Update app.py

app.py

@@ -5,6 +5,7 @@ import joblib
 import pandas as pd
 import numpy as np
 from sklearn.metrics.pairwise import cosine_similarity
+from sklearn.preprocessing import StandardScaler
 
 # Load the emotion prediction model
 emotion_model = load_model('lstm_model.h5')
@@ -18,9 +19,9 @@ emotion_model = load_model('lstm_model.h5')
 # Load the tokenizer (ensure it's the one used during training)
 tokenizer = joblib.load('tokenizer.pkl')
 
-# Load the dataset and preprocess
+# Load the dataset
 df = pd.read_csv('df1.csv')
-df = df.drop(['Unnamed: 0', 'lyrics_filename', 'analysis_url', 'track_href', "type", "id", "uri", 'mood'], axis=1)
+df = df.drop(['Unnamed: 0', 'lyrics_filename', 'analysis_url', 'track_href', "type", "id", "uri"], axis=1)
 
 # Load the similarity matrix
 similarity_matrix = np.load('similarity_matrix.npy')
@@ -40,6 +41,30 @@ knn = joblib.load('knn_model.joblib')
 # Load the KNN recommendation function
 recommend_knn = joblib.load('recommendation_knn_function.joblib')
 
+# Preprocess for content-based
+audio_features = df[['danceability', 'energy', 'key', 'loudness', 'mode', 'speechiness',
+                     'acousticness', 'instrumentalness', 'liveness', 'valence', 'tempo',
+                     'duration_ms', 'time_signature']]
+mood_cats = df[['mood_cats']]
+
+scaler = StandardScaler()
+audio_features_scaled = scaler.fit_transform(audio_features)
+audio_features_df = pd.DataFrame(audio_features_scaled, columns=audio_features.columns)
+mood_cats_df = pd.DataFrame(mood_cats)
+combined_features_content = pd.concat([mood_cats_df, audio_features_df], axis=1)
+
+# Preprocess for KNN
+audio_features_knn = df[['danceability', 'energy', 'key', 'loudness', 'mode', 'speechiness',
+                          'acousticness', 'instrumentalness', 'liveness', 'valence', 'tempo',
+                          'duration_ms', 'time_signature']]
+mood_cats_knn = df[['mood_cats']]
+
+scaler_knn = StandardScaler()
+audio_features_scaled_knn = scaler_knn.fit_transform(audio_features_knn)
+audio_features_df_knn = pd.DataFrame(audio_features_scaled_knn, columns=audio_features_knn.columns)
+mood_cats_df_knn = pd.DataFrame(mood_cats_knn)
+combined_features_knn = pd.concat([mood_cats_df_knn, audio_features_df_knn], axis=1)
+
 # Set up the title of the app
 st.title('Emotion and Audio Feature-based Song Recommendation System')
 
@@ -48,11 +73,11 @@ query_data = df.iloc[0]
 
 # Process the lyrics
 sequence = tokenizer.texts_to_sequences([query_data['lyrics']])
-padded_sequence = pad_sequences(sequence, maxlen=50)  # Adjust the maxlen to match the expected input size
+padded_sequence = pad_sequences(sequence, maxlen=50)
 emotion = emotion_model.predict(padded_sequence).flatten()
 
 # Combine emotion and audio features for recommendation
-combined_features = np.concatenate([emotion, query_data[audio_feature_columns].values])
+combined_features_hybrid = np.concatenate([emotion, query_data[audio_feature_columns].values])
 
 # Generate recommendations using the hybrid model
 hybrid_recs = hybrid_recommendation(song_index=0)