Upload folder using huggingface_hub

app.py

@@ -57,7 +57,7 @@ if st.button("Predict Engine Condition", type="primary"):
         load_index = float(engine_rpm * fuel_pressure / 100)
         thermal_stress = float(coolant_temp - lub_oil_temp)
 
-        # Create DataFrame with EXACT column names and order from your error message
+        # Create DataFrame with EXACT column names and order
         input_data = pd.DataFrame([[
             engine_rpm, 
             lub_oil_pressure, 
@@ -79,15 +79,23 @@ if st.button("Predict Engine Condition", type="primary"):
         st.divider()
         st.subheader("Results")
 
-        if prediction == 0: # Assuming 0 is Good based on typical maintenance datasets
+        # Visual Feedback based on Prediction
+        if prediction == 0: # Assuming 0 is Good
              st.success("**Status: Engine is in Good Condition**")
         else:
              st.error("**Status: Maintenance Required (Potential Fault)**")
 
-        # Confidence Visual
-        health_score = prediction_proba[0] # Probability of "Healthy"
+        # --- FIX START: Handling st.progress error ---
+        # 1. Extract the probability for the "Healthy" class (Class 0)
+        health_score = float(prediction_proba[0])
+        
+        # 2. Clip the value to ensure it is strictly between 0.0 and 1.0
+        # This prevents crashes if the model returns 1.0000001 or -0.0000001
+        safe_progress_value = max(0.0, min(1.0, health_score))
+
         st.write(f"**Confidence Score:** {max(prediction_proba)*100:.2f}%")
-        st.progress(health_score)
+        st.progress(safe_progress_value)
+        # --- FIX END ---
         
     else:
         st.error("Model could not be loaded. Check your Hugging Face Repo ID.")