modeloriented / dalex-docs Goto Github PK

1. add an example of mitigation to dalex-fairness.ipynb
2. add a missing dalex-fairness.ipynb file as a source to python-dalex-fairness2.html

mentions ModelOriented/DALEX#432

and i honestly don't know how to fix it

update dalex-fairness vignette

Hello,

I'm trying the arena service for my model and data,

I saw all the possibilities that the web page gives, it is really useful!

But I found a problem when trying to check the shap - breakdown - ceteris values on a custom observation. I'm changing it in the Observation Details tab, where an slider appears to change the values of the variables.

The value print on the terminal is:

ValueError: Input X contains NaN.
RandomForestClassifier does not accept missing values encoded as NaN natively. For supervised learning, you might want to consider sklearn.ensemble.HistGradientBoostingClassifier and Regressor which accept missing values encoded as NaNs natively. Alternatively, it is possible to preprocess the data, for instance by using an imputer transformer in a pipeline or drop samples with missing values. See https://scikit-learn.org/stable/modules/impute.html You can find a list of all estimators that handle NaN values at the following page: https://scikit-learn.org/stable/modules/impute.html#estimators-that-handle-nan-values

Any idea?

Thanks!

PD: I'm using a random forest model and the titanic dataset with a few modifications.

In fairness notebook numerical columns were omitted by the model. It should be fixed.

https://medium.com/responsibleml/basic-xai-with-dalex-part-6-lime-method-f6aab0af058a
https://medium.com/responsibleml/basic-xai-with-dalex-part-7-ceteris-paribus-profiles-e84a36edd750
to both: dalex.drwhy and python.drwhy

Both of these could use some help <3

v1.4.0

DALEX:

• Responsible Machine Learning ?
• XAI for R & Python ?
• ...

two .md files

update https://github.com/ModelOriented/DALEX-docs/blob/master/dalex/dalex-class.png and https://github.com/ModelOriented/DALEX-docs/blob/master/dalex/dalex-class.xml

and maybe change layout for two rows instead of one

produce new images where appropriate https://python.drwhy.ai

please

..and add vignettes: teaching2, BreakDown/LIME/Shapley comparison

https://medium.com/responsibleml/basic-xai-with-dalex-part-7-ceteris-paribus-profiles-e84a36edd750