Carlos Ramos Carreño

It appears that the topic of constant predictions has been already discussed in scikit-learn: https://github.com/scikit-learn/scikit-learn/issues/16857 We should probably try to adapt the same solution as them, either in this PR...

Some of the methods of `BasisSmoother`, such as the method to compute the hat matrix, make no sense in the sparse context. For now we can ignore it, but we...

Same problem here.

I think this problem is because Numba uses very long names for its cache files, but Windows has a small length limit for the whole path. When the cache files...

As you requested me further help, I will post this here in case it helps others too: Description of the error: - Numba is a (transitive) dependency of scikit-fda, used...

I just negated `reverse` in the call to `get_sorted_rank_groups` and it worked for my case... not sure if in general.

I am really interested in this. It is also useful to remove the silencing of false positives when they are fixed.

Hum, it worked in my PC. I can't see the errors in Azure (maybe because the runs have been removed). Can you relaunch the tests?

It seems that it is working only in Python 3.8 and newer because of this: https://docs.python.org/3/whatsnew/3.8.html#tokenize . I could attempt to manually add a newline to have it working in...

> @vnmabus Agreed, if it makes your code significantly simpler, we can remove support for Python 3.7 I don't think it would be too hard to fix it either. But...