Update app.py

app.py

@@ -43,18 +43,37 @@ if len(data) > 0:
     else:
         data['rating_ratio'] = 0.5  # Default neutral rating
     
-    # Add playtime features if available
+    # Add playtime features if available - FIX: Added proper checks for empty data
     if 'average_playtime_forever' in data.columns:
         # Log transform to handle skewed distribution
         data['log_playtime'] = np.log1p(data['average_playtime_forever'])
-        scaler = MinMaxScaler()
-        data['playtime_scaled'] = scaler.fit_transform(data[['log_playtime']])
+        
+        # FIX: Check if we have valid data before scaling
+        if len(data) > 0 and not data['log_playtime'].isna().all():
+            try:
+                scaler = MinMaxScaler()
+                # FIX: Only scale non-NA values
+                valid_playtime_mask = ~data['log_playtime'].isna()
+                if valid_playtime_mask.any():  # Only if we have any valid values
+                    data.loc[valid_playtime_mask, 'playtime_scaled'] = scaler.fit_transform(
+                        data.loc[valid_playtime_mask, ['log_playtime']]
+                    )
+                else:
+                    data['playtime_scaled'] = 0.5
+            except Exception as e:
+                print(f"Error in playtime scaling: {e}")
+                data['playtime_scaled'] = 0.5
+        else:
+            data['playtime_scaled'] = 0.5
     else:
         data['playtime_scaled'] = 0.5
     
-    # Add user score features if available
+    # Add user score features if available - FIX: Added proper checks
     if 'user_score' in data.columns:
-        data['user_score_scaled'] = data['user_score'] / 100.0  # Assuming user_score is out of 100
+        # FIX: Handle potential non-numeric values
+        data['user_score'] = pd.to_numeric(data['user_score'], errors='coerce')
+        # FIX: Check for NaN values before scaling
+        data['user_score_scaled'] = data['user_score'].fillna(50) / 100.0  # Assuming user_score is out of 100
     else:
         data['user_score_scaled'] = 0.5
     
@@ -84,16 +103,21 @@ if len(data) > 0:
     
     # Vectorize with improved parameters
     try:
-        # Use more n-grams and increased max_features for better semantic understanding
-        vectorizer = TfidfVectorizer(
-            stop_words='english', 
-            ngram_range=(1, 3),  # Capture phrases up to 3 words
-            max_features=10000,   # Increase features for more nuanced relationships
-            min_df=2,             # Ignore very rare terms
-            max_df=0.9            # Ignore very common terms
-        )
-        feature_vectors = vectorizer.fit_transform(data['combined_features'])
-        print(f"Vectorization complete. Shape: {feature_vectors.shape}")
+        # FIX: Check if we have enough data for vectorization
+        if len(data) > 1:
+            # Use more n-grams and increased max_features for better semantic understanding
+            vectorizer = TfidfVectorizer(
+                stop_words='english', 
+                ngram_range=(1, 3),  # Capture phrases up to 3 words
+                max_features=10000,   # Increase features for more nuanced relationships
+                min_df=2,             # Ignore very rare terms
+                max_df=0.9            # Ignore very common terms
+            )
+            feature_vectors = vectorizer.fit_transform(data['combined_features'])
+            print(f"Vectorization complete. Shape: {feature_vectors.shape}")
+        else:
+            print("Not enough data for vectorization")
+            feature_vectors = np.zeros((len(data), 1))
     except Exception as e:
         print(f"Vectorization error: {e}")
         feature_vectors = np.zeros((len(data), 1))
@@ -102,8 +126,23 @@ if len(data) > 0:
     if 'positive_ratings' in data.columns and len(data) > 0:
         # Log transform to handle skewed distribution of ratings
         data['log_ratings'] = np.log1p(data['positive_ratings'])
-        scaler = MinMaxScaler()
-        data['positive_ratings_scaled'] = scaler.fit_transform(data[['log_ratings']])
+        
+        # FIX: Check for valid data before scaling
+        if not data['log_ratings'].isna().all():
+            try:
+                scaler = MinMaxScaler()
+                valid_ratings_mask = ~data['log_ratings'].isna()
+                if valid_ratings_mask.any():
+                    data.loc[valid_ratings_mask, 'positive_ratings_scaled'] = scaler.fit_transform(
+                        data.loc[valid_ratings_mask, ['log_ratings']]
+                    )
+                else:
+                    data['positive_ratings_scaled'] = 0
+            except Exception as e:
+                print(f"Error in ratings scaling: {e}")
+                data['positive_ratings_scaled'] = 0
+        else:
+            data['positive_ratings_scaled'] = 0
     else:
         data['positive_ratings_scaled'] = 0