Guillaume Lemaitre

MAINT introduce black to loky

9

Introducing black with 79 characters.

In scikit-learn, we have the following use-case. We want a class method (e.g. `predict_proba`) to only be defined under some condition meaning that `hasattr(instance, "class_method")` should return `False` or `True`...

This PR isolate the call to score in the hyperparameter notebook. It is linked to the comment: https://github.com/INRIA/scikit-learn-mooc/pull/464#discussion_r768893865 In this PR, we should therefore address the concern of @ogrisel: >...

I wrote an example in scikit-learn which I think would be useful in the inspection part of the course: https://github.com/scikit-learn/scikit-learn/pull/18821

Add figure of the effect of StandardScaler in the M2 wrap quiz correction

2

@ogrisel proposes to add an additional figure: https://gitlab.inria.fr/learninglab/mooc-scikit-learn/mooc-scikit-learn-coordination/-/merge_requests/8#note_477599 It is probably not dramatic for the next course but we should address it. I am putting this issue as a reminder.

We should add a small section to show how: - how to handle missing data - illustrate the usage of an imputer in a pipeline - illustrate the usage of...

Better error message when launching ramp-test outside from the ramp-kit

1

`ramp-test` will expect to find the file `problem.py` and will fail when it is run and it will not fail gracefully in other case. It could be nicer to have...

Reverting some changes from my previous PR. We should use the built-in functionalities from pandas in this case.

FIX use balanced accuracy from scikit-learn

8

closes #127

[RFC] Workflows refactoring

1

The workflows should be derived from a metaclass to enforce having the parameter `element_names`. I would like also to investigate: * If workflows should be derived from the scikit-learn Pipeline...