Skip to content
New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

[r] ebm_show local version? #222

Open
brianfhead opened this issue Apr 6, 2021 · 1 comment
Open

[r] ebm_show local version? #222

brianfhead opened this issue Apr 6, 2021 · 1 comment
Labels
enhancement New feature or request good first issue Good for newcomers

Comments

@brianfhead
Copy link

Hi,

I see in the 'code' section that the python version has both a global and local explain. However, in the R 'interpret' package I'm only seeing "ebm_show" which appears to be a global explanation. I've tried searching around, but haven't had any luck.

Is there currently a way to get a local level explain? Will this feature be added to the R version soon?

I've tested EBM against existing ML algos we used on a recent project (h2o automl stacked ensemble) and it performed comparably. I'd really love to be able to use it as a big issue for the project is interpretability/explainability.

Look forward to your reply,
Brian

@interpret-ml
Copy link
Collaborator

Hi @brianfhead -- We're happy to hear that EBMs are performing comparably on your datasets. You are correct that the current R implementation only supports global explanations. The R package has a subset of the functionality available in python in several areas, like this one. We do plan to improve this and eventually match the python package in terms of EBM functionality, but we don't have a timeline for that work yet. Improving EBMs is still an active area of research and we've been spending more of our time there recently, so we would welcome community contributions like local explanation graphs for R in the meantime.

-InterpretML team

@paulbkoch paulbkoch added enhancement New feature or request good first issue Good for newcomers labels Jan 29, 2023
@paulbkoch paulbkoch mentioned this issue Jan 29, 2023
# for free to join this conversation on GitHub. Already have an account? # to comment
Labels
enhancement New feature or request good first issue Good for newcomers
Development

No branches or pull requests

3 participants