Update app.py

app.py

@@ -13,11 +13,24 @@ if not os.path.exists(model_path):
 with open(model_path, "rb") as file:
     model = pickle.load(file)
 
+# Example mapping for categorical variables (if encoded during training)
+season_mapping = {"Kharif": 0, "Rabi": 1, "Zaid": 2}
+district_mapping = {"District A": 0, "District B": 1, "District C": 2}
+crop_mapping = {"Wheat": 0, "Rice": 1, "Maize": 2, "Sugarcane": 3}
+
 # Prediction Function
-def predict_yield(feature1, feature2, feature3):
+def predict_yield(area, season, district, crop):
     try:
+        # Convert categorical inputs to numerical if necessary
+        season_encoded = season_mapping.get(season, -1)
+        district_encoded = district_mapping.get(district, -1)
+        crop_encoded = crop_mapping.get(crop, -1)
+
+        if -1 in [season_encoded, district_encoded, crop_encoded]:
+            return "Error: Invalid categorical input value."
+
         # Prepare input
-        features = np.array([[feature1, feature2, feature3]])
+        features = np.array([[area, season_encoded, district_encoded, crop_encoded]])
         prediction = model.predict(features)
         return f"Predicted Crop Yield: {float(prediction[0]):.2f}"
     except Exception as e:
@@ -27,13 +40,14 @@ def predict_yield(feature1, feature2, feature3):
 demo = gr.Interface(
     fn=predict_yield,
     inputs=[
-        gr.Number(label="Feature 1"),
-        gr.Number(label="Feature 2"),
-        gr.Number(label="Feature 3")
+        gr.Number(label="Area (in acres)"),
+        gr.Dropdown(["Kharif", "Rabi", "Zaid"], label="Season"),
+        gr.Dropdown(["District A", "District B", "District C"], label="District"),
+        gr.Dropdown(["Wheat", "Rice", "Maize", "Sugarcane"], label="Crop"),
     ],
     outputs=gr.Textbox(label="Prediction Result"),
     title="Crop Yield Prediction",
-    description="Enter the feature values to predict crop yield.",
+    description="Enter the details to predict crop yield.",
 )
 
 if __name__ == "__main__":