1896100.8625000.74359000.7587240.0604210.7206380.80.879098
210.1849150.3118970.6325700.0000000.8181820.1020200.7083520.2295480.759066NaN...NaN1.0000000.4843750.61538500.2409340.1166320.8774790.60.537910
220.6807840.5787780.4809160.0019020.4545450.4616160.5076260.3900460.3547751.000000...NaNNaNNaN0.56410300.6452250.2463160.7438690.60.686475
230.6518590.4748120.3933840.0000000.0000000.1121210.6863370.3451510.0223710.610390...0.3766230.0649350.6826920.89743600.9776290.0677890.6795960.80.420082
240.3915270.5530550.5073790.0000000.1818180.1040400.5185660.4746550.7661990.324675...0.3246750.3246751.0000000.57692300.2338010.4526320.8007570.80.536885
250.0738610.4715970.6030530.0000000.1818180.5878790.6643210.3665960.8920160.880825...1.0000001.0000000.9393380.62820500.1079840.0753680.7823630.80.372951
260.0317130.5937830.8259540.0000000.3636360.9161620.7727640.2476660.951836NaN...NaNNaNNaN0.57692300.0481640.1555790.4308410.80.909836
270.7809910.3204720.3180660.0105110.2727270.0564650.6065820.2353110.765547NaN...0.3246750.324675NaN0.67948700.2344530.0587370.6266350.80.401639
280.2479320.9431940.5491090.0483480.1818180.2424240.5758510.3082810.460971NaN...0.3445380.5233000.6360290.57692300.5390290.1800000.6089180.80.978484
290.5082630.8156480.5857510.0062660.1818180.6161620.6183840.4085190.544441NaN...NaN0.4211500.4107140.73076900.4555590.1233680.5831150.80.340164
..................................................................
12440.2561970.4726690.8961830.0113110.0000000.1222220.6732180.3387600.2291890.212121...0.3246750.3246750.5416670.32051300.7708110.1913680.7539120.80.713115
12450.9018590.3922830.3882950.0000030.4545450.1313130.5416250.4171330.9048580.189610...0.7974030.5948050.7937500.34615400.0951420.0772630.5575820.60.185451
12460.0933860.3558410.6824430.0000130.8181820.6464650.6817520.3690460.6147140.282468...0.7889610.0292210.2695310.35897400.3852860.1656840.5320110.80.139344
12470.0032830.4758840.9450380.0000790.0000000.3494950.6841130.4701920.0768260.021944...0.0568740.1267350.0754310.29487200.9231740.1705260.6891650.80.401639
12480.1260310.3397640.8951650.0168170.1818180.2555560.7884700.3349920.3728330.815821...0.6316410.3553720.6250000.37179500.6271670.2189470.6011110.80.431352
12490.0020740.2111470.6580150.0056060.0000000.2353540.6584200.1466920.9341400.220779...0.2207790.2987010.2067310.28205100.0658600.1353680.8160290.60.407787
12500.3347090.2454450.7333330.0001150.8181820.9838380.7152970.1641830.3924500.212121...0.5497840.0995670.6562500.26923100.6075500.1050530.6960380.80.365779
12510.1818160.4501610.8768450.8488490.1818180.2414140.6880160.2990800.3728330.815821...0.6316410.3553720.6250000.28205100.6271670.1075790.6102020.80.542008
12520.9690080.4469450.2987280.0000330.4545450.1010100.5617790.3732080.9048580.189610...0.7974030.5948050.7937500.28205100.0951420.0698950.5846590.80.559426
12530.2024770.1961410.7231550.0004430.3636360.1222220.6672270.1415210.9341400.220779...0.2207790.2987010.2067310.26923100.0658600.1334740.8044090.60.461066
12540.2654940.4737410.7323160.0000730.8181820.1585860.6315480.5227100.4954670.146958...0.0136710.1736160.7014800.38461500.5045330.1566320.6357680.80.611680
12550.0437990.3751340.8432570.0006230.3636360.1121210.5047660.3584490.4704470.217237...0.2632820.1251480.2968750.33333300.5295530.4210530.6407040.80.431352
12560.4225190.3215430.5399490.0061360.1818180.1353540.6053110.3294470.7487380.298701...0.688312NaN0.1274040.33333300.2512620.0701050.4531670.80.253074
12570.0026010.2325830.7811700.0775780.1818180.1969700.7379480.2884100.3684840.145292...0.0186690.1769480.7421880.33333300.6315160.3221050.3750890.80.276639
12580.7148750.4351550.5633590.0021920.6363640.4000000.5031770.1765270.6399010.276438...1.0000000.5658630.5580360.29487200.3600990.1332630.6764220.80.657787
12590.0579520.5026800.9348600.0000700.3636360.8393940.6494780.6649660.8912460.263282...0.3553720.1711920.5312500.29487200.1087540.1545260.5483210.80.614754
12600.1342950.3515540.7190840.0857860.1818180.3727270.7088970.2279110.4646470.113636...NaN0.1136360.4843750.29487200.5353530.0913680.6860860.80.767418
12610.1415270.3933550.7435110.0000010.8181820.1828280.7441670.3078250.443151NaN...0.050325NaN0.7421880.29487200.5568490.3621050.5451470.80.597336
12620.7417350.2625940.4493640.9639640.1818180.1555560.4578760.7408180.3728330.815821...0.6316410.3553720.6250000.30769200.6271670.0861050.3056240.80.045799
12630.3316100.4962490.8849870.0000200.8181820.4262630.6534730.5165490.4954670.146958...0.0136710.1736160.7014800.26923100.5045330.3600000.6339630.80.400615
12640.7427680.3976420.5491090.0000380.3636360.3747470.5893330.1768430.6399010.276438...1.0000000.5658630.5580360.25641000.3600990.1111580.6783500.80.644467
12650.1983450.3612000.8870230.0001370.3636360.1313130.6741260.3341820.4704470.217237...0.2632820.1251480.2968750.29487200.5295530.2168420.6223860.80.517418
12660.6146680.2465170.7129770.0000950.8181820.2252530.7038580.2260980.443151NaN...0.050325NaN0.7421880.24359000.5568490.1324210.9559160.60.492828
12670.0730350.2829580.8860050.0225230.1818180.6010100.6526100.2342220.3684840.145292...0.0186690.1769480.7421880.24359000.6315160.1604210.6796050.80.431352
12680.3584700.6548770.7272260.0000000.0909090.1242420.6531550.2876470.6230950.348794...0.4935060.2040820.5580360.26923100.3769050.2736840.6564170.80.680328
12690.9607440.2454450.3791350.4294290.1818180.3565660.6215620.0514060.181364NaN...NaNNaNNaN0.33333300.8186360.0791580.3445540.80.057582
12700.0490680.5991430.8666670.0001760.4545450.7838380.8253740.7234280.6548750.140496...0.0484060.5702480.6562500.47435900.3451250.3578950.7403870.80.623975
12710.0113610.5391210.8625950.0038240.4545450.6050510.7548800.7075390.6548750.140496...0.0484060.5702480.6562500.37179500.3451250.3410530.7341530.80.564549
12720.1745850.4941050.6854960.0000010.7272730.1141410.7490240.4930600.8452930.099567...0.1839830.0151520.1979170.50000000.1547070.1286320.5254170.60.361680
12730.2097090.3386920.8819340.0000160.7272730.9939390.7587830.3070140.8452930.099567...0.1839830.0151520.1979170.38461500.1547070.2315790.5697180.60.282787
\n", "

1274 rows × 26 columns

\n", "
" ], "text/plain": [ " acousticness danceability energy instrumentalness key \\\n", "0 0.439048 0.848875 0.759796 0.627628 0.181818 \n", "1 0.140494 0.554126 0.594911 0.834835 0.181818 \n", "2 0.711776 0.248660 0.482952 0.938939 0.090909 \n", "3 0.065907 0.435155 0.696692 0.896897 0.363636 \n", "4 0.869834 0.364416 0.320102 0.833834 0.000000 \n", "5 0.220039 0.311897 0.975573 0.908909 0.545455 \n", "6 0.220039 0.311897 0.975573 0.908909 0.545455 \n", "7 0.146692 0.000000 0.000000 1.000000 0.909091 \n", "8 0.523759 0.622722 0.818830 0.000000 0.000000 \n", "9 0.268593 0.404073 0.915522 0.000000 0.363636 \n", "10 0.398759 0.525188 0.710941 0.000000 0.636364 \n", "11 0.093386 0.604502 0.824936 0.000004 0.636364 \n", "12 0.123965 0.712755 0.779135 0.000004 0.545455 \n", "13 0.065597 0.405145 0.672265 0.000000 0.181818 \n", "14 0.151857 0.505895 0.854453 0.000003 0.636364 \n", "15 0.145659 0.696677 0.774046 0.343343 0.181818 \n", "16 0.011568 0.413719 0.601018 0.000014 0.909091 \n", "17 0.004801 0.697749 0.891094 0.000003 0.363636 \n", "18 0.100618 0.778135 0.697710 0.000035 0.636364 \n", "19 0.487602 0.576635 0.469720 0.000000 1.000000 \n", "20 0.372932 0.275456 0.402545 0.000087 0.090909 \n", "21 0.184915 0.311897 0.632570 0.000000 0.818182 \n", "22 0.680784 0.578778 0.480916 0.001902 0.454545 \n", "23 0.651859 0.474812 0.393384 0.000000 0.000000 \n", "24 0.391527 0.553055 0.507379 0.000000 0.181818 \n", "25 0.073861 0.471597 0.603053 0.000000 0.181818 \n", "26 