dataset from lastfm by tag

build_dataset_by_tag.py

import pylast
import pandas as pd
from tqdm import tqdm
API_KEY_lastfm = "key_here"
API_SECRET_lastfm = "	key_here"
username_lastfm = "username"

def build_dataset_by_tag(tag_list,output_file):
    print('[INFO] Bulding Dataset by Tag....')
    network_lastfm = pylast.LastFMNetwork(api_key=API_KEY_lastfm, api_secret=API_SECRET_lastfm,username=username_lastfm)
    df = pd.DataFrame( columns=['uuid','style','artist','title_music','lastfm_music_url','lastfm_artist_url'])
    tqdm.pandas()
    for style in tag_list:
        dict_to_add = {
            'style': style,
        }
        tracks = network_lastfm.get_tag(style).get_top_tracks(limit=1000)
        for track in tracks:
            dict_to_add['artist'] = track.item.get_artist()
            dict_to_add['title_music'] = track.item.get_name()
            dict_to_add['lastfm_music_url'] = track.item.get_artist().get_url()
            dict_to_add['lastfm_artist_url'] = track.item.get_url()
            df = df.append(dict_to_add , ignore_index=True)
    df['uuid'] = df.progress_apply(create_uuid_by_movie, axis=1)
    df['videoID_youtube'] = "missing"
    df.to_csv(output_file, sep=";", header=True, index=False)
    print('[INFO] Bulding Dataset by Tag DONE')
    
output_file ='data/tracks_list.csv'
build_dataset_by_tag(['country','rap'], output_file)