Update app.py

app.py

@@ -1,67 +1,28 @@
 import gradio as gr
-import numpy as np
 import joblib
-import os
-import pandas as pd
-
-from evidently.report import Report
-from evidently.metrics import DataDriftPreset
+import numpy as np
 
 # Load trained model
-model_path = "model/house_price_model.pkl"
-if not os.path.exists(model_path):
-    raise FileNotFoundError("Model file not found. Make sure 'model/house_price_model.pkl' exists.")
-model = joblib.load(model_path)
-
-# Sample reference and current data (fake for demo, replace with real if available)
-reference_data = pd.DataFrame({
-    "RM": np.random.normal(6, 0.5, 100),
-    "LSTAT": np.random.normal(12, 5, 100),
-    "PTRATIO": np.random.normal(18, 1.5, 100),
-})
-current_data = pd.DataFrame({
-    "RM": np.random.normal(6.2, 0.6, 100),
-    "LSTAT": np.random.normal(14, 5, 100),
-    "PTRATIO": np.random.normal(19, 2, 100),
-})
-
-# Generate drift report and save it
-def generate_drift_report():
-    report = Report(metrics=[DataDriftPreset()])
-    report.run(reference_data=reference_data, current_data=current_data)
-    report.save_html("evidently_report.html")
-    return "evidently_report.html"
+model = joblib.load("model/house_price_model.pkl")
 
-# Predict function
-def predict(rm, lstat, ptratio):
+# Define prediction function
+def predict_price(rm, lstat, ptratio):
     input_data = np.array([[rm, lstat, ptratio]])
     prediction = model.predict(input_data)[0]
     return round(prediction, 2)
 
-# Wrapper to return the HTML drift report
-def show_report():
-    if not os.path.exists("evidently_report.html"):
-        generate_drift_report()
-    with open("evidently_report.html", "r", encoding="utf-8") as f:
-        html_content = f.read()
-    return gr.HTML(html_content)
-
-# Gradio UI setup
-with gr.Blocks() as demo:
-    gr.Markdown("# 🏡 Boston Housing Price Predictor + Drift Report")
-
-    with gr.Tab("🔮 Predict Price"):
-        rm = gr.Slider(3.0, 9.0, step=0.1, label="Average Rooms (RM)")
-        lstat = gr.Slider(1.0, 40.0, step=0.1, label="Lower Status Population (%) (LSTAT)")
-        ptratio = gr.Slider(12.0, 22.0, step=0.1, label="Pupil-Teacher Ratio (PTRATIO)")
-        predict_btn = gr.Button("Predict")
-        result = gr.Number(label="🏠 Predicted Price ($1000s)")
-        predict_btn.click(predict, inputs=[rm, lstat, ptratio], outputs=result)
-
-    with gr.Tab("📊 View Data Drift Report"):
-        report_output = gr.HTML()
-        generate_btn = gr.Button("Generate Report")
-        generate_btn.click(fn=show_report, outputs=report_output)
-
-if __name__ == "__main__":
-    demo.launch()
+# Define Gradio interface
+iface = gr.Interface(
+    fn=predict_price,
+    inputs=[
+        gr.Slider(3.0, 9.0, step=0.1, label="Average Rooms (RM)"),
+        gr.Slider(1.0, 40.0, step=0.1, label="Lower Status Population (%) (LSTAT)"),
+        gr.Slider(12.0, 22.0, step=0.1, label="Pupil-Teacher Ratio (PTRATIO)")
+    ],
+    outputs=gr.Number(label="Predicted House Price (in $1000s)"),
+    title="🏠 Boston Housing Price Predictor",
+    description="Enter the house features to predict the price using a linear regression model.",
+)
+
+# Launch app
+iface.launch()