Model Builders for the Gradient Boosting Frameworks
Note
Scikit-learn patching functionality in daal4py was deprecated and moved to a separate package, Intel(R) Extension for Scikit-learn*. All future patches will be available only in Intel(R) Extension for Scikit-learn*. Use the scikit-learn-intelex package instead of daal4py for the scikit-learn acceleration.
Introduction
Gradient boosting on decision trees is one of the most accurate and efficient machine learning algorithms for classification and regression. The most popular implementations of it are:
XGBoost*
LightGBM*
CatBoost*
daal4py Model Builders deliver the accelerated models inference of those frameworks. The inference is performed by the oneDAL GBT implementation tuned for the best performance on the Intel(R) Architecture.
Conversion
The first step is to convert already trained model. The API usage for different frameworks is the same:
XGBoost:
import daal4py as d4p
d4p_model = d4p.mb.convert_model(xgb_model)
LightGBM:
import daal4py as d4p
d4p_model = d4p.mb.convert_model(lgb_model)
CatBoost:
import daal4py as d4p
d4p_model = d4p.mb.convert_model(cb_model)
Note
Convert model only once and then use it for the inference.
Classification and Regression Inference
The API is the same for classification and regression inference.
Based on the original model passed to the convert_model
, d4p_prediction
is either the classification or regression output.
d4p_prediction = d4p_model.predict(test_data)
Here, the predict()
method of d4p_model
is being used to make predictions on the test_data
dataset.
The d4p_prediction
variable stores the predictions made by the predict()
method.
Scikit-learn-style Estimators
You can also use the scikit-learn-style classes GBTDAALClassifier
and GBTDAALRegressor
to convert and infer your models. For example:
from daal4py.sklearn.ensemble import GBTDAALRegressor
reg = xgb.XGBRegressor()
reg.fit(X, y)
d4p_predt = GBTDAALRegressor.convert_model(reg).predict(X)
Examples
Model Builders models conversion