I am using CatBoostRegressor for my research as you can see below:
''' model = CatBoostRegressor(loss_function='RMSE', random_seed= seed, verbose=False)
model.fit(train_XN, train_Y) '''
My question is whether CatBoostRegressor automatically uses a portion of the training data for validation and early stopping in case of overfitting even if I don't explicitly specify an evaluation set (eval_set=(X_val, y_val)) during the fitting process.