Results 16 issues of Quentin Bertrand

Algorithmically implicit differentiation is composed of 2 steps: - 1 compute the solution of the optimization problem - 2 solve a linear system. This linear system can be large, and...

cla: yes

Is there a reason why the adaptive lasso: https://github.com/mathurinm/celer/pull/131/commits/582c74eaeda7547985fbcedcee7da0ff965ddc34 has not been included in the examples: https://github.com/mathurinm/celer/tree/master/examples and is thus not in the doc https://mathurinm.github.io/celer/auto_examples/index.html ?

## What does this PR do? Add a script to generate multiple images for the CIFAR-10 examples Fixes #73 ## Before submitting - [x] Did you make sure **title is...

The parameter `num_images` is currently unused in the cifar10 example https://github.com/atong01/conditional-flow-matching/blob/21cd0c888186f6e2b76deb393800361b8a850e9b/examples/cifar10/train_cifar10.py#L52 It would be great to have a `generate_cifar10.py` to generate a given number of images, as in the edm...

documentation
enhancement

## Description of the feature In multiple issues, it seems that practitioners had problems selecting the regularization parameter `alpha`, and a lot of them do not know the value of...

enhancement
needs discussion

Currently `penalty.value` on takes `w` in argument https://github.com/scikit-learn-contrib/skglm/blob/99d0a0636b678876e26e93f28bf1d87b5c0d0b97/skglm/penalties/separable.py#L77 Computation could be made faster by using `penalty.value(w, ws)` (one cannot simply call `penalty.value(w[ws])` because of the weighted l1 norm

+ harmonizing first-order methods in examples

Currently the name of the classes encapsulating the estimators is the same as the name of the estimators: for example Lasso is the name of the estimator from sklearn/ celer...

We have been thinking about API question for a while. The current API is a little bit heavy, we have been scratching our head a lot to make it better,...