Spaces:
Sleeping
Sleeping
| import streamlit as st | |
| import pickle | |
| import sklearn | |
| import pandas as pd | |
| # Load trained model, encoders, and feature column order | |
| with open("best_model.pkl", "rb") as f: | |
| model = pickle.load(f) | |
| with open("encoders.pkl", "rb") as f: | |
| encoders = pickle.load(f) | |
| with open("columns.pkl", "rb") as f: | |
| feature_columns = pickle.load(f) | |
| st.title("🧠 Autism Prediction System") | |
| st.markdown("### 📝 Please fill the following details:") | |
| # A1 to A10 - Screening Scores | |
| user_inputs = { | |
| "A1_Score": st.selectbox("A1 Score (Social interaction)", [0, 1]), | |
| "A2_Score": st.selectbox("A2 Score (Eye contact)", [0, 1]), | |
| "A3_Score": st.selectbox("A3 Score (Responds to name)", [0, 1]), | |
| "A4_Score": st.selectbox("A4 Score (Repetitive behavior)", [0, 1]), | |
| "A5_Score": st.selectbox("A5 Score (Preference for routine)", [0, 1]), | |
| "A6_Score": st.selectbox("A6 Score (Sensitive to sounds, lights, etc.)", [0, 1]), | |
| "A7_Score": st.selectbox("A7 Score (Speech or communication delay)", [0, 1]), | |
| "A8_Score": st.selectbox("A8 Score (Struggles recognizing emotions)", [0, 1]), | |
| "A9_Score": st.selectbox("A9 Score (Unusual interests or fixations)", [0, 1]), | |
| "A10_Score": st.selectbox("A10 Score (Strong attention to detail)", [0, 1]), | |
| # Demographic and medical history | |
| "age": st.number_input("Age (in years)", min_value=1, max_value=120), | |
| "gender": st.selectbox("Gender (Biological sex)", encoders["gender"].classes_), | |
| "ethnicity": st.selectbox("Ethnicity (Cultural background)", encoders["ethnicity"].classes_), | |
| "jaundice": st.selectbox("Jaundice at Birth (Medical history)", encoders["jaundice"].classes_), | |
| "contry_of_res": st.selectbox("Country of Residence (Current location)", encoders["contry_of_res"].classes_), | |
| "austim": st.selectbox("Family history of Autism?", encoders["austim"].classes_), | |
| "used_app_before": st.selectbox("Used Screening App Before?", encoders["used_app_before"].classes_), | |
| "result": st.number_input("Screening Score Result (Total score from A1-A10)", min_value=0, max_value=20), | |
| "relation": st.selectbox("Relation to Patient (Who is filling this form?)", encoders["relation"].classes_), | |
| } | |
| if st.button("🔍 Predict"): | |
| try: | |
| # Convert to DataFrame | |
| input_df = pd.DataFrame([user_inputs]) | |
| # Encode categorical columns | |
| for col in encoders: | |
| input_df[col] = encoders[col].transform(input_df[col]) | |
| # Reorder columns as used in training | |
| input_df = input_df[feature_columns] | |
| # Predict | |
| prediction = model.predict(input_df)[0] | |
| result = "🧠 ASD Positive" if prediction == 1 else "✅ ASD Negative" | |
| st.success(f"Prediction: {result}") | |
| except Exception as e: | |
| st.error(f"⚠️ Error during prediction: {e}") | |