Update app.py

app.py

@@ -1,43 +1,38 @@
-import gradio as gr
+import streamlit as st
 import numpy as np
 import joblib
 from sklearn import datasets
 
-# Load diabetes dataset just to get feature names (order must match training)
+# Get feature names from load_diabetes
 diabetes = datasets.load_diabetes()
-feature_names = diabetes.feature_names  # ['age', 'sex', 'bmi', 'bp', 's1', ..., 's6']
+feature_names = diabetes.feature_names
 
-# Load model and scaler ONCE (fast)
+# Load model and scaler
 model = joblib.load("lr.pkl")
 scaler = joblib.load("scaler.pkl")
 
-def predict(*features):
-    # features = 10 numbers (age, sex, bmi, bp, s1...s6)
-    x = np.array(features, dtype=float).reshape(1, -1)
+st.title("Diabetes Prediction (load_diabetes + lr + scaler)")
+st.write("This app uses a saved model (lr.pkl) and scaler (scaler.pkl) to make predictions.")
+
+inputs = []
+
+st.subheader("Enter feature values")
+for name in feature_names:
+    val = st.number_input(f"{name}", value=0.0, format="%.4f")
+    inputs.append(val)
+
+if st.button("Predict"):
+    x = np.array(inputs, dtype=float).reshape(1, -1)
     x_scaled = scaler.transform(x)
 
     y = model.predict(x_scaled)[0]
 
-    # If it's a classification model (LogisticRegression)
     if hasattr(model, "predict_proba"):
-        proba = model.predict_proba(x_scaled)[0]
-        # assume binary classification
-        label = "Positive (class 1)" if int(y) == 1 else "Negative (class 0)"
-        return f"Prediction: {label}\nProbability of class 1: {proba[1]:.3f}"
+        proba = model.predict_proba(x_scaled)[0][1]
+        label = "Class 1 (Positive)" if int(y) == 1 else "Class 0 (Negative)"
+        st.subheader("Prediction")
+        st.write(label)
+        st.write(f"Probability of class 1: {proba:.3f}")
     else:
-        # Regression case (score)
-        return f"Predicted disease progression score: {float(y):.2f}"
-
-# 10 numeric inputs (same order as training)
-inputs = [gr.Number(label=name) for name in feature_names]
-
-demo = gr.Interface(
-    fn=predict,
-    inputs=inputs,
-    outputs=gr.Textbox(label="Result"),
-    title="Diabetes Prediction (load_diabetes + lr + scaler)",
-    description="Enter the 10 feature values in the same format used during training."
-)
-
-if __name__ == "__main__":
-    demo.launch()
+        st.subheader("Predicted Disease Progression Score")
+        st.write(f"{float(y):.2f}")