project.py

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np

from sklearn.linear_model import LinearRegression
from sklearn.ensemble import RandomForestRegressor
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import mean_squared_error , mean_absolute_error , r2_score
from sklearn.model_selection import train_test_split

df = pd.read_csv(r"/content/sample_data/california_housing_test.csv")

X = df["longitude"]
Y = df["latitude"]

plt.scatter(X,Y)
plt.xlabel("longitude")
plt.ylabel("latitude")
plt.title("california housing")
plt.show()

lr_model = LinearRegression()

a = df[["median_income","population","households","housing_median_age","longitude",
        "latitude","total_rooms","total_bedrooms"]]
b = df["median_house_value"]

a_trainval,a_test,b_trainval,b_test = train_test_split(a,b,test_size=0.25,random_state = 42)

a_train,a_val,b_train,b_val = train_test_split(a_trainval,b_trainval,test_size = 0.2,random_state = 42)

Scaler = StandardScaler()
a_train = Scaler.fit_transform(a_train)
a_val = Scaler.transform(a_val)
a_test = Scaler.transform(a_test)

lr_model.fit(a_train,b_train)

b_pred = lr_model.predict(a_test)

mae = mean_absolute_error(b_test,b_pred)
mse = mean_squared_error(b_test,b_pred)
rmse = np.sqrt(mse)
r2 = r2_score(b_test,b_pred)
print("Evaluation for Linear Regression")
print("Mean Absolute Error:", mae)
print("Mean Squared Error:", mse)
print("Root Mean Squared Error:", rmse)
print("R2 score:", r2)

plt.scatter(b_test,b_pred,s =10)
plt.xlabel("Actual value")
plt.ylabel("predicted value")
plt.title("california housing prediction (linear regression)")
plt.show()

rf_model = RandomForestRegressor(n_estimators = 200,n_jobs = -1,random_state = 42)

rf_model.fit(a_train, b_train)
y_pred = rf_model.predict(a_test)

mae = mean_absolute_error(b_test,y_pred)
mse = mean_squared_error(b_test,y_pred)
rmse = np.sqrt(mse)
print("Evaluation for Randoom Forest Regresssor")
print("Mean Absolute Error:", mae)
print("Mean Squared Error:", mse)
print("Root Mean Squared Error:", rmse)
print("R2:", r2_score(b_test, y_pred))

plt.scatter(b_test,y_pred,s =10)
plt.xlabel("Actual value")
plt.ylabel("predicted value")
plt.title("california housing prediction (random forest regressor)")
plt.show()

print("Prediction on new values") 

features = ["median_income","population","households","housing_median_age","longitude",
        "latitude","total_rooms","total_bedrooms"]

def get_input():
  val ={}

  for i in features:
    while True:
      try:
          val[i] = float(input(f"{i}: "))
          break
      except ValueError:
          print("Enter a valid number.")

  return val 

def model_activation(user_input,a,scaler,lr_model,rf_model):
    input_df = pd.DataFrame([user_input],columns = features)

    scaled_df = scaler.transform(input_df)
    lr_pred = lr_model.predict(scaled_df)[0]
    rf_pred = rf_model.predict(scaled_df)[0]

    print("Linear Regression Prediction:", lr_pred)
    print("Random Forest Prediction:", rf_pred)
    
    return lr_pred,rf_pred

user_input = get_input()
model_activation(user_input,a,Scaler,lr_model,rf_model)