app.py
--- CHANGED BLOCK START ---
import gradio as gr
import joblib
import spaces
We load the model once when the app starts
deployed_lr = joblib.load('my_first_ml_model.pkl')
--- ZERO-GPU DECORATOR AND PREDICTION LOGIC ---
@spaces.GPU
def predict_rent(size_of_prop):
# The model expects a 2D array: [[size]]
prediction = deployed_lr.predict([[size_of_prop]])
# Extract the single prediction value and format it
return f"Estimated Rent: {prediction[0]:.2f}"
Create the web interface
interface = gr.Interface(
fn=predict_rent,
inputs=gr.Number(label="Please Enter the Size of Your Property for rent"),
outputs=gr.Text(label="Predicted Rent"),
title="Property Rent Predictor",
description="Enter the property size to get a rent estimate powered by Machine Learning."
)
if name == "main":
interface.launch()
--- CHANGED BLOCK END ---
