final.py

# ... Previous code for fetching and processing songs based on mood ...

# Assuming you have a trained mood classification model called 'classifier'

# Function to predict mood for a given song's lyrics
def predict_mood(song_lyrics):
    # Preprocess the lyrics data as needed (cleaning, tokenization, vectorization)
    lyrics_tfidf = tfidf_vectorizer.transform([song_lyrics])
    
    # Predict the mood label for the song
    mood_label = classifier.predict(lyrics_tfidf)[0]
    
    return mood_label

# Initialize a list to store recommended songs
recommended_songs = []

# Loop through the fetched songs
for song_data in all_lyrics:
    song_title = song_data['title']
    artist_name = song_data['artist']
    lyrics = song_data['lyrics']
    
    # Predict the mood for the song's lyrics
    predicted_mood = predict_mood(lyrics)
    
    # Check if the predicted mood matches the user's specified mood
    if predicted_mood == desired_mood:
        recommended_songs.append({'title': song_title, 'artist': artist_name, 'lyrics': lyrics})

# Now, recommended_songs contains songs that match the user's specified mood
for song_data in recommended_songs:
    print(f"Title: {song_data['title']}")
    print(f"Artist: {song_data['artist']}")
    print(f"Lyrics: {song_data['lyrics']}\n")