Update app.py

app.py

@@ -4,25 +4,28 @@ import joblib
 from huggingface_hub import hf_hub_download
 import os
 
-# Define the model path and file name on Hugging Face Hub
+# Define the model path and file name on Hugging Face Hub.
+# The filename now includes the 'data/' subdirectory.
 repo_id = "Web4/LS-W4-Mini-RF_Addiction_Impact"
-model_file = "LS-W4-Mini-RF_Addiction_Impact.joblib"
+model_file = "data/LS-W4-Mini-RF_Addiction_Impact.joblib"
 
-# Get the Hugging Face token from environment variables
+# Get the Hugging Face token from environment variables.
+# This is required for gated repositories.
 token = os.environ.get("HF_TOKEN")
 
-# Download the model file from the Hugging Face Hub using the token
+# Download the model file from the Hugging Face Hub using the token.
 try:
     model_path = hf_hub_download(repo_id=repo_id, filename=model_file, token=token)
     print(f"Model downloaded to: {model_path}")
 except Exception as e:
+    # This error indicates that the file was not found or access was denied.
     print(f"Error downloading model: {e}")
     raise
 
-# Load the scikit-learn pipeline from the downloaded joblib file
+# Load the scikit-learn pipeline from the downloaded joblib file.
 pipeline = joblib.load(model_path)
 
-# Define the prediction function for the Gradio interface
+# Define the prediction function for the Gradio interface.
 def predict_impact(
     gender,
     academic_level,
@@ -35,7 +38,11 @@ def predict_impact(
     addicted_score,
     conflicts_over_social_media
 ):
-    # Create a pandas DataFrame from the user inputs
+    """
+    Takes user inputs, creates a pandas DataFrame, and makes a prediction
+    using the loaded scikit-learn pipeline.
+    """
+    # Create a pandas DataFrame from the user inputs.
     input_data = pd.DataFrame({
         'Gender': [gender],
         'Academic_Level': [academic_level],
@@ -49,16 +56,16 @@ def predict_impact(
         'Conflicts_Over_Social_Media': [conflicts_over_social_media]
     })
     
-    # Make a prediction using the loaded pipeline
+    # Make a prediction. The pipeline handles the preprocessing automatically.
     prediction = pipeline.predict(input_data)[0]
     
-    # Return a user-friendly result based on the prediction
+    # Return a user-friendly result based on the prediction.
     if prediction == 1:
         return "Prediction: Yes, social media use is likely to impact academic performance."
     else:
         return "Prediction: No, social media use is likely not to impact academic performance."
 
-# Define the Gradio interface components
+# Define the Gradio interface components.
 demo = gr.Interface(
     fn=predict_impact,
     inputs=[
@@ -75,9 +82,9 @@ demo = gr.Interface(
     ],
     outputs="text",
     title="Social Media Addiction Impact on Academic Performance",
-    description="A Random Forest model to predict if social media use impacts a student's academic performance."
+    description="A Random Forest model to predict if social media use impacts a student's academic performance. This is not a diagnostic tool."
 )
 
-# Launch the Gradio app
+# Launch the Gradio app.
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()