interaction_constraints

  • Available in: XGBoost

  • Hyperparameter: no

Description

Specify the feature column interactions which are allowed to interact during tree building. Use column names to define which features can interact together. This option defaults to None/Null, which means that all column features are included.

Note: This option can only be used when the categorical encoding is set to AUTO (one_hot_internal or OneHotInternal).

Example

library(h2o)
h2o.init()

# import the prostate dataset:
prostate = h2o.importFile("http://s3.amazonaws.com/h2o-public-test-data/smalldata/prostate/prostate.csv.zip")

# convert the CAPSULE column to a factor
prostate$CAPSULE <- as.factor(prostate$CAPSULE)
response <- "CAPSULE"

# train a model using the interaction_constraints option
prostate_gbm <- h2o.gbm(y = response,
                        interaction_constraints = list(list("AGE", "DCAPS")),
                        seed = 1234,
                        training_frame = prostate)
import h2o
from h2o.estimators.gbm import H2OGradientBoostingEstimator
h2o.init()

# import the prostate dataset:
prostate = h2o.import_file("http://s3.amazonaws.com/h2o-public-test-data/smalldata/prostate/prostate.csv.zip")

# convert the CAPSULE column to a factor
prostate["CAPSULE"] = prostate["CAPSULE"].asfactor()
response = "CAPSULE"
seed = 1234

# train a model using the interaction_constraints option
gbm_model = H2OGradientBoostingEstimator(seed=seed, interaction_constraints=[["AGE", "DCAPS"]])
gbm_model.train(y=response, ignored_columns=["ID"], training_frame=prostate)