Update app.py

app.py

@@ -6,44 +6,61 @@ import pandas as pd
 with open("model.pkl", "rb") as f:
     model = pickle.load(f)
 
-# Define prediction function
+# Define categorical mappings (must match training!)
+label_encoders = {
+    "School_Type": {"Public": 0, "Private": 1},
+    "School_Location": {"Urban": 0, "Rural": 1},
+    "Extra_Tutorials": {"No": 0, "Yes": 1},
+    "Access_To_Learning_Materials": {"No": 0, "Yes": 1},
+    "Parent_Involvement": {"Low": 0, "Medium": 1, "High": 2},
+    "IT_Knowledge": {"Low": 0, "Medium": 1, "High": 2},
+    "Gender": {"Male": 0, "Female": 1},
+    "Socioeconomic_Status": {"Low": 0, "Medium": 1, "High": 2},
+    "Parent_Education_Level": {"None": 0, "Primary": 1, "Secondary": 2, "Tertiary": 3}
+}
+
+# Prediction function
 def predict_jamb(study_hours, attendance, teacher_quality, distance,
                  school_type, school_location, extra_tutorials, learning_materials,
                  parent_involvement, it_knowledge, age, gender, socioeconomic_status,
                  parent_education, assignments_completed):
 
-    # Create dataframe from inputs
-    input_data = pd.DataFrame({
-        "Study_Hours_Per_Week": [study_hours],
-        "Attendance_Rate": [attendance],
-        "Teacher_Quality": [teacher_quality],
-        "Distance_To_School": [distance],
-        "School_Type": [school_type],
-        "School_Location": [school_location],
-        "Extra_Tutorials": [extra_tutorials],
-        "Access_To_Learning_Materials": [learning_materials],
-        "Parent_Involvement": [parent_involvement],
-        "IT_Knowledge": [it_knowledge],
-        "Age": [age],
-        "Gender": [gender],
-        "Socioeconomic_Status": [socioeconomic_status],
-        "Parent_Education_Level": [parent_education],
-        "Assignments_Completed": [assignments_completed]
-    })
+    # Encode categorical inputs
+    input_dict = {
+        "Study_Hours_Per_Week": study_hours,
+        "Attendance_Rate": attendance,
+        "Teacher_Quality": teacher_quality,
+        "Distance_To_School": distance,
+        "School_Type": label_encoders["School_Type"][school_type],
+        "School_Location": label_encoders["School_Location"][school_location],
+        "Extra_Tutorials": label_encoders["Extra_Tutorials"][extra_tutorials],
+        "Access_To_Learning_Materials": label_encoders["Access_To_Learning_Materials"][learning_materials],
+        "Parent_Involvement": label_encoders["Parent_Involvement"][parent_involvement],
+        "IT_Knowledge": label_encoders["IT_Knowledge"][it_knowledge],
+        "Age": age,
+        "Gender": label_encoders["Gender"][gender],
+        "Socioeconomic_Status": label_encoders["Socioeconomic_Status"][socioeconomic_status],
+        "Parent_Education_Level": label_encoders["Parent_Education_Level"][parent_education],
+        "Assignments_Completed": assignments_completed
+    }
+
+    # Convert to dataframe
+    input_data = pd.DataFrame([input_dict])
 
+    # Predict
     prediction = model.predict(input_data)[0]
 
     if prediction == 0:
-        return "Below Average"
+        return "📉 Below Average"
     elif prediction == 1:
-        return "Average"
+        return "⚖️ Average"
     else:
-        return "Pass"
+        return "✅ Pass"
 
-# Gradio interface
+# Gradio UI
 with gr.Blocks() as demo:
     gr.Markdown("# 🎓 JAMB Score Prediction App")
-    gr.Markdown("Predict whether a student's JAMB score will be **Below Average**, **Average**, or **Pass**.")
+    gr.Markdown("Predict student's JAMB performance (Below Average, Average, Pass).")
 
     with gr.Row():
         study_hours = gr.Slider(0, 40, value=10, label="Study Hours per Week")
@@ -80,6 +97,5 @@ with gr.Blocks() as demo:
         outputs=output
     )
 
-# Launch app
 if __name__ == "__main__":
     demo.launch()