0.031713 0.593783 0.825954 0.000000 0.363636 \n", "27 0.780991 0.320472 0.318066 0.010511 0.272727 \n", "28 0.247932 0.943194 0.549109 0.048348 0.181818 \n", "29 0.508263 0.815648 0.585751 0.006266 0.181818 \n", "... ... ... ... ... ... \n", "1244 0.256197 0.472669 0.896183 0.011311 0.000000 \n", "1245 0.901859 0.392283 0.388295 0.000003 0.454545 \n", "1246 0.093386 0.355841 0.682443 0.000013 0.818182 \n", "1247 0.003283 0.475884 0.945038 0.000079 0.000000 \n", "1248 0.126031 0.339764 0.895165 0.016817 0.181818 \n", "1249 0.002074 0.211147 0.658015 0.005606 0.000000 \n", "1250 0.334709 0.245445 0.733333 0.000115 0.818182 \n", "1251 0.181816 0.450161 0.876845 0.848849 0.181818 \n", "1252 0.969008 0.446945 0.298728 0.000033 0.454545 \n", "1253 0.202477 0.196141 0.723155 0.000443 0.363636 \n", "1254 0.265494 0.473741 0.732316 0.000073 0.818182 \n", "1255 0.043799 0.375134 0.843257 0.000623 0.363636 \n", "1256 0.422519 0.321543 0.539949 0.006136 0.181818 \n", "1257 0.002601 0.232583 0.781170 0.077578 0.181818 \n", "1258 0.714875 0.435155 0.563359 0.002192 0.636364 \n", "1259 0.057952 0.502680 0.934860 0.000070 0.363636 \n", "1260 0.134295 0.351554 0.719084 0.085786 0.181818 \n", "1261 0.141527 0.393355 0.743511 0.000001 0.818182 \n", "1262 0.741735 0.262594 0.449364 0.963964 0.181818 \n", "1263 0.331610 0.496249 0.884987 0.000020 0.818182 \n", "1264 0.742768 0.397642 0.549109 0.000038 0.363636 \n", "1265 0.198345 0.361200 0.887023 0.000137 0.363636 \n", "1266 0.614668 0.246517 0.712977 0.000095 0.818182 \n", "1267 0.073035 0.282958 0.886005 0.022523 0.181818 \n", "1268 0.358470 0.654877 0.727226 0.000000 0.090909 \n", "1269 0.960744 0.245445 0.379135 0.429429 0.181818 \n", "1270 0.049068 0.599143 0.866667 0.000176 0.454545 \n", "1271 0.011361 0.539121 0.862595 0.003824 0.454545 \n", "1272 0.174585 0.494105 0.685496 0.000001 0.727273 \n", "1273 0.209709 0.338692 0.881934 0.000016 0.727273 \n", "\n", " liveness loudness lyrical_density neg nnrc_anger ... \\\n", "0 0.796970 0.557149 0.057550 0.360472 NaN ... \n", "1 0.120202 0.608488 0.007807 0.622129 NaN ... \n", "2 0.066869 0.732229 0.006360 0.537797 NaN ... \n", "3 0.231313 0.780844 0.000207 0.537797 NaN ... \n", "4 0.113131 0.254290 0.007784 0.687087 NaN ... \n", "5 0.335354 0.576714 0.000000 0.537797 NaN ... \n", "6 0.335354 0.576714 0.000000 0.537797 NaN ... \n", "7 0.000000 0.368634 0.083879 0.504531 NaN ... \n", "8 0.109091 0.865729 0.365594 0.532734 NaN ... \n", "9 0.074747 0.919655 0.556854 0.536918 0.366883 ... \n", "10 0.314141 0.812982 0.443268 0.520248 NaN ... \n", "11 0.128283 0.743078 0.408980 0.369287 NaN ... \n", "12 0.126263 0.681843 0.289830 0.726211 NaN ... \n", "13 0.184848 0.673128 0.608999 0.596823 0.392208 ... \n", "14 0.601010 0.791148 0.482591 0.605477 NaN ... \n", "15 0.217172 0.838629 0.332965 0.608455 0.263282 ... \n", "16 0.088889 0.715343 0.181030 0.595786 0.392208 ... \n", "17 0.128283 0.767499 0.614553 0.629600 0.240260 ... \n", "18 0.241414 0.576577 0.305533 0.759113 NaN ... \n", "19 0.171717 0.658284 0.271464 0.574013 0.662338 ... \n", "20 0.070909 0.529778 0.323723 0.241276 0.189610 ... \n", "21 0.102020 0.708352 0.229548 0.759066 NaN ... \n", "22 0.461616 0.507626 0.390046 0.354775 1.000000 ... \n", "23 0.112121 0.686337 0.345151 0.022371 0.610390 ... \n", "24 0.104040 0.518566 0.474655 0.766199 0.324675 ... \n", "25 0.587879 0.664321 0.366596 0.892016 0.880825 ... \n", "26 0.916162 0.772764 0.247666 0.951836 NaN ... \n", "27 0.056465 0.606582 0.235311 0.765547 NaN ... \n", "28 0.242424 0.575851 0.308281 0.460971 NaN ... \n", "29 0.616162 0.618384 0.408519 0.544441 NaN ... \n", "... ... ... ... ... ... ... \n", "1244 0.122222 0.673218 0.338760 0.229189 0.212121 ... \n", "1245 0.131313 0.541625 0.417133 0.904858 0.189610 ... \n", "1246 0.646465 0.681752 0.369046 0.614714 0.282468 ... \n", "1247 0.349495 0.684113 0.470192 0.076826 0.021944 ... \n", "1248 0.255556 0.788470 0.334992 0.372833 0.815821 ... \n", "1249 0.235354 0.658420 0.146692 0.934140 0.220779 ... \n", "1250 0.983838 0.715297 0.164183 0.392450 0.212121 ... \n", "1251 0.241414 0.688016 0.299080 0.372833 0.815821 ... \n", "1252 0.101010 0.561779 0.373208 0.904858 0.189610 ... \n", "1253 0.122222 0.667227 0.141521 0.934140 0.220779 ... \n", "1254 0.158586 0.631548 0.522710 0.495467 0.146958 ... \n", "1255 0.112121 0.504766 0.358449 0.470447 0.217237 ... \n", "1256 0.135354 0.605311 0.329447 0.748738 0.298701 ... \n", "1257 0.196970 0.737948 0.288410 0.368484 0.145292 ... \n", "1258 0.400000 0.503177 0.176527 0.639901 0.276438 ... \n", "1259 0.839394 0.649478 0.664966 0.891246 0.263282 ... \n", "1260 0.372727 0.708897 0.227911 0.464647 0.113636 ... \n", "1261 0.182828 0.744167 0.307825 0.443151 NaN ... \n", "1262 0.155556 0.457876 0.740818 0.372833 0.815821 ... \n", "1263 0.426263 0.653473 0.516549 0.495467 0.146958 ... \n", "1264 0.374747 0.589333 0.176843 0.639901 0.276438 ... \n", "1265 0.131313 0.674126 0.334182 0.470447 0.217237 ... \n", "1266 0.225253 0.703858 0.226098 0.443151 NaN ... \n", "1267 0.601010 0.652610 0.234222 0.368484 0.145292 ... \n", "1268 0.124242 0.653155 0.287647 0.623095 0.348794 ... \n", "1269 0.356566 0.621562 0.051406 0.181364 NaN ... \n", "1270 0.783838 0.825374 0.723428 0.654875 0.140496 ... \n", "1271 0.605051 0.754880 0.707539 0.654875 0.140496 ... \n", "1272 0.114141 0.749024 0.493060 0.845293 0.099567 ... \n", "1273 0.993939 0.758783 0.307014 0.845293 0.099567 ... \n", "\n", " nnrc_sadness nnrc_surprise nnrc_trust popularity popularity0 \\\n", "0 NaN NaN NaN 0.564103 0 \n", "1 NaN NaN NaN 0.397436 0 \n", "2 NaN NaN NaN 0.435897 0 \n", "3 NaN NaN NaN 0.346154 0 \n", "4 NaN NaN NaN 0.576923 0 \n", "5 NaN NaN NaN 0.435897 0 \n", "6 NaN NaN NaN 0.410256 0 \n", "7 NaN NaN NaN 0.000000 0 \n", "8 NaN 0.155844 0.140625 0.692308 0 \n", "9 0.620130 NaN 0.097656 0.743590 0 \n", "10 NaN NaN 0.312500 0.820513 0 \n", "11 NaN 0.046600 0.090074 0.692308 0 \n", "12 NaN 0.276438 0.705357 0.756410 0 \n", "13 0.189610 0.290909 0.278125 0.743590 0 \n", "14 NaN 0.064935 0.048077 0.679487 0 \n", "15 0.079103 0.171192 0.625000 0.679487 0 \n", "16 0.594805 NaN 0.381250 0.871795 0 \n", "17 0.113636 0.746753 0.484375 0.666667 0 \n", "18 NaN 0.276438 0.410714 0.628205 0 \n", "19 NaN NaN NaN 0.602564 0 \n", "20 0.324675 0.189610 0.862500 0.743590 0 \n", "21 NaN 1.000000 0.484375 0.615385 0 \n", "22 NaN NaN NaN 0.564103 0 \n", "23 0.376623 0.064935 0.682692 0.897436 0 \n", "24 0.324675 0.324675 1.000000 0.576923 0 \n", "25 1.000000 1.000000 0.939338 0.628205 0 \n", "26 NaN NaN NaN 0.576923 0 \n", "27 0.324675 0.324675 NaN 0.679487 0 \n", "28 0.344538 0.523300 0.636029 0.576923 0 \n", "29 NaN 0.421150 0.410714 0.730769 0 \n", "... ... ... ... ... ... \n", "1244 0.324675 0.324675 0.541667 0.320513 0 \n", "1245 0.797403 0.594805 0.793750 0.346154 0 \n", "1246 0.788961 0.029221 0.269531 0.358974 0 \n", "1247 0.056874 0.126735 0.075431 0.294872 0 \n", "1248 0.631641 0.355372 0.625000 0.371795 0 \n", "1249 0.220779 0.298701 0.206731 0.282051 0 \n", "1250 0.549784 0.099567 0.656250 0.269231 0 \n", "1251 0.631641 0.355372 0.625000 0.282051 0 \n", "1252 0.797403 0.594805 0.793750 0.282051 0 \n", "1253 0.220779 0.298701 0.206731 0.269231 0 \n", "1254 0.013671 0.173616 0.701480 0.384615 0 \n", "1255 0.263282 0.125148 0.296875 0.333333 0 \n", "1256 0.688312 NaN 0.127404 0.333333 0 \n", "1257 0.018669 0.176948 0.742188 0.333333 0 \n", "1258 1.000000 0.565863 0.558036 0.294872 0 \n", "1259 0.355372 0.171192 0.531250 0.294872 0 \n", "1260 NaN 0.113636 0.484375 0.294872 0 \n", "1261 0.050325 NaN 0.742188 0.294872 0 \n", "1262 