Upgrade to RFR model

app.py

@@ -2,6 +2,7 @@ import gradio as gr
 import pandas as pd
 import numpy as np
 from sklearn.linear_model import LinearRegression
+from sklearn.ensemble import RandomForestRegressor
 from sklearn.datasets import fetch_california_housing
 
 def trainme(MedInc, HouseAge, AveRooms, AveBedrms, Population, AveOccup, Latitude, Longitude, Price):
@@ -21,9 +22,9 @@ def trainme(MedInc, HouseAge, AveRooms, AveBedrms, Population, AveOccup, Latitud
   X = housing_df.drop('Price', axis=1).to_numpy()
   
   #create a machine learning model and train it
-  lineareg = LinearRegression()
-  lineareg.fit(X,y)
-  
+  regressor = RandomForestRegressor(n_estimators = 100, random_state = 0)
+  regressor.fit(X,y)
+
   #Create a redable/clean feature list
   clean_features = ['Median income',
  'Median house age',
@@ -39,12 +40,12 @@ def trainme(MedInc, HouseAge, AveRooms, AveBedrms, Population, AveOccup, Latitud
   explainer = lime_tabular.LimeTabularExplainer(X, mode="regression", feature_names=clean_features)
 
   #Create the expl
-  explanation = explainer.explain_instance(X[0], lineareg.predict, num_features=len(housing.feature_names))
+  explanation = explainer.explain_instance(X[0], regressor.predict, num_features=len(housing.feature_names))
   listing = explanation.as_list()
   #explanation.show_in_notebook(show_table=True)
 
   #Get pred and actual scores
-  Pred_value = lineareg.predict(X[0].reshape(1,-1))*100000
+  Pred_value = regressor.predict(X[0].reshape(1,-1))*100000
   Actual_value = y[0]*100000
   Error_rate = ((Pred_value - Actual_value)/Actual_value) *100