search

mathurinm / andersoncd

This code is no longer maintained. The codebase has been moved to https://github.com/scikit-learn-contrib/skglm. This repository only serves to reproduce the results of the AISTATS 2021 paper "Anderson acceleration of coordinate descent" by Quentin Bertrand and Mathurin Massias.
BSD 3-Clause "New" or "Revised" License
18 stars 6 forks source link

Missing transpose #24

Closed idc9 closed 3 years ago

idc9 commented 3 years ago

There is a transpose missing in https://github.com/mathurinm/andersoncd/blob/af2123b241e5f82f7c51b2bbf5196fb02723b582/andersoncd/weighted_lasso.py#L227

i.e. this line should read as

alpha_max = np.max(np.abs(X[:, penalized].T @ y / weights[penalized])) / n_samples
mathurinm commented 3 years ago

Thanks Iain, you're right ! This part of the code probably isn't covered by tests.

Can you send a quick PR to fix the transpose ?