0.631641 0.355372 0.625000 0.307692 0 \n", "1263 0.013671 0.173616 0.701480 0.269231 0 \n", "1264 1.000000 0.565863 0.558036 0.256410 0 \n", "1265 0.263282 0.125148 0.296875 0.294872 0 \n", "1266 0.050325 NaN 0.742188 0.243590 0 \n", "1267 0.018669 0.176948 0.742188 0.243590 0 \n", "1268 0.493506 0.204082 0.558036 0.269231 0 \n", "1269 NaN NaN NaN 0.333333 0 \n", "1270 0.048406 0.570248 0.656250 0.474359 0 \n", "1271 0.048406 0.570248 0.656250 0.371795 0 \n", "1272 0.183983 0.015152 0.197917 0.500000 0 \n", "1273 0.183983 0.015152 0.197917 0.384615 0 \n", "\n", " pos speechiness tempo time_signature valence \n", "0 0.639528 0.106526 0.425867 0.8 0.155738 \n", "1 0.377871 0.086947 0.335331 0.8 0.760246 \n", "2 0.462203 0.069263 0.329760 0.6 0.156762 \n", "3 0.462203 0.063368 0.739918 0.8 0.478484 \n", "4 0.312913 0.068421 0.671074 0.8 0.682377 \n", "5 0.462203 0.117263 0.854050 0.8 0.130123 \n", "6 0.462203 0.117263 0.854050 0.8 0.130123 \n", "7 0.495469 0.000000 0.000000 0.0 0.000000 \n", "8 0.467266 0.066947 0.644934 0.8 0.991803 \n", "9 0.463082 0.101263 0.357808 0.8 0.934426 \n", "10 0.479752 0.100211 0.619264 0.8 0.887295 \n", "11 0.630713 0.059579 0.425615 0.8 0.934426 \n", "12 0.273789 0.064632 0.651131 0.8 0.748975 \n", "13 0.403177 0.062737 0.502470 0.8 0.663934 \n", "14 0.394523 0.118316 0.372792 0.8 0.861680 \n", "15 0.391545 0.061895 0.519226 0.8 0.588115 \n", "16 0.404214 0.054947 0.697336 0.8 0.545082 \n", "17 0.370400 0.065895 0.640974 0.8 0.985656 \n", "18 0.240887 0.067368 0.516885 0.8 0.953893 \n", "19 0.425987 0.081684 0.279509 0.8 0.539959 \n", "20 0.758724 0.060421 0.720638 0.8 0.879098 \n", "21 0.240934 0.116632 0.877479 0.6 0.537910 \n", "22 0.645225 0.246316 0.743869 0.6 0.686475 \n", "23 0.977629 0.067789 0.679596 0.8 0.420082 \n", "24 0.233801 0.452632 0.800757 0.8 0.536885 \n", "25 0.107984 0.075368 0.782363 0.8 0.372951 \n", "26 0.048164 0.155579 0.430841 0.8 0.909836 \n", "27 0.234453 0.058737 0.626635 0.8 0.401639 \n", "28 0.539029 0.180000 0.608918 0.8 0.978484 \n", "29 0.455559 0.123368 0.583115 0.8 0.340164 \n", "... ... ... ... ... ... \n", "1244 0.770811 0.191368 0.753912 0.8 0.713115 \n", "1245 0.095142 0.077263 0.557582 0.6 0.185451 \n", "1246 0.385286 0.165684 0.532011 0.8 0.139344 \n", "1247 0.923174 0.170526 0.689165 0.8 0.401639 \n", "1248 0.627167 0.218947 0.601111 0.8 0.431352 \n", "1249 0.065860 0.135368 0.816029 0.6 0.407787 \n", "1250 0.607550 0.105053 0.696038 0.8 0.365779 \n", "1251 0.627167 0.107579 0.610202 0.8 0.542008 \n", "1252 0.095142 0.069895 0.584659 0.8 0.559426 \n", "1253 0.065860 0.133474 0.804409 0.6 0.461066 \n", "1254 0.504533 0.156632 0.635768 0.8 0.611680 \n", "1255 0.529553 0.421053 0.640704 0.8 0.431352 \n", "1256 0.251262 0.070105 0.453167 0.8 0.253074 \n", "1257 0.631516 0.322105 0.375089 0.8 0.276639 \n", "1258 0.360099 0.133263 0.676422 0.8 0.657787 \n", "1259 0.108754 0.154526 0.548321 0.8 0.614754 \n", "1260 0.535353 0.091368 0.686086 0.8 0.767418 \n", "1261 0.556849 0.362105 0.545147 0.8 0.597336 \n", "1262 0.627167 0.086105 0.305624 0.8 0.045799 \n", "1263 0.504533 0.360000 0.633963 0.8 0.400615 \n", "1264 0.360099 0.111158 0.678350 0.8 0.644467 \n", "1265 0.529553 0.216842 0.622386 0.8 0.517418 \n", "1266 0.556849 0.132421 0.955916 0.6 0.492828 \n", "1267 0.631516 0.160421 0.679605 0.8 0.431352 \n", "1268 0.376905 0.273684 0.656417 0.8 0.680328 \n", "1269 0.818636 0.079158 0.344554 0.8 0.057582 \n", "1270 0.345125 0.357895 0.740387 0.8 0.623975 \n", "1271 0.345125 0.341053 0.734153 0.8 0.564549 \n", "1272 0.154707 0.128632 0.525417 0.6 0.361680 \n", "1273 0.154707 0.231579 0.569718 0.6 0.282787 \n", "\n", "[1274 rows x 26 columns]" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "all_pre_df" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Columns to drop\n", "Find the covariances of the different metrics, and use that to drop the columns with the highest covariances. " ] }, { "cell_type": "code", "execution_count": 21, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/plain": [ "nnrc_positive speechiness 0.000007\n", "key time_signature 0.000020\n", "lyrical_density neg 0.000029\n", " pos 0.000029\n", "danceability nnrc_anger 0.000056\n", "lyrical_density tempo 0.000064\n", " popularity 0.000105\n", "nnrc_trust speechiness 0.000106\n", "tempo time_signature 0.000118\n", "key lyrical_density 0.000121\n", "neg time_signature 0.000125\n", "pos time_signature 0.000125\n", "popularity time_signature 0.000177\n", "instrumentalness key 0.000184\n", "pos speechiness 0.000223\n", "neg speechiness 0.000223\n", "danceability loudness 0.000223\n", "nnrc_sadness time_signature 0.000224\n", "nnrc_fear speechiness 0.000230\n", "nnrc_anticipation tempo 0.000233\n", "nnrc_surprise valence 0.000275\n", "key liveness 0.000287\n", "instrumentalness nnrc_sadness 0.000307\n", "danceability nnrc_trust 0.000310\n", "key nnrc_surprise 0.000311\n", "nnrc_anticipation time_signature 0.000340\n", "danceability nnrc_disgust 0.000340\n", "nnrc_surprise time_signature 0.000406\n", "nnrc_anger nnrc_anticipation 0.000430\n", "nnrc_joy popularity 0.000440\n", " ... \n", "nnrc_anger nnrc_joy 0.016843\n", "acousticness loudness 0.017604\n", "nnrc_anticipation nnrc_surprise 0.020270\n", "nnrc_negative pos 0.020766\n", "neg nnrc_negative 0.020766\n", "energy liveness 0.021465\n", " loudness 0.021603\n", "nnrc_anger nnrc_positive 0.021650\n", "danceability valence 0.021851\n", "nnrc_fear nnrc_joy 0.022436\n", "nnrc_joy nnrc_trust 0.022657\n", "nnrc_disgust nnrc_sadness 0.023548\n", "nnrc_positive nnrc_sadness 0.023987\n", "nnrc_disgust nnrc_fear 0.024015\n", "nnrc_anger nnrc_sadness 0.024682\n", "nnrc_anticipation nnrc_trust 0.024758\n", "nnrc_positive nnrc_trust 0.025335\n", "nnrc_anger nnrc_disgust 0.027317\n", "nnrc_fear nnrc_positive 0.028906\n", " nnrc_sadness 0.033809\n", "nnrc_disgust nnrc_negative 0.036158\n", "nnrc_anger nnrc_fear 0.036829\n", " nnrc_negative 0.037309\n", "acousticness energy 0.037537\n", "nnrc_joy nnrc_negative 0.039953\n", "nnrc_fear nnrc_negative 0.045366\n", "neg pos 0.045879\n", "nnrc_negative nnrc_positive 0.051093\n", " nnrc_sadness 0.051542\n", "nnrc_joy nnrc_positive 0.053021\n", "Length: 300, dtype: float64" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" } ], "source": [ "adc = all_pre_df.drop(['popularity0'], axis=1).cov().stack().abs().sort_values()\n", "adc[adc.index.get_level_values(0) < adc.index.get_level_values(1)]" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [], "source": [ "columns_to_drop = ['popularity0', 'nnrc_positive', 'nnrc_negative', \n", " 'pos', 'neg', 'energy',\n", " 'time_signature', 'speechiness', 'instrumentalness', 'liveness',\n", " 'popularity', 'loudness', 'lyrical_density']" ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/plain": [ "danceability nnrc_anger 0.000056\n", "nnrc_anticipation tempo 0.000233\n", "nnrc_surprise valence 0.000275\n", "danceability nnrc_trust 0.000310\n", "key nnrc_surprise 0.000311\n", "danceability nnrc_disgust 0.000340\n", "nnrc_anger nnrc_anticipation 0.000430\n", "danceability nnrc_fear 0.000506\n", " nnrc_surprise 0.000510\n", "key nnrc_disgust 0.000525\n", "nnrc_anger tempo 0.000574\n", "nnrc_trust tempo 0.000600\n", "nnrc_sadness tempo 0.000880\n", "nnrc_disgust tempo 0.000897\n", "key nnrc_fear 0.000939\n", " tempo 0.001029\n", " valence 0.001139\n", "acousticness nnrc_trust 0.001177\n", "danceability key 0.001214\n", "key nnrc_anticipation 0.001242\n", "acousticness nnrc_disgust 0.001261\n", "nnrc_surprise tempo 0.001318\n", "nnrc_disgust nnrc_trust 0.001386\n", "nnrc_fear tempo 