celer
celer copied to clipboard
mathurinm
Fast solver for L1-type problems: Lasso, sparse Logisitic regression, Group Lasso, weighted Lasso, Multitask Lasso, etc.
There seems to be a problem with the `climate.target_region` function which incorrectly extracts one column to the right of the intended one. https://github.com/mathurinm/celer/blob/main/celer/datasets/climate.py#L60-L72 The `pos_Lx` value range is supposed to...
It is not used in the Lasso case, and for the logreg case we can avoid this design with the declaration of inv_lc_j inside the various if/else of the cd...
Thank you again for this remarkably fast library. In the celer_path the default value of prune is 0 while in Lasso, MultiTaskLasso, and GroupLasso it's set to 1 (True).
I was wondering whether you could (easily) implement the MultiTask GroupLasso as in, or similar to, equation 1 here: http://psb.stanford.edu/psb-online/proceedings/psb22/nouira.pdf Thanks !
Example(s) should exhibit the advantages of the ElasticNet estimator over the Lasso and OLS estimators, namely - [x] Feature selection (case p >> n) - [ ] Grouping effects -...
https://github.com/mathurinm/celer/blob/4230db117a916e5158cfae85b2cd1a7249cf5475/celer/group_fast.pyx#L237-L238 https://github.com/mathurinm/celer/blob/4230db117a916e5158cfae85b2cd1a7249cf5475/celer/multitask_fast.pyx#L332-L333
This is the classic Fenchel-Rockafellar primal dual link. It will be easier to check the maths, notably in the enet case.
I think this is a dangerous design as it silences eventual errors/bugs. I suggest raising a warning in this case. https://github.com/mathurinm/celer/blob/83c46284b712c42d499044ab4ef29e4fe2a0df63/celer/PN_logreg.pyx#L426-L428