Guillaume Lemaitre

FEA add PredictionErrorDisplay

14

Add a `PredictionErrorDisplay` which is useful when dealing with regression. Edit: Partially addresses #16608.

Add a new display to plot the output of `learning_curve` with sensible default.

Opening an issue for a bug reported in scikit-learn by @AnotherSamWilson. xref: https://github.com/scikit-learn/scikit-learn/issues/20867 Unfortunately, I was not able to reproduce it on a Linux machine. ### Describe the bug When...

Sampler should also sample sample_weight and return it

5

Some scikit-learn estimators rely on the sample_weight but the current Sampler do not accept it. We should be able to at least sample the sample_weight as well. However, it should...

API duck-typing for n_neighbors in CNN and deprecate estimator_

1

This PR does: - deprecate `estimator_` in favour of `n_neighbors_` - accept any arbitrary knn classifier that has a `n_neighbors` parameter

ENH make Random*Sampler accept dask array and dataframe

13

POC to see if we can make at least the `RandomUnderSampler` and `RandomOverSampler` accept some dask array and dataframe Note: - https://github.com/dask/dask-ml/issues/317 - #701

[WIP] ENH: Resample additional arrays apart from X and y

7

Implement the last point of #462 and should be merged after it. Partially addressing #460

[ENH] Add sample_indices_ for SMOTE/ADASYN classes

5

SMOTE/ADASYN classes currently do not provide a `sample_indices_` attribute since they are generating samples that do not belong to the original dataset. However, we could create a new semantic for...

New methods

48

This is a non exhaustive list of the methods that can be added for the next release. Oversampling: - [ ] SPIDER - [ ] MWMOTE - [ ] SMOTE-SL...

Create a meta-estimator using BalanceCascade sampler

4

As the BalancedBaggingEnsemble, it could be possible to create a meta-estimator using the BalanceEnsemble sampler.