datatable.models.Ftrl.feature_importances¶
feature_importances
Feature importances as calculated during the model training and
normalized to [0; 1]. The normalization is done by dividing
the accumulated feature importances over the maximum value.
Parameters¶
return
FrameA frame with two columns: feature_name that has stype str32,
and feature_importance that has stype float32 or float64
depending on whether the .double_precision
option is False or True.
The content on this page is licensed under the Creative Commons Attribution 4.0 License
(CC BY 4.0) ,
and code samples are licensed under the MIT License.