JPMML-CatBoost

Java library and command-line application for converting CatBoost models to PMML.

Prerequisites

• CatBoost 0.15.2 or newer.
• Java 1.8 or newer.

Installation

Enter the project root directory and build using Apache Maven:

mvn clean install

The build produces an executable uber-JAR file target/jpmml-catboost-executable-1.0-SNAPSHOT.jar.

Usage

A typical workflow can be summarized as follows:

1. Use CatBoost to train a model.
2. Save the model to a CBM file in a local filesystem.
3. Use the JPMML-CatBoost command-line converter application to turn this CBM file to a PMML file.

The CatBoost side of operations

Using the catboost package to train a regression model for the example Boston housing dataset:

from pandas import DataFrame, Series
from sklearn.datasets import load_boston

boston = load_boston()

X = DataFrame(boston.data, columns = boston.feature_names)
y = Series(boston.target)

from catboost import CatBoostRegressor

cb = CatBoostRegressor(iterations = 131)
cb.fit(X, y)

cb.save_model("catboost.cbm", format = "cbm")

The JPMML-CatBoost side of operations

Converting the CBM file catboost.cbm to a PMML file catboost.pmml:

java -jar target/jpmml-catboost-executable-1.0-SNAPSHOT.jar --cbm-input catboost.cbm --pmml-output catboost.pmml

Getting help:

java -jar target/jpmml-catboost-executable-1.0-SNAPSHOT.jar --help

License

JPMML-CatBoost is dual-licensed under the GNU Affero General Public License (AGPL) version 3.0, and a commercial license.

Additional information

JPMML-CatBoost is developed and maintained by Openscoring Ltd, Estonia.

Interested in using JPMML software in your application? Please contact info@openscoring.io