Changes in deployment++

main.py

@@ -12,6 +12,7 @@ origins = [
     "http://localhost:5501",
     "http://127.0.0.1:5500",
     "http://localhost:5500",
+    "https://agro-vision-frontend.vercel.app"
 ]
 
 app.add_middleware(

routes/predictions.py

@@ -264,7 +264,8 @@ def predict_commodity_price(commodity: str):
         
         if not test_df.empty:
             FEATURES = [col for col in test_df.columns if col != 'modal_price']
-            # FIX: Use DataFrame with columns for XGBoost
+            
+            # FIX: Ensure DataFrame format for XGBoost
             X_input = pd.DataFrame(test_df[FEATURES].values, columns=FEATURES, index=test_df.index)
             
             predictions = model.predict(X_input)
@@ -286,7 +287,7 @@ def predict_commodity_price(commodity: str):
         for date, row in daily_forecast_df.iterrows():
             price = row['forecast']
             
-            # --- CRITICAL FIX: Handle NaN values to prevent JSON crash ---
+            # --- CRITICAL FIX: Handle NaN values safely ---
             if pd.isna(price) or np.isnan(price):
                 final_price = None 
             else:

services/market_services.py

@@ -13,7 +13,7 @@ models = {}
 if os.path.exists(MODELS_DIR):
     for model_file in os.listdir(MODELS_DIR):
         if model_file.endswith('.pkl'):
-            # Normalize filename to commodity name (handle slash replacement if needed)
+            # Normalize filename to commodity name
             commodity_name = model_file.replace('.pkl', '').replace('_', '/')
             try:
                 models[commodity_name] = joblib.load(os.path.join(MODELS_DIR, model_file))
@@ -48,8 +48,8 @@ def _create_features(df):
     df['rolling_mean_30'] = df['modal_price'].shift(1).rolling(window=30).mean()
     df['rolling_std_30'] = df['modal_price'].shift(1).rolling(window=30).std()
     
-    # Return features. Note: We do NOT dropna here because we need to generate 
-    # features for the future row even if it has NaNs initially.
+    # CRITICAL: Do NOT dropna() here. We need the future row (which has NaNs) to survive
+    # so we can predict it.
     return df
 
 def get_market_prediction(model, df_full, commodity, last_known_date):
@@ -69,8 +69,7 @@ def get_market_prediction(model, df_full, commodity, last_known_date):
     # 3. Combine history and future placeholder
     df_extended = pd.concat([df_daily, future_df])
     
-    # 4. Get the correct list of features from a valid sample
-    # We take the last 50 valid days to determine feature columns
+    # 4. Determine feature columns from a valid historical sample
     valid_sample = _create_features(df_daily.tail(50)).dropna()
     FEATURES = [col for col in valid_sample.columns if col != 'modal_price']
 
@@ -81,25 +80,24 @@ def get_market_prediction(model, df_full, commodity, last_known_date):
         # Safety check: ensure we have enough data for 30-day rolling window
         if len(subset) < 35: continue 
         
-        # Generate features for this specific date
+        # Generate features
         featured_subset = _create_features(subset)
         
-        # Extract the row for the current prediction date
+        # Get the row for the prediction date
         if date not in featured_subset.index: continue
         featured_row = featured_subset.loc[[date]]
         
-        # --- CRITICAL FIX START ---
-        # XGBoost requires a DataFrame with specific column names. 
-        # We explicitly recreate the DataFrame to ensure headers are present.
+        # --- FIX: Ensure strict DataFrame format for XGBoost ---
         try:
+            # Reconstruct DataFrame with explicit columns to satisfy XGBoost
             X_input = pd.DataFrame(featured_row[FEATURES].values, columns=FEATURES, index=featured_row.index)
+            
             prediction = model.predict(X_input)[0]
             df_extended.loc[date, 'modal_price'] = prediction
         except Exception as e:
-            # Log error but don't crash the loop immediately
+            # If prediction fails, we break. The NaNs will remain and be handled by the route.
             print(f"Prediction error for {date}: {e}")
             break
-        # --- CRITICAL FIX END ---
     
     daily_forecast_df = df_extended.loc[future_dates].copy()
     daily_forecast_df.rename(columns={'modal_price': 'forecast'}, inplace=True)