0.001460\n", "nnrc_anticipation nnrc_sadness 0.001786\n", "acousticness nnrc_surprise 0.001940\n", "key nnrc_trust 0.001954\n", "acousticness danceability 0.002006\n", "nnrc_disgust valence 0.002225\n", "acousticness nnrc_anger 0.002381\n", " ... \n", "nnrc_joy tempo 0.003588\n", "acousticness nnrc_fear 0.003706\n", "danceability tempo 0.005166\n", "nnrc_fear valence 0.005228\n", " nnrc_trust 0.005561\n", "key nnrc_sadness 0.006327\n", "nnrc_disgust nnrc_surprise 0.006644\n", "acousticness tempo 0.007438\n", "nnrc_anger nnrc_surprise 0.007465\n", "nnrc_fear nnrc_surprise 0.007833\n", "nnrc_joy nnrc_surprise 0.008925\n", "acousticness nnrc_joy 0.009298\n", "nnrc_sadness nnrc_surprise 0.009950\n", "acousticness valence 0.011231\n", "nnrc_surprise nnrc_trust 0.012670\n", "nnrc_anticipation nnrc_joy 0.013639\n", "nnrc_disgust nnrc_joy 0.013900\n", "nnrc_joy nnrc_sadness 0.014611\n", "nnrc_anger nnrc_joy 0.016843\n", "nnrc_anticipation nnrc_surprise 0.020270\n", "danceability valence 0.021851\n", "nnrc_fear nnrc_joy 0.022436\n", "nnrc_joy nnrc_trust 0.022657\n", "nnrc_disgust nnrc_sadness 0.023548\n", " nnrc_fear 0.024015\n", "nnrc_anger nnrc_sadness 0.024682\n", "nnrc_anticipation nnrc_trust 0.024758\n", "nnrc_anger nnrc_disgust 0.027317\n", "nnrc_fear nnrc_sadness 0.033809\n", "nnrc_anger nnrc_fear 0.036829\n", "Length: 78, dtype: float64" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "adc = all_pre_df.drop(columns_to_drop, axis=1).cov().stack().abs().sort_values()\n", "adc[adc.index.get_level_values(0) < adc.index.get_level_values(1)]" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
countmeanstdmin25%50%75%max
acousticness1274.00.2561500.2719140.00.0246880.1503080.4070231.0
danceability1274.00.5087710.1680090.00.3850480.5058950.6248661.0
key1274.00.4480520.3178170.00.1818180.4545450.7272731.0
nnrc_anger1005.00.3026030.2314910.00.1317250.2402600.4155841.0
nnrc_anticipation1147.00.4626750.2880160.00.2250000.4259260.6555561.0
nnrc_disgust886.00.2699330.2251520.00.1061880.1896100.3766231.0
nnrc_fear1082.00.3647790.2705970.00.1389610.2987010.4935061.0
nnrc_joy1186.00.5715780.2932770.00.3086420.6011400.8271601.0
nnrc_sadness1082.00.3936880.2654990.00.1896100.3246750.5658631.0
nnrc_surprise994.00.2832970.2113390.00.1220780.2282000.4009901.0
nnrc_trust1174.00.4238060.2700940.00.2067310.3812500.5989581.0
tempo1274.00.5824510.1444020.00.4800080.5749150.6679051.0
valence1274.00.5379960.2539150.00.3401640.5420080.7407791.0
\n", "
" ], "text/plain": [ " count mean std min 25% 50% \\\n", "acousticness 1274.0 0.256150 0.271914 0.0 0.024688 0.150308 \n", "danceability 1274.0 0.508771 0.168009 0.0 0.385048 0.505895 \n", "key 1274.0 0.448052 0.317817 0.0 0.181818 0.454545 \n", "nnrc_anger 1005.0 0.302603 0.231491 0.0 0.131725 0.240260 \n", "nnrc_anticipation 1147.0 0.462675 0.288016 0.0 0.225000 0.425926 \n", "nnrc_disgust 886.0 0.269933 0.225152 0.0 0.106188 0.189610 \n", "nnrc_fear 1082.0 0.364779 0.270597 0.0 0.138961 0.298701 \n", "nnrc_joy 1186.0 0.571578 0.293277 0.0 0.308642 0.601140 \n", "nnrc_sadness 1082.0 0.393688 0.265499 0.0 0.189610 0.324675 \n", "nnrc_surprise 994.0 0.283297 0.211339 0.0 0.122078 0.228200 \n", "nnrc_trust 1174.0 0.423806 0.270094 0.0 0.206731 0.381250 \n", "tempo 1274.0 0.582451 0.144402 0.0 0.480008 0.574915 \n", "valence 1274.0 0.537996 0.253915 0.0 0.340164 0.542008 \n", "\n", " 75% max \n", "acousticness 0.407023 1.0 \n", "danceability 0.624866 1.0 \n", "key 0.727273 1.0 \n", "nnrc_anger 0.415584 1.0 \n", "nnrc_anticipation 0.655556 1.0 \n", "nnrc_disgust 0.376623 1.0 \n", "nnrc_fear 0.493506 1.0 \n", "nnrc_joy 0.827160 1.0 \n", "nnrc_sadness 0.565863 1.0 \n", "nnrc_surprise 0.400990 1.0 \n", "nnrc_trust 0.598958 1.0 \n", "tempo 0.667905 1.0 \n", "valence 0.740779 1.0 " ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "pipeline = [\n", " {'$match': {'lyrics': {'$exists': True}, 'sentiment': {'$exists': True}, 'valence': {'$exists': True}}},\n", " {'$project': projection_dict}\n", "]\n", "all_raw_df = pd.DataFrame(list(tracks.aggregate(pipeline)))\n", "all_raw_df.drop(columns_to_drop, axis=1, inplace=True)\n", "\n", "all_df=(all_raw_df-all_raw_df.min())/(all_raw_df.max()-all_raw_df.min())\n", "all_df.popularity0 = 0\n", "all_df.describe().T" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgEAAAD8CAYAAADudXePAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ndjUgtg6iE0MAAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "ax = all_df.plot.kde()\n", "ax.legend(loc='center left', bbox_to_anchor=(1, 0.5))" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAesAAAD8CAYAAABXV4w2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "ax = all_df[['acousticness']].plot.kde()\n", "ax.legend(loc='center left', bbox_to_anchor=(1, 0.5))" ] }, { "cell_type": "code", "execution_count": 27, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
acousticnessdanceabilitykeynnrc_angernnrc_anticipationnnrc_disgustnnrc_fearnnrc_joynnrc_sadnessnnrc_surprisennrc_trusttempovalence
00.4390480.8488750.181818NaNNaNNaNNaN0.222222NaNNaNNaN0.4258670.155738
10.1404940.5541260.181818NaN1.000000NaNNaNNaNNaNNaNNaN0.3353310.760246
20.7117760.2486600.090909NaNNaNNaNNaNNaNNaNNaNNaN0.3297600.156762
30.0659070.4351550.363636NaNNaNNaNNaNNaNNaNNaNNaN0.7399180.478484
40.8698340.3644160.000000NaNNaNNaNNaNNaNNaNNaNNaN0.6710740.682377
50.2200390.3118970.545455NaNNaNNaNNaNNaNNaNNaNNaN0.8540500.130123
60.2200390.3118970.545455NaNNaNNaNNaNNaNNaNNaNNaN0.8540500.130123
70.1466920.0000000.909091NaNNaNNaNNaNNaNNaNNaNNaN0.0000000.000000
80.5237590.6227220.000000NaN0.483333NaNNaN1.000000NaN0.1558440.1406250.6449340.991803
90.2685930.4040730.3636360.3668830.3541670.1769480.3668830.8055560.620130NaN0.0976560.3578080.934426
100.3987590.5251880.636364NaN0.311111NaN1.0000000.654321NaNNaN0.3125000.6192640.887295
110.0933860.6045020.636364NaN0.452941NaNNaN1.000000NaN0.0466000.0900740.4256150.934426
120.1239650.7127550.545455NaN0.704762NaNNaN0.703704NaN0.2764380.7053570.6511310.748975
130.0655970.4051450.1818180.3922081.000000NaN0.0883120.3777780.1896100.2909090.2781250.5024700.663934
140.1518570.5058950.636364NaN0.0461540.0649350.025974NaNNaN0.0649350.0480770.3727920.861680
150.1456590.6966770.1818180.2632820.4363640.079103NaN0.5286200.0791030.1711920.6250000.5192260.588115
160.0115680.4137190.9090910.3922080.3800000.5948050.7974030.1703700.594805NaN0.3812500.6973360.545082
170.0048010.6977490.3636360.2402600.6125000.2402600.1136360.7407410.1136360.7467530.4843750.6409740.985656
180.1006180.7781350.636364NaN0.409524NaNNaN0.407407NaN0.2764380.4107140.5168850.953893
190.4876020.5766351.0000000.662338NaN0.324675NaNNaNNaNNaNNaN0.2795090.539959
200.3729320.2754560.0909090.1896100.311111NaN1.0000000.4469140.3246750.1896100.8625000.7206380.879098
210.1849150.3118970.818182NaN1.000000NaN1.0000000.481481NaN1.0000000.4843750.8774790.537910
220.6807840.5787780.4545451.000000NaNNaNNaNNaNNaNNaNNaN0.7438690.686475
230.6518590.4748120.0000000.6103900.205128NaN0.0649350.2022790.3766230.0649350.6826920.6795960.420082
240.3915270.5530550.1818180.3246750.6555560.662338NaN1.0000000.3246750.3246751.0000000.8007570.536885
250.0738610.4715970.1818180.8808251.0000000.8808251.0000001.0000001.0000001.0000000.9393380.7823630.372951
260.0317130.5937830.363636NaNNaNNaNNaN0.740741NaNNaNNaN0.4308410.909836
270.7809910.3204720.272727NaN1.000000NaN0.099567NaN0.3246750.324675NaN0.6266350.401639
280.2479320.9431940.181818NaN0.513725NaN0.0466001.0000000.3445380.5233000.6360290.6089180.978484
290.5082630.8156480.181818NaN1.000000NaNNaN0.407407NaN0.4211500.4107140.5831150.340164
..........................................
