20/02/2025, 12:12 Untitled1.

ipynb - Colab

import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder, StandardScaler, PolynomialFeatures
from sklearn.impute import SimpleImputer
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error

# Load the dataset

file_path = "housing.csv" # Update this if needed
df = pd.read_csv(file_path)
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 20640 entries, 0 to 20639
Data columns (total 10 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 longitude 20640 non-null float64
1 latitude 20640 non-null float64
2 housing_median_age 20640 non-null float64
3 total_rooms 20640 non-null float64
4 total_bedrooms 20433 non-null float64
5 population 20640 non-null float64
6 households 20640 non-null float64
7 median_income 20640 non-null float64
8 median_house_value 20640 non-null float64
9 ocean_proximity 20640 non-null object
dtypes: float64(9), object(1)
memory usage: 1.6+ MB

# Handle missing values in 'total_bedrooms' using median imputation

imputer = SimpleImputer(strategy="median")
df["total_bedrooms"] = imputer.fit_transform(df[["total_bedrooms"]])
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 20640 entries, 0 to 20639
Data columns (total 10 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 longitude 20640 non-null float64
1 latitude 20640 non-null float64
2 housing_median_age 20640 non-null float64
3 total_rooms 20640 non-null float64
4 total_bedrooms 20640 non-null float64
5 population 20640 non-null float64
6 households 20640 non-null float64
7 median_income 20640 non-null float64
8 median_house_value 20640 non-null float64
9 ocean_proximity 20640 non-null object
dtypes: float64(9), object(1)
memory usage: 1.6+ MB

# One-hot encode 'ocean_proximity'

encoder = OneHotEncoder(sparse_output=False, drop="first") # Fixed parameter
encoded_ocean_proximity = encoder.fit_transform(df[["ocean_proximity"]])

# Convert encoded categories to a DataFrame

encoded_df = pd.DataFrame(encoded_ocean_proximity, columns=encoder.get_feature_names_out())

# Combine numerical and categorical features

df_final = pd.concat([df.drop(columns=["ocean_proximity"]), encoded_df], axis=1)

# Feature Engineering: Creating new features

df_final["rooms_per_household"] = df_final["total_rooms"] / df_final["households"]
df_final["bedrooms_per_room"] = df_final["total_bedrooms"] / df_final["total_rooms"]
df_final["population_per_household"] = df_final["population"] / df_final["households"]

# Feature Selection: Drop weakly correlated features

correlation_matrix = df_final.corr()
correlations = correlation_matrix["median_house_value"].sort_values(ascending=False)
low_corr_features = correlations[abs(correlations) < 0.1].index.tolist()
df_optimized = df_final.drop(columns=low_corr_features)

# Remove outliers: Keep only data within 1.5 * IQR range

Q1 = df optimized.quantile(0.25)
https://colab.research.google.com/drive/1WBO5ExB0N6FFLotiWgTunr-opDpq45OO#scrollTo=N5iBhMIREy_7&printMode=true 1/3
20/02/2025, 12:12 Untitled1.ipynb - Colab
Q1 df_optimized.quantile(0.25)
Q3 = df_optimized.quantile(0.75)
IQR = Q3 - Q1
df_filtered = df_optimized[~((df_optimized < (Q1 - 1.5 * IQR)) | (df_optimized > (Q3 + 1.5 * IQR))).any(axis=1)]

# Separate features and target variable

X_opt = df_filtered.drop(columns=["median_house_value"])
y_opt = df_filtered["median_house_value"]

# Split into training (80%) and test (20%) sets

X_train_opt, X_test_opt, y_train_opt, y_test_opt = train_test_split(X_opt, y_opt, test_size=0.2)
X_train_opt.info(),X_test_opt.info()

<class 'pandas.core.frame.DataFrame'>
Index: 10588 entries, 13355 to 3065
Data columns (total 9 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 latitude 10588 non-null float64
1 housing_median_age 10588 non-null float64
2 total_rooms 10588 non-null float64
3 median_income 10588 non-null float64
4 ocean_proximity_INLAND 10588 non-null float64
5 ocean_proximity_NEAR BAY 10588 non-null float64
6 ocean_proximity_NEAR OCEAN 10588 non-null float64
7 rooms_per_household 10588 non-null float64
8 bedrooms_per_room 10588 non-null float64
dtypes: float64(9)
memory usage: 827.2 KB
<class 'pandas.core.frame.DataFrame'>
Index: 2647 entries, 7842 to 2441
Data columns (total 9 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 latitude 2647 non-null float64
1 housing_median_age 2647 non-null float64
2 total_rooms 2647 non-null float64
3 median_income 2647 non-null float64
4 ocean_proximity_INLAND 2647 non-null float64
5 ocean_proximity_NEAR BAY 2647 non-null float64
6 ocean_proximity_NEAR OCEAN 2647 non-null float64
7 rooms_per_household 2647 non-null float64
8 bedrooms_per_room 2647 non-null float64
dtypes: float64(9)
memory usage: 206.8 KB
(None, None)

# Standardize numerical features

scaler = StandardScaler()
X_train_scaled_opt = scaler.fit_transform(X_train_opt)
X_test_scaled_opt = scaler.transform(X_test_opt)

# Apply Polynomial Features (degree=2)

poly = PolynomialFeatures(degree=2, include_bias=False)
X_train_poly = poly.fit_transform(X_train_scaled_opt)
X_test_poly = poly.transform(X_test_scaled_opt)

# Train the Linear Regression model on polynomial features

model = LinearRegression()
model.fit(X_train_poly, y_train_opt)

# Make predictions
y_train_poly_pred = model.predict(X_train_poly)
y_test_poly_pred = model.predict(X_test_poly)

# Compute Mean Squared Error (MSE)

train_mse_poly = mean_squared_error(y_train_opt, y_train_poly_pred)
test_mse_poly = mean_squared_error(y_test_opt, y_test_poly_pred)

# Print results
print(f"Training MSE: {train_mse_poly:.2f}")
print(f"Test MSE: {test_mse_poly:.2f}")

Training MSE: 2872509441.01

Test MSE: 2915943708.52

Start coding or generate with AI.

https://colab.research.google.com/drive/1WBO5ExB0N6FFLotiWgTunr-opDpq45OO#scrollTo=N5iBhMIREy_7&printMode=true 2/3
20/02/2025, 12:12 Untitled1.ipynb - Colab

https://colab.research.google.com/drive/1WBO5ExB0N6FFLotiWgTunr-opDpq45OO#scrollTo=N5iBhMIREy_7&printMode=true 3/3