Paul Koch

Hi @ditskihh -- The interaction parameter will accept specific interactions. You can set it like this: ``` ebm=ExplainableBoostingRegressor(interactions=[(1,2), (3, 7), (1, 11)]) ebm.fit(...) ``` We do not currently support any...

Hi @gmartinonQM -- Custom losses are not implemented currently. It's something that we've put a lot of work into, but the feature is not complete and we're currently working on...

I'll close this issue for now since we're tracking custom losses in issue https://github.com/interpretml/interpret/issues/281

HI @BullAJ -- The interaction strength API is available now. It's a simple utility function that you can call like this: ``` from interpret.utils import measure_interactions interaction_strengths = measure_interactions(X, y)...

Hi @elinrose -- We've done a lot of work on alternative losses, but they are not ready yet. We do plan to complete this sometime in the future, but we're...

Closing this issue since it is a duplicate of Issue #31. We'll post any updates for custom validation sets in that issue.

This might be related to issue https://github.com/interpretml/interpret/issues/322 even though the error reported is different. The error there suggests using a Pandas dataframe, so that could be one thing to try...

Thank you @prateekiiest for this fix!

Hi @progressEdd, @svjunkie and @Echochi -- We now have a conda-forge package (https://anaconda.org/conda-forge/interpret). The new installation instructions are: ``` conda install -c conda-forge interpret ```

This can now be accomplished with the bags parameter. Details in our docs: https://interpret.ml/docs/ebm.html#interpret.glassbox.ExplainableBoostingClassifier.fit