12440.2561970.4726690.0000000.2121210.5407410.4372290.0995670.4238680.3246750.3246750.5416670.7539120.713115
12450.9018590.3922830.4545450.1896100.5866670.1896100.1896100.7925930.7974030.5948050.7937500.5575820.185451
12460.0933860.3558410.8181820.2824680.3972220.4090910.3246750.4814810.7889610.0292210.2695310.5320110.139344
12470.0032830.4758840.0000000.0219440.4655170.0219440.0219440.8927200.0568740.1267350.0754310.6891650.401639
12480.1260310.3397640.1818180.8158210.2484850.3553720.8158210.4343430.6316410.3553720.6250000.6011110.431352
12490.0020740.2111470.0000000.2207790.3641030.7662340.142857NaN0.2207790.2987010.2067310.8160290.407787
12500.3347090.2454450.8181820.2121210.655556NaN0.2121210.7695470.5497840.0995670.6562500.6960380.365779
12510.1818160.4501610.1818180.8158210.2484850.3553720.8158210.4343430.6316410.3553720.6250000.6102020.542008
12520.9690080.4469450.4545450.1896100.5866670.1896100.1896100.7925930.7974030.5948050.7937500.5846590.559426
12530.2024770.1961410.3636360.2207790.3641030.7662340.142857NaN0.2207790.2987010.2067310.8044090.461066
12540.2654940.4737410.8181820.1469580.836842NaN0.0403280.8362570.0136710.1736160.7014800.6357680.611680
12550.0437990.3751340.3636360.2172370.4363640.8618650.4014170.0572390.2632820.1251480.2968750.6407040.431352
12560.4225190.3215430.1818180.2987010.0461540.2207790.6883120.2022790.688312NaN0.1274040.4531670.253074
12570.0026010.2325830.1818180.1452920.838542NaN0.0503250.8703700.0186690.1769480.7421880.3750890.276639
12580.7148750.4351550.6363640.2764380.557143NaN0.2764380.7037041.0000000.5658630.5580360.6764220.657787
12590.0579520.5026800.3636360.2632820.5303030.171192NaN0.3400670.3553720.1711920.5312500.5483210.614754
12600.1342950.3515540.1818180.1136360.6125000.1136360.1136360.481481NaN0.1136360.4843750.6860860.767418
12610.1415270.3933550.818182NaN0.0958330.556818NaN0.0925930.050325NaN0.7421880.5451470.597336
12620.7417350.2625940.1818180.8158210.2484850.3553720.8158210.4343430.6316410.3553720.6250000.3056240.045799
12630.3316100.4962490.8181820.1469580.836842NaN0.0403280.8362570.0136710.1736160.7014800.6339630.400615
12640.7427680.3976420.3636360.2764380.557143NaN0.2764380.7037041.0000000.5658630.5580360.6783500.644467
12650.1983450.3612000.3636360.2172370.4363640.8618650.4014170.0572390.2632820.1251480.2968750.6223860.517418
12660.6146680.2465170.818182NaN0.0958330.556818NaN0.0925930.050325NaN0.7421880.9559160.492828
12670.0730350.2829580.1818180.1452920.838542NaN0.0503250.8703700.0186690.1769480.7421880.6796050.431352
12680.3584700.6548770.0909090.3487940.3357140.6382190.3487940.4814810.4935060.2040820.5580360.6564170.680328
12690.9607440.2454450.181818NaN0.311111NaNNaNNaNNaNNaNNaN0.3445540.057582
12700.0490680.5991430.4545450.1404961.000000NaN0.0791030.6228960.0484060.5702480.6562500.7403870.623975
12710.0113610.5391210.4545450.1404961.000000NaN0.0791030.6228960.0484060.5702480.6562500.7341530.564549
12720.1745850.4941050.7272730.0995670.0527780.1277060.0995670.7407410.1839830.0151520.1979170.5254170.361680
12730.2097090.3386920.7272730.0995670.0527780.1277060.0995670.7407410.1839830.0151520.1979170.5697180.282787
\n", "

1274 rows × 13 columns

\n", "
" ], "text/plain": [ " acousticness danceability key nnrc_anger nnrc_anticipation \\\n", "0 0.439048 0.848875 0.181818 NaN NaN \n", "1 0.140494 0.554126 0.181818 NaN 1.000000 \n", "2 0.711776 0.248660 0.090909 NaN NaN \n", "3 0.065907 0.435155 0.363636 NaN NaN \n", "4 0.869834 0.364416 0.000000 NaN NaN \n", "5 0.220039 0.311897 0.545455 NaN NaN \n", "6 0.220039 0.311897 0.545455 NaN NaN \n", "7 0.146692 0.000000 0.909091 NaN NaN \n", "8 0.523759 0.622722 0.000000 NaN 0.483333 \n", "9 0.268593 0.404073 0.363636 0.366883 0.354167 \n", "10 0.398759 0.525188 0.636364 NaN 0.311111 \n", "11 0.093386 0.604502 0.636364 NaN 0.452941 \n", "12 0.123965 0.712755 0.545455 NaN 0.704762 \n", "13 0.065597 0.405145 0.181818 0.392208 1.000000 \n", "14 0.151857 0.505895 0.636364 NaN 0.046154 \n", "15 0.145659 0.696677 0.181818 0.263282 0.436364 \n", "16 0.011568 0.413719 0.909091 0.392208 0.380000 \n", "17 0.004801 0.697749 0.363636 0.240260 0.612500 \n", "18 0.100618 0.778135 0.636364 NaN 0.409524 \n", "19 0.487602 0.576635 1.000000 0.662338 NaN \n", "20 0.372932 0.275456 0.090909 0.189610 0.311111 \n", "21 0.184915 0.311897 0.818182 NaN 1.000000 \n", "22 0.680784 0.578778 0.454545 1.000000 NaN \n", "23 0.651859 0.474812 0.000000 0.610390 0.205128 \n", "24 0.391527 0.553055 0.181818 0.324675 0.655556 \n", "25 0.073861 0.471597 0.181818 0.880825 1.000000 \n", "26 0.031713 0.593783 0.363636 NaN NaN \n", "27 0.780991 0.320472 0.272727 NaN 1.000000 \n", "28 0.247932 0.943194 0.181818 NaN 0.513725 \n", "29 0.508263 0.815648 0.181818 NaN 1.000000 \n", "... ... ... ... ... ... \n", "1244 0.256197 0.472669 0.000000 0.212121 0.540741 \n", "1245 0.901859 0.392283 0.454545 0.189610 0.586667 \n", "1246 0.093386 0.355841 0.818182 0.282468 0.397222 \n", "1247 0.003283 0.475884 0.000000 0.021944 0.465517 \n", "1248 0.126031 0.339764 0.181818 0.815821 0.248485 \n", "1249 0.002074 0.211147 0.000000 0.220779 0.364103 \n", "1250 0.334709 0.245445 0.818182 0.212121 0.655556 \n", "1251 0.181816 0.450161 0.181818 0.815821 0.248485 \n", "1252 0.969008 0.446945 0.454545 0.189610 0.586667 \n", "1253 0.202477 0.196141 0.363636 0.220779 0.364103 \n", "1254 0.265494 0.473741 0.818182 0.146958 0.836842 \n", "1255 0.043799 0.375134 0.363636 0.217237 0.436364 \n", "1256 0.422519 0.321543 0.181818 0.298701 0.046154 \n", "1257 0.002601 0.232583 0.181818 0.145292 0.838542 \n", "1258 0.714875 0.435155 0.636364 0.276438 0.557143 \n", "1259 0.057952 0.502680 0.363636 0.263282 0.530303 \n", "1260 0.134295 0.351554 0.181818 0.113636 0.612500 \n", "1261 0.141527 0.393355 0.818182 NaN 0.095833 \n", "1262 0.741735 0.262594 0.181818 0.815821 0.248485 \n", "1263 0.331610 0.496249 0.818182 0.146958 0.836842 \n", "1264 0.742768 0.397642 0.363636 0.276438 0.557143 \n", "1265 0.198345 0.361200 0.363636 0.217237 0.436364 \n", "1266 0.614668 0.246517 0.818182 NaN 0.095833 \n", "1267 0.073035 0.282958 0.181818 0.145292 0.838542 \n", "1268 0.358470 0.654877 0.090909 0.348794 0.335714 \n", "1269 0.960744 0.245445 0.181818 NaN 0.311111 \n", "1270 0.049068 0.599143 0.454545 0.140496 1.000000 \n", "1271 0.011361 0.539121 0.454545 0.140496 1.000000 \n", "1272 0.174585 0.494105 0.727273 0.099567 0.052778 \n", "1273 0.209709 0.338692 0.727273 0.099567 0.052778 \n", "\n", " nnrc_disgust nnrc_fear nnrc_joy nnrc_sadness nnrc_surprise \\\n", "0 NaN NaN 0.222222 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", "5 NaN NaN NaN NaN NaN \n", "6 NaN NaN NaN NaN NaN \n", "7 NaN NaN NaN NaN NaN \n", "8 NaN NaN 1.000000 NaN 0.155844 \n", "9 0.176948 0.366883 0.805556 0.620130 NaN \n", "10 NaN 1.000000 0.654321 NaN NaN \n", "11 NaN NaN 1.000000 NaN 0.046600 \n", "12 NaN NaN 0.703704 NaN 0.276438 \n", "13 NaN 0.088312 0.377778 0.189610 0.290909 \n", "14 0.064935 0.025974 NaN NaN 0.064935 \n", "15 0.079103 NaN 0.528620 0.079103 0.171192 \n", "16 0.594805 0.797403 0.170370 0.594805 NaN \n", "17 0.240260 0.113636 0.740741 0.113636 0.746753 \n", "18 NaN NaN 0.407407 NaN 0.276438 \n", "19 0.324675 NaN NaN NaN NaN \n", "20 NaN 1.000000 0.446914 0.324675 0.189610 \n", "21 NaN 1.000000 0.481481 NaN 1.000000 \n", "22 NaN NaN NaN NaN NaN \n", "23 NaN 0.064935 0.202279 0.376623 0.064935 \n", "24 0.662338 NaN 1.000000 0.324675 0.324675 \n", "25 0.880825 1.000000 1.000000 1.000000 1.000000 \n", "26 NaN NaN 0.740741 NaN NaN \n", "27 NaN 0.099567 NaN 0.324675 0.324675 \n", "28 NaN 0.046600 1.000000 0.344538 0.523300 \n", "29 NaN NaN 0.407407 NaN 0.421150 \n", "... ... ... ... ... ... \n", "1244 0.437229 0.099567 0.423868 0.324675 0.324675 \n", "1245 0.189610 0.189610 0.792593 0.797403 0.594805 \n", "1246 0.409091 0.324675 0.481481 0.788961 0.029221 \n", "1247 0.021944 0.021944 0.892720 0.056874 0.126735 \n", "1248 0.355372 0.815821 0.434343 0.631641 0.355372 \n", "1249 0.766234 0.142857 NaN 0.220779 0.298701 \n", "1250 NaN 0.212121 0.769547 0.549784 0.099567 \n", "1251 0.355372 0.815821 0.434343 0.631641 0.355372 \n", "1252 0.189610 0.189610 0.792593 0.797403 0.594805 \n", "1253 0.766234 0.142857 NaN 0.220779 0.298701 \n", "1254 NaN 0.040328 0.836257 0.013671 0.173616 \n", "1255 0.861865 0.401417 0.057239 0.263282 0.125148 \n", "1256 0.220779 0.688312 0.202279 0.688312 NaN \n", "1257 NaN 0.050325 0.870370 0.018669 0.176948 \n", "1258 NaN 0.276438 0.703704 1.000000 