Update app.py

app.py

@@ -1,32 +1,17 @@
+import pandas as pd
 import streamlit as st
 import numpy as np
 import pickle
-from sklearn.compose import ColumnTransformer
-from sklearn.pipeline import Pipeline
-from sklearn.preprocessing import OneHotEncoder, OrdinalEncoder, RobustScaler
 
-# Load Model
+# Load model
 try:
-    with open("final_model_2.pkl", "rb") as f:
+    with open("final_model_3.pkl", "rb") as f:
         model = pickle.load(f)
     st.success("✅ Model loaded successfully!")
 except FileNotFoundError:
     st.error("❌ Model file not found! Please upload `final_model.pkl`.")
     model = None
 
-# Define your preprocessing pipeline (assuming it's already defined in your model)
-# Example preprocessing pipeline
-nom_pl = Pipeline(steps=[('Encoding', OneHotEncoder(sparse_output=False, handle_unknown='ignore'))])
-ord_pl = Pipeline(steps=[('Encoding', OrdinalEncoder())])
-scale_pl = Pipeline(steps=[('Scaling', RobustScaler())])
-
-ct = ColumnTransformer(
-    transformers=[
-        ('nom_pl', nom_pl, [0, 4]),  # Example for categorical columns
-        ('ord_pl', ord_pl, [1, 2, 3, 6, 7]),  # Example for ordinal columns
-        ('scale_pl', scale_pl, [5, 8, 9])  # Example for numerical columns
-    ])
-
 # Title of the application
 st.markdown("<h1 class='title'>🏡 House Price Predictor</h1>", unsafe_allow_html=True)
 
@@ -44,16 +29,14 @@ with st.expander("🔹 **Property Details**", expanded=True):
     LATITUDE = st.number_input("LATITUDE", min_value=-121.761248, max_value=152.962676, value=77.324137)
 
 if st.button("🔍 Predict Price"):
-    # Create input data for prediction
-    input_data = [[POSTED_BY, UNDER_CONSTRUCTION, RERA, BHK_NO_, BHK_OR_RK, SQUARE_FT,
-                   READY_TO_MOVE, RESALE, LONGITUDE, LATITUDE]]
-
-    # Transform input data using the ColumnTransformer (fit before prediction)
-    # You need to transform the input data before prediction
-    input_data_transformed = ct.transform(input_data)
+    # Create input data for prediction, using pandas DataFrame with proper column names
+    input_data = pd.DataFrame([[POSTED_BY, UNDER_CONSTRUCTION, RERA, BHK_NO_, BHK_OR_RK, SQUARE_FT,
+                               READY_TO_MOVE, RESALE, LONGITUDE, LATITUDE]],
+                              columns=["POSTED_BY", "UNDER_CONSTRUCTION", "RERA", "BHK_NO_", "BHK_OR_RK", "SQUARE_FT", 
+                                       "READY_TO_MOVE", "RESALE", "LONGITUDE", "LATITUDE"])
 
     # Make prediction using the model
-    predicted_price = model.predict(input_data_transformed)[0]
+    predicted_price = model.predict(input_data)[0]
 
     # Display predicted price
     st.markdown(f"<div class='result-box'>🏠 Predicted Price: ₹ {predicted_price:.2f} Lakhs</div>", unsafe_allow_html=True)