from data_processing import load_data, spotify_data, path import pandas import nltk from nltk.corpus import stopwords from nltk.tokenize import word_tokenize import string #---------------------------Download the requirements NLTK-------------------------------- #nltk.download('punkt') #nltk.download('stopwords') def clean_lyrics(lyrics): # TokenizaciĆ³n tokens = word_tokenize(lyrics) # To lower case tokens = [word.lower() for word in tokens] # Delete signs table = str.maketrans('', '', string.punctuation) stripped_tokens = [word.translate(table) for word in tokens] # Stop Words stop_words = set(stopwords.words('english')) tokens_without_sw = [word for word in stripped_tokens if word not in stop_words] return tokens_without_sw # Apply clean spotify_data['cleaned_text'] = spotify_data['text'].apply(clean_lyrics) spotify_data.to_csv('spotify_data_processed.csv', index=False) #print(spotify_data['cleaned_text'].head())