skada

skada copied to clipboard

During the validation step of the training loop, we should think about the strategy for the dataloading. Should we use the same as the training set? Should we provide a...

tgnassou

Merge all OT mapping methods in one class

2

Should we merge OT mapping methods? it seems like the use of entropic regularization or group lasso could be added easily as a parameter to the same class. We should...

rflamary

[ON_HOLD] Additional utility helper to check structure of the `sample_domain` array

1

The checker makes sure that each domain label is used either as a source or as a target, except for the 'lodo' packing (for which we check that subsets are...

kachayev

https://projet.liris.cnrs.fr/imagine/pub/proceedings/CVPR2012/data/papers/261_O2D-03.pdf

rflamary

Multidomain source target split and merge

2

Add the ability to have: - `source_target_merge` from utils.py to accept `*List[arrays]` with each list corresponding to one sample domain. Ex: List[X_domain_1, y_domain_1], List[X_domain_2, y_domain_2] ... - `source_target_split` from utils.py...

YanisLalou

[WIP] Tabular dataset + Doc for testing the advantage of using DA techniques

1

Issue #19 Dataset implemented: https://tableshift.org/datasets.html#childhood-lead Still need to be done: - For the plot_domain_adaptation_advantages.py: - Plots ? - Understand why score gap between with and without DA technique is so...

YanisLalou

[API] fit, predict and other functions should accept X,y,smaple_weight as first parameters by default

2

In order to stay compatible with `skleran`API we want all estimatoer (and Adapters) to have functions that accept unnamed `sample_weight`by defalult. For the moment we have for Adapter and DAEstimator:...

rflamary

This functionality was planned... but was not implemented from what I can tell. It seems like a good safety guard. It is also important for those algorithms that rely on...

kachayev

Last doc build: https://output.circle-artifacts.com/output/job/d77ed2a4-3c2a-48e2-838d-75bc08bb00a7/artifacts/0/dev/auto_examples/validation/plot_cross_val_score_for_da.html#sphx-glr-auto-examples-validation-plot-cross-val-score-for-da-py In this example we compare 2 estimators for a binary classification task. The first estimator uses DA methods and the second one doesn't. The 1st estimator...

YanisLalou

Currently, every DAdataset packing function is only compatible with `List[str]` for the `as_sources` and `as_targets` arguments. However, in cases where we have only one target and one source, it would...

YanisLalou