Yngve Mardal Moe

Oh, to easily deal with the intercept, I simply padded the data matrix with a column of ones. As for the `losses_` attribute, that is computed during training if the...

LOG_LOSSES specifies that the loss per iteration should be stored in a list, which is very useful for debugging. For LogisticGroupLasso, I always use the overparametrised softmax formulation, which should...

That would be a very nice addition. However, while non-negativity constraints are very straightforward for LASSO, it is not straightforward to implement non negative group lasso. It seems like we...

There is currently no way of specifying that. However, it is on my wishlist for the future. The reason this is not supported yet is that it is not immediately...

The main reason for why it will be difficult to implement the latent group lasso is that there is no longer an easy closed-form solution to the Lipschitz bound of...

The package is now using a line search, so this is possible to implement, but I don't currently have the time for that.

I reinstalled poetry and now everything works. I'm not sure why. **Edit:** I tried again and the issue occured again, so I am unsure if it actually worked before or...

@JeanKossaifi, I tried looking at this, and I do not understand what the problem with the orthogonalisation is as it stands. Is orthogonalisation used as a trick to improve the...

I did some testing with the example code above, and the call time went from ~80 ns to ~500 ns. So a big relative change, but still a very small...

Hmmm. I think we could do something like that :) But I think it would be nice to discuss it in a separate issue/PR to make it easier to see...