0.565863 \n", "1259 0.171192 NaN 0.340067 0.355372 0.171192 \n", "1260 0.113636 0.113636 0.481481 NaN 0.113636 \n", "1261 0.556818 NaN 0.092593 0.050325 NaN \n", "1262 0.355372 0.815821 0.434343 0.631641 0.355372 \n", "1263 NaN 0.040328 0.836257 0.013671 0.173616 \n", "1264 NaN 0.276438 0.703704 1.000000 0.565863 \n", "1265 0.861865 0.401417 0.057239 0.263282 0.125148 \n", "1266 0.556818 NaN 0.092593 0.050325 NaN \n", "1267 NaN 0.050325 0.870370 0.018669 0.176948 \n", "1268 0.638219 0.348794 0.481481 0.493506 0.204082 \n", "1269 NaN NaN NaN NaN NaN \n", "1270 NaN 0.079103 0.622896 0.048406 0.570248 \n", "1271 NaN 0.079103 0.622896 0.048406 0.570248 \n", "1272 0.127706 0.099567 0.740741 0.183983 0.015152 \n", "1273 0.127706 0.099567 0.740741 0.183983 0.015152 \n", "\n", " nnrc_trust tempo valence \n", "0 NaN 0.425867 0.155738 \n", "1 NaN 0.335331 0.760246 \n", "2 NaN 0.329760 0.156762 \n", "3 NaN 0.739918 0.478484 \n", "4 NaN 0.671074 0.682377 \n", "5 NaN 0.854050 0.130123 \n", "6 NaN 0.854050 0.130123 \n", "7 NaN 0.000000 0.000000 \n", "8 0.140625 0.644934 0.991803 \n", "9 0.097656 0.357808 0.934426 \n", "10 0.312500 0.619264 0.887295 \n", "11 0.090074 0.425615 0.934426 \n", "12 0.705357 0.651131 0.748975 \n", "13 0.278125 0.502470 0.663934 \n", "14 0.048077 0.372792 0.861680 \n", "15 0.625000 0.519226 0.588115 \n", "16 0.381250 0.697336 0.545082 \n", "17 0.484375 0.640974 0.985656 \n", "18 0.410714 0.516885 0.953893 \n", "19 NaN 0.279509 0.539959 \n", "20 0.862500 0.720638 0.879098 \n", "21 0.484375 0.877479 0.537910 \n", "22 NaN 0.743869 0.686475 \n", "23 0.682692 0.679596 0.420082 \n", "24 1.000000 0.800757 0.536885 \n", "25 0.939338 0.782363 0.372951 \n", "26 NaN 0.430841 0.909836 \n", "27 NaN 0.626635 0.401639 \n", "28 0.636029 0.608918 0.978484 \n", "29 0.410714 0.583115 0.340164 \n", "... ... ... ... \n", "1244 0.541667 0.753912 0.713115 \n", "1245 0.793750 0.557582 0.185451 \n", "1246 0.269531 0.532011 0.139344 \n", "1247 0.075431 0.689165 0.401639 \n", "1248 0.625000 0.601111 0.431352 \n", "1249 0.206731 0.816029 0.407787 \n", "1250 0.656250 0.696038 0.365779 \n", "1251 0.625000 0.610202 0.542008 \n", "1252 0.793750 0.584659 0.559426 \n", "1253 0.206731 0.804409 0.461066 \n", "1254 0.701480 0.635768 0.611680 \n", "1255 0.296875 0.640704 0.431352 \n", "1256 0.127404 0.453167 0.253074 \n", "1257 0.742188 0.375089 0.276639 \n", "1258 0.558036 0.676422 0.657787 \n", "1259 0.531250 0.548321 0.614754 \n", "1260 0.484375 0.686086 0.767418 \n", "1261 0.742188 0.545147 0.597336 \n", "1262 0.625000 0.305624 0.045799 \n", "1263 0.701480 0.633963 0.400615 \n", "1264 0.558036 0.678350 0.644467 \n", "1265 0.296875 0.622386 0.517418 \n", "1266 0.742188 0.955916 0.492828 \n", "1267 0.742188 0.679605 0.431352 \n", "1268 0.558036 0.656417 0.680328 \n", "1269 NaN 0.344554 0.057582 \n", "1270 0.656250 0.740387 0.623975 \n", "1271 0.656250 0.734153 0.564549 \n", "1272 0.197917 0.525417 0.361680 \n", "1273 0.197917 0.569718 0.282787 \n", "\n", "[1274 rows x 13 columns]" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "all_df" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Analysis and calculation of the convex hull\n", "\n", "`artist_features()` extract the data for one artist and scales the various scores according to the `all_raw_df` values.\n", "\n", "`convex_hull_volume()` does the actual calculation. The shuffling is done determinsitically (with a fixed random seed) to make the calculations repeatable." ] }, { "cell_type": "code", "execution_count": 28, "metadata": { "scrolled": true }, "outputs": [], "source": [ "def artist_features(artist_id):\n", "\n", " pipeline = [\n", " {'$match': {'lyrics': {'$exists': True}, 'sentiment': {'$exists': True}, 'valence': {'$exists': True},\n", " 'artist_id': artist_id}},\n", " {'$project': projection_dict}\n", " ]\n", " raw_df = pd.DataFrame(list(tracks.aggregate(pipeline)))\n", " raw_df.drop(columns_to_drop, axis=1, inplace=True)\n", " raw_df.fillna(0, inplace=True)\n", " df = (raw_df-all_raw_df.min()) / (all_raw_df.max()-all_raw_df.min())\n", " df.popularity0 = 0\n", " return df" ] }, { "cell_type": "code", "execution_count": 29, "metadata": {}, "outputs": [], "source": [ "def convex_hull_volume(artist_df, state=42, groups=4):\n", " artist_s_df = artist_df.sample(frac=1, random_state=state)\n", " rows_per_subframe = math.ceil(len(artist_s_df) / groups)\n", " subframes = [i[1] for i in artist_s_df.groupby(np.arange(len(artist_s_df))//rows_per_subframe)]\n", " total_vol = 0\n", " for subframe in subframes:\n", " sub_ar = subframe.as_matrix()\n", " hull = ConvexHull(sub_ar, qhull_options='QJ')\n", " total_vol += hull.volume\n", " return total_vol" ] }, { "cell_type": "code", "execution_count": 30, "metadata": {}, "outputs": [], "source": [ "# sub_ar = subframes[0].as_matrix()\n", "# hull = ConvexHull(sub_ar, qhull_options='QJ')\n", "# hull.volume" ] }, { "cell_type": "code", "execution_count": 31, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
acousticnessdanceabilitykeynnrc_angernnrc_anticipationnnrc_disgustnnrc_fearnnrc_joynnrc_sadnessnnrc_surprisennrc_trusttempovalence
00.4390480.8488750.181818-0.012987-0.033333-0.012987-0.0129870.222222-0.012987-0.012987-0.0312500.4258670.155738
10.5237590.6227220.000000-0.0129870.483333-0.012987-0.0129871.000000-0.0129870.1558440.1406250.6449340.991803
20.2685930.4040730.3636360.3668830.3541670.1769480.3668830.8055560.620130-0.0129870.0976560.3578080.934426
30.3987590.5251880.636364-0.0129870.311111-0.0129871.0000000.654321-0.012987-0.0129870.3125000.6192640.887295
40.0933860.6045020.636364-0.0129870.452941-0.012987-0.0129871.000000-0.0129870.0466000.0900740.4256150.934426
50.1239650.7127550.545455-0.0129870.704762-0.012987-0.0129870.703704-0.0129870.2764380.7053570.6511310.748975
60.0655970.4051450.1818180.3922081.000000-0.0129870.0883120.3777780.1896100.2909090.2781250.5024700.663934
70.1518570.5058950.636364-0.0129870.0461540.0649350.025974-0.037037-0.0129870.0649350.0480770.3727920.861680
80.1456590.6966770.1818180.2632820.4363640.079103-0.0129870.5286200.0791030.1711920.6250000.5192260.588115
90.0115680.4137190.9090910.3922080.3800000.5948050.7974030.1703700.594805-0.0129870.3812500.6973360.545082
100.0048010.6977490.3636360.2402600.6125000.2402600.1136360.7407410.1136360.7467530.4843750.6409740.985656
110.1006180.7781350.636364-0.0129870.409524-0.012987-0.0129870.407407-0.0129870.2764380.4107140.5168850.953893
120.4876020.5766351.0000000.662338-0.0333330.324675-0.012987-0.037037-0.012987-0.012987-0.0312500.2795090.539959
130.3729320.2754560.0909090.1896100.311111-0.0129871.0000000.4469140.3246750.1896100.8625000.7206380.879098
140.1849150.3118970.818182-0.0129871.000000-0.0129871.0000000.481481-0.0129871.0000000.4843750.8774790.537910
150.6807840.5787780.4545451.000000-0.033333-0.012987-0.012987-0.037037-0.012987-0.012987-0.0312500.7438690.686475
160.6518590.4748120.0000000.6103900.205128-0.0129870.0649350.2022790.3766230.0649350.6826920.6795960.420082
170.3915270.5530550.1818180.3246750.6555560.662338-0.0129871.0000000.3246750.3246751.0000000.8007570.536885
180.0738610.4715970.1818180.8808251.0000000.8808251.0000001.0000001.0000001.0000000.9393380.7823630.372951
190.0317130.5937830.363636-0.012987-0.033333-0.012987-0.0129870.740741-0.012987-0.012987-0.0312500.4308410.909836
200.7809910.3204720.272727-0.0129871.000000-0.0129870.099567-0.0370370.3246750.324675-0.0312500.6266350.401639
210.2479320.9431940.181818-0.0129870.513725-0.0129870.0466001.0000000.3445380.5233000.6360290.6089180.978484
220.5082630.8156480.181818-0.0129871.000000-0.012987-0.0129870.407407-0.0129870.4211500.4107140.5831150.340164
230.0311960.5712750.8181820.8733770.1604170.1136360.9366880.1574070.0503250.1136360.0976560.7816570.191598
240.2045440.4244370.000000-0.0129870.586667-0.012987-0.0129871.0000000.5948050.7974031.0000000.6325090.386270
250.6497930.8188640.1818180.4211500.1142860.6382190.421150-0.0370370.4211500.2764380.1897320.6211300.743852
260.0130140.4523040.363636-0.012987-0.033333-0.0129870.4935060.4814810.6623380.2402600.4843750.8239260.562500
270.1601220.6709540.0909090.0750990.1463770.0310560.0310560.6843800.0310560.0310560.2377720.4368800.747951
280.0240680.3965700.1818181.000000-0.0333331.0000001.000000-0.0370371.000000-0.012987-0.0312500.5482360.430328
290.0350180.5969990.8181820.0196900.8333330.0196900.0196901.0000000.0196900.8039381.0000000.6118980.403689
..........................................
