DALEX

DALEX copied to clipboard

Add aSHAP - aggregated SHAP values for a set of observations

2

Issue #519

adrianstando

[fastai] 'DataFrame' object has no attribute 'to_frame' with fastai

12

I'm trying to wrap a `fastai` tabular `learner` with DALEX. I got `'DataFrame' object has no attribute 'to_frame'` error with `dx.Explainer(learn, xs, y, label = "Deep NN")`. Any potential problems...

ming-cui

fix stopping Arena with `flask`

1

closes #503 hopefully fixes our CI and conda CI

hbaniecki

python model_parts: Is there show_boxplots parameter?

1

In R package show_boxplots = TRUE would plot the permutations in feature importance In python model_parts, this parameter does not seem to exist. Or it is called something else?

asheetal

aSHAP values can be used to explain a set of observations, then they may by visualized with a waterfall plot.

adrianstando

run python tests and see if they broke due to changes in `flask` v2.1

1

possibly the cause of broken CI in https://github.com/conda-forge/dalex-feedstock/pull/6 set a temporary requirement for the dependencies: `werkzeug` < 2.1

hbaniecki

ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 96 from C header, got 88 from PyObject

3

Some random `numpy` error breaks our Python 3.10 CI https://stackoverflow.com/questions/71448239/valueerror-numpy-ndarray-size-changed-may-indicate-binary-incompatibility-exp

hbaniecki

...and possibly remove the conversion from pandas to numpy in ShapWrapper. Then, `explainer.model_info['arrays_accepted']` would become redundant, which in fact fixes https://github.com/ModelOriented/DALEX/issues/507 by making the scikit Pipeline fail gracefully in the...

hbaniecki

* codes are added to model_performance function. * proper documentation is added to DALEX reference. * NEWS.md is updated.

mcavs

update code for newer versions of `pandas`

hbaniecki