Marco Maggi

> The issue seems to be caused by `cross_product`: > > https://github.com/holoviz/panel/blob/4bf8756a336e6017e3ddb6b5542df813f869daae/panel/io/embed.py#L296 > > And the fact that `cross_product` contains all combinations of widget values, in order -- for the...

> Hmm, this seems to be in contradiction with #14877 and #14745? Sorry for the mistake and thank you for the references. Perhaps I should just add the fact that...

@lorentzenchr Could you please elaborate more on your need? When looking at the "filters" option https://mkdocstrings.github.io/python/usage/configuration/members/#with-filters-hello it seems to me that it is used to include/exclude Python objects rather than...

@lorentzenchr do you have any reference for implementing this? This feature sounds very useful and I would be happy to contribute

@lorentzenchr thanks for clarifying. To put it in pseudo code, `compute_score` should take as argument at least `score_per_obs` and at some point call ```python3 score_per_obs_de_meaned = score_per_obs - np.mean(score_per_obs) scipy.special.stdtr(len(score_per_obs)...

> I'm thinking of a new function `compute_score` as the `compute_bias`. It's just simple t-tests, see the code of compute_bias. the t-test in compute_bias is testing whether the bias per...

Thanks for your reply @paulbkoch. I did a fork to show you what I mean by smoothing only some variables. It's currently hardcoded: what's your opinion on having an argument...

ok, let me try first to understand the methodology with the two separate EMBs. I tried to modify the regression with synthetic data example as follows: ```python from interpret.glassbox import...

> ``` > from interpret.glassbox import ExplainableBoostingRegressor > from interpret.glassbox import merge_ebms > > exclude = [(i,) for i in range(X_train.shape[1]) if i > 3] > > ebm_smooth = ExplainableBoostingRegressor(...

ok, I don't know why I get the "decreasing" effect for feature 4 if I use the merged ebm, I even got it on two different machines and OS (Windows...