Passing n-dimensional array to explain_function

[Bar-Kal]

According the docs of the explain_function, it is only possible to pass a 2-dimensional array with shape (1, n_features). Is there a way to pass an array with the shape (n, n_features)? If not, would you just run the explain function in a loop to achieve this?
For instance the shap package has this functionality where passing a matrix or a data frame is possible.
Many thanks!

[mmschlk]

@Bar-Kal, hi good point. Yes for this we have the Explainer.explain_X function. This basically is a wrapper around the explain function with some additional suppport for parallelization via joblib. We opted to split this from the main explain function such that users are not surprised when they get a list back as a return type and not a single InteractionValues as from the basic explain function.

Does this help you?

[Bar-Kal]

Many thanks for the hint! Works as expected.

[mmschlk]

Yes! Perfect! :) The docs are not so well maintained again. I really need to get on this! :O