train.py

import pandas as pd
from sklearn.preprocessing import LabelEncoder
from sklearn.model_selection import train_test_split
from sklearn.ensemble import GradientBoostingRegressor
from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score, median_absolute_error
from joblib import dump

# Load the dataset
df = pd.read_csv('cleaned_housesTRAIN.csv')

# Apply label encoding to 'Area' and 'Suburb'
le_area = LabelEncoder()
df['Area'] = le_area.fit_transform(df['Area'])

le_suburb = LabelEncoder()
df['Suburb'] = le_suburb.fit_transform(df['Suburb'])

# Save the label encoders
dump(le_area, 'le_area.joblib')
dump(le_suburb, 'le_suburb.joblib')

# Shuffle the dataframe
df = df.sample(frac=1)

# Split the data into features (X) and target (y)
X = df.drop('Rent', axis=1)
y = df['Rent']

# Split the data into training and test sets (90/10 split)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=42)

# Create a Gradient Boosting regressor with specified hyperparameters
gb = GradientBoostingRegressor(n_estimators=850, learning_rate=0.195, max_depth=7, random_state=42)

# Train the model
gb.fit(X_train, y_train)

# Make predictions on the test set
y_pred = gb.predict(X_test)

# Calculate MAE, MSE, and R2
mae = mean_absolute_error(y_test, y_pred)
mse = mean_squared_error(y_test, y_pred)
r2 = r2_score(y_test, y_pred)
medae = median_absolute_error(y_test, y_pred)

print(f"MAE: {mae}, MSE: {mse}, R2: {r2}, MedAE: {medae}")

# Save the model
dump(gb, 'bestmodelyet.joblib')