使用跟踪URI从Spotify提取音频特征

Question

我正在尝试使用track URI从Spotify中提取音频特征。我有一个500k的列表，并希望为所有人提取音频特征。我有一个可行的代码下面，可以提取80首歌曲的特征。我需要一些帮助来修改下面的代码，一次提取80，这样我就不会违反Spotify的限制。下面是该列表的一个示例

['spotify:track:2d7LPtieXdIYzf7yHPooWd',
 'spotify:track:0y4TKcc7p2H6P0GJlt01EI',
 'spotify:track:6q4c1vPRZREh7nw3wG7Ixz',
 'spotify:track:54KFQB6N4pn926IUUYZGzK',
 'spotify:track:0NeJjNlprGfZpeX2LQuN6c']

client_id = 'xxx'
client_secret = 'xxx'
client_credentials_manager = SpotifyClientCredentials(client_id=client_id, client_secret=client_secret)
sp = spotipy.Spotify(client_credentials_manager=client_credentials_manager)

def get_audio_features(saved_uris):

    artist = []
    track = []

    danceability = []
    energy = []
    key = []
    loudness = []
    mode = []
    speechiness = []
    acousticness = []
    instrumentalness = []
    liveness = []
    valence = []
    tempo = []
    duration_ms = []


    for uri in saved_uris:

            x = sp.audio_features(uri)
            y = sp.track(uri)

        for audio_features in x:
            danceability.append(audio_features['danceability'])
            energy.append(audio_features['energy'])
            key.append(audio_features['key'])
            loudness.append(audio_features['loudness'])
            mode.append(audio_features['mode'])
            speechiness.append(audio_features['speechiness'])
            acousticness.append(audio_features['acousticness'])
            instrumentalness.append(audio_features['instrumentalness'])
            liveness.append(audio_features['liveness'])
            valence.append(audio_features['valence'])
            tempo.append(audio_features['tempo'])
            duration_ms.append(audio_features['duration_ms'])

        artist.append(y['album']['artists'][0]['name'])
        track.append(y['name'])

    df = pd.DataFrame()
    df['artist'] = artist
    df['track'] = track
    df['danceability'] = danceability
    df['energy'] = energy
    df['key'] = key
    df['loudness'] = loudness
    df['mode'] = mode
    df['speechiness'] = speechiness
    df['acousticness'] = acousticness
    df['instrumentalness'] = instrumentalness
    df['liveness'] = liveness
    df['valence'] = valence
    df['tempo'] = tempo
    df['duration_ms'] = duration_ms

    df.to_csv('data/xxx.csv')

    return df

我的输出是一个dataframe，它看起来像这样，为了提高可靠性，我删减了一些列：

artist               track   danceability   energy  key     loudness    
Sleeping At Last    Chasing    Cars          0.467  0.157      11 

Answer 1

此代码将返回所需的数据帧。

import spotipy
import time
from spotipy.oauth2 import SpotifyClientCredentials #To access authorised Spotify data4
import pandas as pd

client_id = 'paste client_id here'
client_secret = 'paste client_secret here'
client_credentials_manager = SpotifyClientCredentials(client_id=client_id, client_secret=client_secret)
sp = spotipy.Spotify(client_credentials_manager=client_credentials_manager)
sp.trace=False

#your uri list goes here
s_list = ['spotify:track:2d7LPtieXdIYzf7yHPooWd','spotify:track:0y4TKcc7p2H6P0GJlt01EI','spotify:track:6q4c1vPRZREh7nw3wG7Ixz','spotify:track:54KFQB6N4pn926IUUYZGzK','spotify:track:0NeJjNlprGfZpeX2LQuN6c']

#put uri to dataframe
df = pd.DataFrame(s_list)
df.columns = ['URI']

df['energy'] = ''*df.shape[0]
df['loudness'] = ''*df.shape[0]
df['speechiness'] = ''*df.shape[0]
df['valence'] = ''*df.shape[0]
df['liveness'] = ''*df.shape[0]
df['tempo'] = ''*df.shape[0]
df['danceability'] = ''*df.shape[0]

for i in range(0,df.shape[0]):
  time.sleep(random.uniform(3, 6))
  URI = df.URI[i]
  features = sp.audio_features(URI)
  df.loc[i,'energy'] = features[0]['energy']
  df.loc[i,'speechiness'] = features[0]['speechiness']
  df.loc[i,'liveness'] = features[0]['liveness']
  df.loc[i,'loudness'] = features[0]['loudness']
  df.loc[i,'danceability'] = features[0]['danceability']
  df.loc[i,'tempo'] = features[0]['tempo']
  df.loc[i,'valence'] = features[0]['valence']
  uri=0

输出：

​

希望这解决了你的问题。