data-scientists

Recently dtreeviz has added support for lightgbm models: https://github.com/parrt/dtreeviz/

So it would be cool to add the same tree visualization in explainerdashboard that already exist for RandomForest, ExtraTrees and XGBoost models.

If someone wants to pitch in to help extract individual predictions of decision trees inside lightgbm boosters and then get them in shape to be used by the

Remove pinned transfomer version from generate_conda_files
https://github.com/interpretml/interpret-text/blob/97416a0a9cc3e60bcb1221f878577762c64df02e/tools/generate_conda_files.py#L65

Since fuseml/fuseml#171 the mlflow-e2e job is taking too long to complete.

This behaviour can only be reproduced when running the workflow on github actions hosts as the CI does and it seems to be related to the change in the kaniko task to push using --insecure.