1930.6621890.5166130.1818180.2632820.060606-0.012987-0.0129871.000000-0.0129870.3553720.0625000.5903910.960041
1940.7024790.4062170.4545450.0995670.4259260.0995670.3246750.8847740.2121210.2121210.4270830.5722770.185451
1950.2634280.4973200.363636-0.0129870.793333-0.0129870.0545450.7925930.7974030.7298700.8625000.8566120.703893
1960.0241710.8038590.1818180.0545451.0000000.1220780.1220780.7925930.1896100.6623380.8625000.5767340.823770
1970.2675600.8531620.0000001.0000000.1545451.0000001.0000000.0572391.0000000.0791030.1562500.5649340.744877
1980.9493800.5176850.363636-0.012987-0.0333330.3668830.1136360.8703700.2402600.2402600.6132810.8073890.196721
1990.7603300.3944270.8181820.7974030.1733330.3922080.1896100.1703700.3922080.1896101.0000000.8463520.337090
2000.2179730.4748120.0000000.7974030.1733330.3922080.1896100.1703700.3922080.1896101.0000000.5250140.297131
2010.3749980.4147910.9090910.7974030.1733330.3922080.1896100.1703700.3922080.1896101.0000000.4626120.241803
2020.6911150.9646300.000000-0.0129870.060606-0.0129870.0791030.3400670.2632820.0791031.0000000.5722200.150615
2030.0352250.6730980.818182-0.0129870.060606-0.0129870.0791030.3400670.2632820.0791031.0000000.5366770.638320
2040.0298530.3933550.6363640.6961040.7933330.6961040.7974030.8962961.0000000.4935061.0000000.4507320.528689
2050.1177660.7213290.090909-0.0129870.483333-0.0129870.1136361.0000000.366883-0.0129870.2265620.5321910.728484
2060.0965890.3483390.181818-0.0129870.261905-0.0129870.0593690.185185-0.0129870.1317250.3370540.6430770.435451
2070.0668370.6420150.0000000.4211501.0000000.5658630.4211500.2592590.2764380.2764380.5580360.5776530.818648
2080.5020650.4415860.0000000.1896100.380000-0.0129870.1896100.1703700.189610-0.0129870.1750000.5309310.561475
2090.9132230.3987140.0000000.3922081.0000000.1896100.0883120.3777780.1896100.2909090.5875000.6329070.210041
2100.1229320.4683820.8181820.0791030.6242420.2632820.1711920.4343430.2632820.1711920.7187500.5273640.603484
2110.3615690.3140410.5454550.0995670.4259260.0995670.3246750.8847740.2121210.2121210.4270830.5929540.386270
2120.0071260.6184350.727273-0.0129870.793333-0.0129870.0545450.7925930.7974030.7298700.8625000.4143220.675205
2130.0003760.5530550.1818180.0545451.0000000.1220780.1220780.7925930.1896100.6623380.8625000.5758390.869877
2140.0002410.6173630.0000001.0000000.1545451.0000001.0000000.0572391.0000000.0791030.1562500.5626650.813525
2150.2623950.3633440.363636-0.012987-0.0333330.3668830.1136360.8703700.2402600.2402600.6132810.3832090.136270
2160.3006180.6248660.0000000.6961040.7933330.6961040.7974030.8962961.0000000.4935061.0000000.4477000.512295
2170.2396680.9785640.545455-0.0129870.483333-0.0129870.1136361.0000000.366883-0.0129870.2265620.5276580.389344
2180.7737600.4533760.181818-0.0129870.261905-0.0129870.0593690.185185-0.0129870.1317250.3370540.4716600.336066
2190.4617760.7824220.0000000.4211501.0000000.5658630.4211500.2592590.2764380.2764380.5580360.6042900.338115
2200.1673530.6334410.4545450.1896100.380000-0.0129870.1896100.1703700.189610-0.0129870.1750000.5382740.656762
2210.9493800.4748120.3636360.3922081.0000000.1896100.0883120.3777780.1896100.2909090.5875000.6046830.188525
2220.4979330.6377280.0000000.0791030.6242420.2632820.1711920.4343430.2632820.1711920.7187500.5292160.549180
\n", "

223 rows × 13 columns

\n", "
" ], "text/plain": [ " acousticness danceability key nnrc_anger nnrc_anticipation \\\n", "0 0.439048 0.848875 0.181818 -0.012987 -0.033333 \n", "1 0.523759 0.622722 0.000000 -0.012987 0.483333 \n", "2 0.268593 0.404073 0.363636 0.366883 0.354167 \n", "3 0.398759 0.525188 0.636364 -0.012987 0.311111 \n", "4 0.093386 0.604502 0.636364 -0.012987 0.452941 \n", "5 0.123965 0.712755 0.545455 -0.012987 0.704762 \n", "6 0.065597 0.405145 0.181818 0.392208 1.000000 \n", "7 0.151857 0.505895 0.636364 -0.012987 0.046154 \n", "8 0.145659 0.696677 0.181818 0.263282 0.436364 \n", "9 0.011568 0.413719 0.909091 0.392208 0.380000 \n", "10 0.004801 0.697749 0.363636 0.240260 0.612500 \n", "11 0.100618 0.778135 0.636364 -0.012987 0.409524 \n", "12 0.487602 0.576635 1.000000 0.662338 -0.033333 \n", "13 0.372932 0.275456 0.090909 0.189610 0.311111 \n", "14 0.184915 0.311897 0.818182 -0.012987 1.000000 \n", "15 0.680784 0.578778 0.454545 1.000000 -0.033333 \n", "16 0.651859 0.474812 0.000000 0.610390 0.205128 \n", "17 0.391527 0.553055 0.181818 0.324675 0.655556 \n", "18 0.073861 0.471597 0.181818 0.880825 1.000000 \n", "19 0.031713 0.593783 0.363636 -0.012987 -0.033333 \n", "20 0.780991 0.320472 0.272727 -0.012987 1.000000 \n", "21 0.247932 0.943194 0.181818 -0.012987 0.513725 \n", "22 0.508263 0.815648 0.181818 -0.012987 1.000000 \n", "23 0.031196 0.571275 0.818182 0.873377 0.160417 \n", "24 0.204544 0.424437 0.000000 -0.012987 0.586667 \n", "25 0.649793 0.818864 0.181818 0.421150 0.114286 \n", "26 0.013014 0.452304 0.363636 -0.012987 -0.033333 \n", "27 0.160122 0.670954 0.090909 0.075099 0.146377 \n", "28 0.024068 0.396570 0.181818 1.000000 -0.033333 \n", "29 0.035018 0.596999 0.818182 0.019690 0.833333 \n", ".. ... ... ... ... ... \n", "193 0.662189 0.516613 0.181818 0.263282 0.060606 \n", "194 0.702479 0.406217 0.454545 0.099567 0.425926 \n", "195 0.263428 0.497320 0.363636 -0.012987 0.793333 \n", "196 0.024171 0.803859 0.181818 0.054545 1.000000 \n", "197 0.267560 0.853162 0.000000 1.000000 0.154545 \n", "198 0.949380 0.517685 0.363636 -0.012987 -0.033333 \n", "199 0.760330 0.394427 0.818182 0.797403 0.173333 \n", "200 0.217973 0.474812 0.000000 0.797403 0.173333 \n", "201 0.374998 0.414791 0.909091 0.797403 0.173333 \n", "202 0.691115 0.964630 0.000000 -0.012987 0.060606 \n", "203 0.035225 0.673098 0.818182 -0.012987 0.060606 \n", "204 0.029853 0.393355 0.636364 0.696104 0.793333 \n", "205 0.117766 0.721329 0.090909 -0.012987 0.483333 \n", "206 0.096589 0.348339 0.181818 -0.012987 0.261905 \n", "207 0.066837 0.642015 0.000000 0.421150 1.000000 \n", "208 0.502065 0.441586 0.000000 0.189610 0.380000 \n", "209 0.913223 0.398714 0.000000 0.392208 1.000000 \n", "210 0.122932 0.468382 0.818182 0.079103 0.624242 \n", "211 0.361569 0.314041 0.545455 0.099567 0.425926 \n", "212 0.007126 0.618435 0.727273 -0.012987 0.793333 \n", "213 0.000376 0.553055 0.181818 0.054545 1.000000 \n", "214 0.000241 0.617363 0.000000 1.000000 0.154545 \n", "215 0.262395 0.363344 0.363636 -0.012987 -0.033333 \n", "216 0.300618 0.624866 0.000000 0.696104 0.793333 \n", "217 0.239668 0.978564 0.545455 -0.012987 0.483333 \n", "218 0.773760 0.453376 0.181818 -0.012987 0.261905 \n", "219 0.461776 0.782422 0.000000 0.421150 1.000000 \n", "220 0.167353 0.633441 0.454545 0.189610 0.380000 \n", "221 0.949380 0.474812 0.363636 0.392208 1.000000 \n", "222 0.497933 0.637728 0.000000 0.079103 0.624242 \n", "\n", " nnrc_disgust nnrc_fear nnrc_joy nnrc_sadness nnrc_surprise \\\n", "0 -0.012987 -0.012987 0.222222 -0.012987 -0.012987 \n", "1 -0.012987 -0.012987 1.000000 -0.012987 0.155844 \n", "2 0.176948 0.366883 0.805556 0.620130 -0.012987 \n", "3 -0.012987 1.000000 0.654321 -0.012987 -0.012987 \n", "4 -0.012987 -0.012987 1.000000 -0.012987 0.046600 \n", "5 -0.012987 -0.012987 0.703704 -0.012987 0.276438 \n", "6 -0.012987 0.088312 0.377778 0.189610 0.290909 \n", "7 0.064935 0.025974 -0.037037 -0.012987 0.064935 \n", "8 0.079103 -0.012987 0.528620 0.079103 0.171192 \n", "9 0.594805 0.797403 0.170370 0.594805 -0.012987 \n", "10 0.240260 0.113636 0.740741 0.113636 0.746753 \n", "11 -0.012987 -0.012987 0.407407 -0.012987 0.276438 \n", "12 0.324675 -0.012987 -0.037037 -0.012987 -0.012987 \n", "13 -0.012987 1.000000 0.446914 0.324675 0.189610 \n", "14 -0.012987 1.000000 0.481481 -0.012987 1.000000 \n", "15 -0.012987 -0.012987 -0.037037 -0.012987 -0.012987 \n", "16 -0.012987 0.064935 0.202279 0.376623 0.064935 \n", "17 0.662338 -0.012987 1.000000 0.324675 0.324675 \n", "18 0.880825 1.000000 1.000000 1.000000 1.000000 \n", "19 -0.012987 -0.012987 0.740741 -0.012987 -0.012987 \n", "20 -0.012987 0.099567 -0.037037 0.324675 0.324675 \n", "21 -0.012987 0.046600 1.000000 0.344538 0.523300 \n", "22 -0.012987 -0.012987 0.407407 -0.012987 0.421150 \n", "23 0.113636 0.936688 0.157407 0.050325 0.113636 \n", "24 -0.012987 -0.012987 1.000000 0.594805 0.797403 \n", "25 0.638219 0.421150 -0.037037 0.421150 0.276438 \n", "26 -0.012987 0.493506 0.481481 0.662338 0.240260 \n", "27 0.031056 0.031056 0.684380 0.031056 0.031056 \n", "28 1.000000 1.000000 -0.037037 1.000000 -0.012987 \n", "29 0.019690 0.019690 1.000000 0.019690 0.803938 \n", ".. ... ... ... ... ... \n", "193 -0.012987 -0.012987 1.000000 -0.012987 0.355372 \n", "194 0.099567 0.324675 0.884774 0.212121 0.212121 \n", "195 -0.012987 0.054545 0.792593 0.797403 0.729870 \n", "196 0.122078 0.122078 0.792593 0.189610 0.662338 \n", "197 1.000000 1.000000 0.057239 1.000000 0.079103 \n", "198 0.366883 0.113636 0.870370 0.240260 0.240260 \n", "199 0.392208 0.189610 0.170370 0.392208 0.189610 \n", "200 0.392208 0.189610 0.170370 0.392208 0.189610 \n", "201 0.392208 0.189610 0.170370 0.392208 0.189610 \n", "202 -0.012987 0.079103 0.340067 0.263282 0.079103 \n", "203 -0.012987 0.079103 0.340067 0.263282 0.079103 \n", "204 0.696104 0.797403 0.896296 1.000000 0.493506 \n", "205 -0.012987 0.113636 1.000000 0.366883 -0.012987 \n", "206 -0.012987 0.059369 0.185185 -0.012987 0.131725 \n", "207 0.565863 0.421150 0.259259 0.276438 0.276438 \n", "208 -0.012987 0.189610 0.170370 0.189610 -0.012987 \n", "209 0.189610 0.088312 0.377778 0.189610 0.290909 \n", "210 0.263282 0.171192 0.434343 0.263282 0.171192 \n", "211 0.099567 0.324675 0.884774 0.212121 0.212121 \n", "212 -0.012987 0.054545 0.792593 0.797403 0.729870 \n", "213 0.122078 0.122078 0.792593 0.189610 0.662338 \n", "214 1.000000 1.000000 0.057239 1.000000 0.079103 \n", "215 0.366883 0.113636 0.870370 0.240260 0.240260 \n", "216 0.696104 0.797403 