Learn CATBOOST with Real Code Examples

from catboost import CatBoostClassifier
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

# Load dataset
data = load_iris()
X_train, X_test, y_train, y_test = train_test_split(data.data, data.target, test_size=0.2, random_state=42)

# Define model
model = CatBoostClassifier(iterations=100, learning_rate=0.1, depth=3, verbose=0)

# Train model
model.fit(X_train, y_train)

# Predict
y_pred = model.predict(X_test)
print('Accuracy:', accuracy_score(y_test, y_pred))

A minimal CatBoost example performing classification on the Iris dataset.

from catboost import CatBoostRegressor
from sklearn.datasets import make_regression
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error

# Generate data
X, y = make_regression(n_samples=200, n_features=5, noise=0.1, random_state=42)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Define model
model = CatBoostRegressor(iterations=200, learning_rate=0.05, depth=4, verbose=0)

# Train model
model.fit(X_train, y_train)

# Predict
y_pred = model.predict(X_test)
print('MSE:', mean_squared_error(y_test, y_pred))

A simple regression using CatBoost on synthetic data.

from catboost import CatBoostClassifier
from sklearn.datasets import load_wine
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

# Load dataset
data = load_wine()
X_train, X_test, y_train, y_test = train_test_split(data.data, data.target, test_size=0.3, random_state=42)

# Define model
model = CatBoostClassifier(iterations=150, learning_rate=0.1, depth=5, verbose=0, loss_function='MultiClass')

# Train model
model.fit(X_train, y_train)

# Predict
y_pred = model.predict(X_test)
print('Accuracy:', accuracy_score(y_test, y_pred))

CatBoost handling multi-class classification on the Wine dataset.

from catboost import CatBoostClassifier, Pool
import pandas as pd

# Sample data
data = pd.DataFrame({
    'feature_num': [1,2,3,4,5,6],
    'feature_cat': ['A','B','A','B','C','C'],
    'label': [0,1,0,1,0,1]
})
X = data[['feature_num','feature_cat']]
y = data['label']

# Define categorical features
cat_features = ['feature_cat']

# Create Pool
data_pool = Pool(X, y, cat_features=cat_features)

# Define model
model = CatBoostClassifier(iterations=50, learning_rate=0.1, depth=3, verbose=0)

# Train model
model.fit(data_pool)

# Predict
y_pred = model.predict(X)
print('Predictions:', y_pred)

Using CatBoost with categorical features in a classification task.

from catboost import CatBoostClassifier
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split

# Load dataset
data = load_iris()
X_train, X_valid, y_train, y_valid = train_test_split(data.data, data.target, test_size=0.2, random_state=42)

# Define model
model = CatBoostClassifier(iterations=500, learning_rate=0.05, depth=4, verbose=0)

# Train with early stopping
model.fit(X_train, y_train, eval_set=(X_valid, y_valid), early_stopping_rounds=20)

# Predict
y_pred = model.predict(X_valid)
print('Predictions:', y_pred)

CatBoost training with early stopping based on validation set.

from catboost import CatBoostRanker
import numpy as np

# Sample ranking data
X_train = np.random.rand(6,3)
y_train = np.array([1,2,3,1,2,3])
group_id = np.array([0,0,0,1,1,1])

# Define model
model = CatBoostRanker(iterations=100, learning_rate=0.1, depth=3, verbose=0)

# Train model
model.fit(X_train, y_train, group_id=group_id)

# Predict
y_pred = model.predict(X_train)
print('Ranking Predictions:', y_pred)

Using CatBoost for a simple ranking problem.

from catboost import CatBoostClassifier
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split

# Load dataset
data = load_iris()
X_train, X_test, y_train, y_test = train_test_split(data.data, data.target, test_size=0.2, random_state=42)

# Custom loss function (logloss)
model = CatBoostClassifier(loss_function='Logloss', iterations=200, learning_rate=0.05, depth=4, verbose=0)

# Train model
model.fit(X_train, y_train)

# Predict
y_pred = model.predict(X_test)
print('Predictions:', y_pred)

Example showing how to use a custom loss function with CatBoost.

from catboost import CatBoostClassifier
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split

# Load dataset
data = load_iris()
X_train, X_test, y_train, y_test = train_test_split(data.data, data.target, test_size=0.2, random_state=42)

# Define model
model = CatBoostClassifier(iterations=100, learning_rate=0.1, depth=3, verbose=0)

# Train model
model.fit(X_train, y_train)

# Feature importance
importance = model.get_feature_importance()
print('Feature Importance:', importance)

Compute and display feature importance using CatBoost.

from catboost import CatBoostClassifier
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split, GridSearchCV

# Load dataset
data = load_iris()
X_train, X_test, y_train, y_test = train_test_split(data.data, data.target, test_size=0.2, random_state=42)

# Define model
model = CatBoostClassifier(verbose=0)

# Define hyperparameter grid
param_grid = {'depth':[3,4,5], 'learning_rate':[0.05,0.1], 'iterations':[100,200]}

# Grid Search
grid_search = GridSearchCV(estimator=model, param_grid=param_grid, cv=3)
grid_search.fit(X_train, y_train)
print('Best Params:', grid_search.best_params_)

Performing hyperparameter tuning using Grid Search with CatBoost.

from catboost import CatBoostClassifier
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split

# Load dataset
data = load_iris()
X_train, X_test, y_train, y_test = train_test_split(data.data, data.target, test_size=0.2, random_state=42)

# Define model
model = CatBoostClassifier(iterations=100, learning_rate=0.1, depth=3, verbose=0)

# Train model
model.fit(X_train, y_train)

# Save model
model.save_model('catboost_model.cbm')

# Load model
loaded_model = CatBoostClassifier()
loaded_model.load_model('catboost_model.cbm')

# Predict
y_pred = loaded_model.predict(X_test)
print('Accuracy:', y_pred)

Saving and loading a trained CatBoost model.