evalml icon indicating copy to clipboard operation
evalml copied to clipboard

Add support for Pipelines and Random Forests to visualize_decision_tree

Open ParthivNaresh opened this issue 4 years ago • 2 comments

Currently visualize_decision_tree supports only Decision Trees. We can expand the functionality to include support for Pipelines (so that the associated estimator can be taken and visualized) and for Random Forests.

Regarding Random Forests, we can choose to add the ability to view the entire forest, or to specify a tree in the forest, or both.

ParthivNaresh avatar Jan 12 '21 17:01 ParthivNaresh

Lightgbm also has a plot method https://lightgbm.readthedocs.io/en/latest/pythonapi/lightgbm.plot_tree.html

Perhaps xgboost does too

dsherry avatar Oct 05 '21 19:10 dsherry

@gsheni Let us know about this: https://gitlab.tue.nl/20040367/pybaobab

freddyaboulton avatar Dec 14 '21 15:12 freddyaboulton