0.896296 1.000000 0.493506 \n", "217 -0.012987 0.113636 1.000000 0.366883 -0.012987 \n", "218 -0.012987 0.059369 0.185185 -0.012987 0.131725 \n", "219 0.565863 0.421150 0.259259 0.276438 0.276438 \n", "220 -0.012987 0.189610 0.170370 0.189610 -0.012987 \n", "221 0.189610 0.088312 0.377778 0.189610 0.290909 \n", "222 0.263282 0.171192 0.434343 0.263282 0.171192 \n", "\n", " nnrc_trust tempo valence \n", "0 -0.031250 0.425867 0.155738 \n", "1 0.140625 0.644934 0.991803 \n", "2 0.097656 0.357808 0.934426 \n", "3 0.312500 0.619264 0.887295 \n", "4 0.090074 0.425615 0.934426 \n", "5 0.705357 0.651131 0.748975 \n", "6 0.278125 0.502470 0.663934 \n", "7 0.048077 0.372792 0.861680 \n", "8 0.625000 0.519226 0.588115 \n", "9 0.381250 0.697336 0.545082 \n", "10 0.484375 0.640974 0.985656 \n", "11 0.410714 0.516885 0.953893 \n", "12 -0.031250 0.279509 0.539959 \n", "13 0.862500 0.720638 0.879098 \n", "14 0.484375 0.877479 0.537910 \n", "15 -0.031250 0.743869 0.686475 \n", "16 0.682692 0.679596 0.420082 \n", "17 1.000000 0.800757 0.536885 \n", "18 0.939338 0.782363 0.372951 \n", "19 -0.031250 0.430841 0.909836 \n", "20 -0.031250 0.626635 0.401639 \n", "21 0.636029 0.608918 0.978484 \n", "22 0.410714 0.583115 0.340164 \n", "23 0.097656 0.781657 0.191598 \n", "24 1.000000 0.632509 0.386270 \n", "25 0.189732 0.621130 0.743852 \n", "26 0.484375 0.823926 0.562500 \n", "27 0.237772 0.436880 0.747951 \n", "28 -0.031250 0.548236 0.430328 \n", "29 1.000000 0.611898 0.403689 \n", ".. ... ... ... \n", "193 0.062500 0.590391 0.960041 \n", "194 0.427083 0.572277 0.185451 \n", "195 0.862500 0.856612 0.703893 \n", "196 0.862500 0.576734 0.823770 \n", "197 0.156250 0.564934 0.744877 \n", "198 0.613281 0.807389 0.196721 \n", "199 1.000000 0.846352 0.337090 \n", "200 1.000000 0.525014 0.297131 \n", "201 1.000000 0.462612 0.241803 \n", "202 1.000000 0.572220 0.150615 \n", "203 1.000000 0.536677 0.638320 \n", "204 1.000000 0.450732 0.528689 \n", "205 0.226562 0.532191 0.728484 \n", "206 0.337054 0.643077 0.435451 \n", "207 0.558036 0.577653 0.818648 \n", "208 0.175000 0.530931 0.561475 \n", "209 0.587500 0.632907 0.210041 \n", "210 0.718750 0.527364 0.603484 \n", "211 0.427083 0.592954 0.386270 \n", "212 0.862500 0.414322 0.675205 \n", "213 0.862500 0.575839 0.869877 \n", "214 0.156250 0.562665 0.813525 \n", "215 0.613281 0.383209 0.136270 \n", "216 1.000000 0.447700 0.512295 \n", "217 0.226562 0.527658 0.389344 \n", "218 0.337054 0.471660 0.336066 \n", "219 0.558036 0.604290 0.338115 \n", "220 0.175000 0.538274 0.656762 \n", "221 0.587500 0.604683 0.188525 \n", "222 0.718750 0.529216 0.549180 \n", "\n", "[223 rows x 13 columns]" ] }, "execution_count": 31, "metadata": {}, "output_type": "execute_result" } ], "source": [ "beatles_df = artist_features(artist_ids['The Beatles'])\n", "beatles_df" ] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "2.113885406456015e-06" ] }, "execution_count": 32, "metadata": {}, "output_type": "execute_result" } ], "source": [ "beatles_volume = convex_hull_volume(beatles_df)\n", "beatles_volume" ] }, { "cell_type": "code", "execution_count": 69, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 69, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgEAAAD8CAYAAADudXePAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# ax = beatles_df.drop(['popularity0'], axis=1).plot.kde()\n", "ax = beatles_df.plot.kde()\n", "ax.legend(loc='center left', bbox_to_anchor=(1, 0.5))" ] }, { "cell_type": "code", "execution_count": 70, "metadata": {}, "outputs": [], "source": [ "# ax = stones_df.plot.kde()\n", "# ax.legend(loc='center left', bbox_to_anchor=(1, 0.5))" ] }, { "cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "13" ] }, "execution_count": 33, "metadata": {}, "output_type": "execute_result" } ], "source": [ "beatles_df.columns.size" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Do the calculation\n", "This cycles through all the known artist IDs and calculates the convex hull volume for each.\n", "\n", "Note the couple of special cases: there are so few Spice Girls tracks, we can't split them into four sections. The Queen analysis blows up the convex hull calculation when `state=42`." ] }, { "cell_type": "code", "execution_count": 76, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "starting Radiohead\n", "Radiohead 1.3735907773308412e-08\n", "starting Foo Fighters\n", "Foo Fighters 7.666641146914435e-09\n", "starting Spice Girls\n", "Spice Girls 2.5497245489351534e-09\n", "starting Led Zeppelin\n", "Led Zeppelin 7.120684861152149e-09\n", "starting Queen\n", "Queen 7.675230324648466e-07\n", "starting The Beatles\n", "The Beatles 1.8584109644151603e-06\n", "starting The Rolling Stones\n", "The Rolling Stones 1.3830666327684997e-06\n", "starting Abba\n", "Abba 3.986374918574468e-09\n" ] } ], "source": [ "results_df = pd.DataFrame([{'name': k, 'id': artist_ids[k]}\n", " for k in artist_ids])\n", "results_df.set_index('id', inplace=True)\n", "results_df['raw_volume'] = 0\n", "\n", "artist_scores = {}\n", "\n", "for artist in artist_ids:\n", " print('starting', artist)\n", " artist_df = artist_features(artist_ids[artist])\n", " if artist == 'Spice Girls':\n", " artist_volume = convex_hull_volume(artist_df, groups=1, state=77777) * 4\n", " elif artist == 'Queen':\n", " artist_volume = convex_hull_volume(artist_df, groups=4, state=77777)\n", " else:\n", " artist_volume = convex_hull_volume(artist_df, state=77777)\n", " artist_scores[artist] = artist_volume\n", " print(artist, artist_volume)\n" ] }, { "cell_type": "code", "execution_count": 77, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{42: {'Abba': 5.682966002661997e-09,\n", " 'Foo Fighters': 1.6874480985875627e-08,\n", " 'Led Zeppelin': 1.0078024641849407e-08,\n", " 'Queen': 7.149182754285038e-07,\n", " 'Radiohead': 1.6984954630731168e-08,\n", " 'Spice Girls': 2.5497245468497206e-09,\n", " 'The Beatles': 2.113885406456015e-06,\n", " 'The Rolling Stones': 1.389105641110557e-06},\n", " 123: {'Abba': 2.7090200835856255e-09,\n", " 'Foo Fighters': 1.2874782699947662e-08,\n", " 'Led Zeppelin': 6.8943416946745336e-09,\n", " 'Queen': 7.149182754285038e-07,\n", " 'Radiohead': 1.3597240666142407e-08,\n", " 'Spice Girls': 2.5497245457016713e-09,\n", " 'The Beatles': 2.276084073574721e-06,\n", " 'The Rolling Stones': 1.2714625609036554e-06},\n", " 999: {'Abba': 3.609183347846526e-09,\n", " 'Foo Fighters': 1.3255723616195349e-08,\n", " 'Led Zeppelin': 6.9007454600734344e-09,\n", " 'Queen': 8.411696372118656e-07,\n", " 'Radiohead': 1.3715723194819635e-08,\n", " 'Spice Girls': 2.5497245472409113e-09,\n", " 'The Beatles': 1.6188840757436778e-06,\n", " 'The Rolling Stones': 1.2579221200636931e-06},\n", " 77777: {'Abba': 3.986374918574468e-09,\n", " 'Foo Fighters': 7.666641146914435e-09,\n", " 'Led Zeppelin': 7.120684861152149e-09,\n", " 'Queen': 7.675230324648466e-07,\n", " 'Radiohead': 1.3735907773308412e-08,\n", " 'Spice Girls': 2.5497245489351534e-09,\n", " 'The Beatles': 1.8584109644151603e-06,\n", " 'The Rolling Stones': 1.3830666327684997e-06}}" ] }, "execution_count": 77, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# seed_artist_scores = {}\n", "seed_artist_scores[77777] = {k: v for k, v in artist_scores.items()}\n", "seed_artist_scores" ] }, { "cell_type": "code", "execution_count": 70, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[('Spice Girls', 218.25340121957652),\n", " ('Abba', 224.1660910874002),\n", " ('Led Zeppelin', 235.62576082803054),\n", " ('Foo Fighters', 247.7599095937317),\n", " ('Radiohead', 248.41091249485208),\n", " ('Queen', 340.9442362055484),\n", " ('The Rolling Stones', 351.6634333304922),\n", " ('The Beatles', 358.5544012084572)]" ] }, "execution_count": 70, "metadata": {}, "output_type": "execute_result" } ], "source": [ "sorted(((k, math.pow(s, 1/13) * 1000) for k, s in artist_scores.items()), key=lambda p: p[1])" ] }, { "cell_type": "code", "execution_count": 78, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "1.0229846740306523" ] }, "execution_count": 78, "metadata": {}, "output_type": "execute_result" } ], "source": [ "math.pow(artist_scores['The Beatles'], 1/13) / math.pow(artist_scores['The Rolling Stones'], 1/13)" ] }, { "cell_type": "code", "execution_count": 79, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "1.0458097129614543" ] }, "execution_count": 79, "metadata": {}, "output_type": "execute_result" } ], "source": [ "s = 123\n", "math.pow(seed_artist_scores[s]['The Beatles'], 1/13) / math.pow(seed_artist_scores[s]['The Rolling Stones'], 1/13)" ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'Abba': 5.682966002661997e-09,\n", " 'Foo Fighters': 1.6874480985875627e-08,\n", " 'Led Zeppelin': 1.0078024641849407e-08,\n", " 'Radiohead': 1.6984954630731168e-08,\n", " 'Spice Girls': 2.5497245468497206e-09,\n", " 'The Beatles': 2.113885406456015e-06,\n", " 'The Rolling Stones': 1.389105641110557e-06}" ] }, "execution_count": 36, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# original_artist_scores = artist_scores\n", "original_artist_scores" ] }, { "cell_type": "code", "execution_count": 52, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(336.7057183984806, 340.9442362055484)" ] }, "execution_count": 52, "metadata": {}, "output_type": "execute_result" } ], "source": [ "math.pow(7.149182754285038e-07, 1/13) * 1000, math.pow(8.411696372118656e-07, 1/13) * 1000" ] }, { "cell_type": "code", "execution_count": 55, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "1.0328247299337718" ] }, "execution_count": 55, "metadata": {}, "output_type": "execute_result" } ], "source": [ "365.98880658455073 / 354.35712950833306" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "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": "3.5.3" } }, "nbformat": 4, "nbformat_minor": 1 }