evalml.pipelines.CatBoostClassificationPipeline

class evalml.pipelines.CatBoostClassificationPipeline(parameters, objective, random_state=0)

CatBoost Pipeline for both binary and multiclass classification. CatBoost is an open-source library and natively supports categorical features.

For more information, check out https://catboost.ai/ Note: impute_strategy must support both string and numeric data

name = 'Cat Boost Classification Pipeline'

summary = 'CatBoost Classifier w/ Simple Imputer'

component_graph = ['Simple Imputer', <evalml.pipelines.components.estimators.classifiers.catboost_classifier.CatBoostClassifier object>]

supported_problem_types = ['binary', 'multiclass']

model_family = 'catboost'

hyperparameters = {'eta': Real(low=0, high=1, prior='uniform', transform='identity'), 'impute_strategy': ['most_frequent'], 'max_depth': Integer(low=1, high=16, prior='uniform', transform='identity'), 'n_estimators': Integer(low=10, high=1000, prior='uniform', transform='identity')}

custom_hyperparameters = {'impute_strategy': ['most_frequent']}

Instance attributes

feature_importances	Return feature importances.
parameters	Returns parameter dictionary for this pipeline

Methods:

__init__	Machine learning pipeline made out of transformers and a estimator.
describe	Outputs pipeline details including component parameters
feature_importance_graph	Generate a bar graph of the pipeline’s feature importances
fit	Build a model
get_component	Returns component by name
graph	Generate an image representing the pipeline graph
load	Loads pipeline at file path
predict	Make predictions using selected features.
predict_proba	Make probability estimates for labels.
save	Saves pipeline at file path
score	Evaluate model performance on current and additional objectives