Support for `XGBRegressor.missing` and `XGBClassifier.missing` attributes

jpmml / jpmml-sklearn

Java library and command-line application for converting Scikit-Learn pipelines to PMML

GNU Affero General Public License v3.0

531 stars 117 forks source link

Support for `XGBRegressor.missing` and `XGBClassifier.missing` attributes #167

Closed vruusmann closed 2 years ago

vruusmann commented 2 years ago

See https://github.com/jpmml/jpmml-sklearn/issues/166#issuecomment-928173571

The pipeline has currently no idea that the XGBoost step interprets data differently from other steps. For example, in the linked issue, XGBoost thinks that 1 values represent missing values, whereas all other steps take 1 values valid values.

vruusmann commented 2 years ago

Was fixed in https://github.com/jpmml/jpmml-sklearn/commit/64a6e5a47a83cfec19152012ea38c150fd814123