Carlos Ramos Carreño

This is supposed to be solved in 0.5.5. If you can, please check that you have no problems with that version.

Currently it is not possible. It could be done, at least for the naive algorithm, propagating the distance function until the point where `pairwise_distances` is called, calling the distance function...

Before performing any work, I advise you to research if that is a sound procedure, that is, if the good properties of distance correlation are conserved if the distance is...

No, as today. The code would need to have a separate branch to handle that case, but it should be relatively easy to implement (adding a new function in `_hypothesis`...

Same problem here, and I agree with the second option.

Maybe you should try using the recently added `Tensor` basis (https://fda.readthedocs.io/en/latest/modules/autosummary/skfda.representation.basis.Tensor.html#skfda.representation.basis.Tensor), which represent a multivariable function in the basis of tensor products of univariate basis functions. Multivariable functions is currently...

May I also ask why are you doing this? ```python # get discretised functional data data = fetch_aemet()['data'].data_matrix fdgrid = FDataGrid(data) ``` AEMET data has `dim_domain=1` and `dim_codomain=3`, but as...

Ok, let me answer your questions. > The AEMET data was just an example to illustrate the problem. I'm just trying to get a fdbasis object from multivariable data with...

But the AEMET dataset has only one dimension in the domain (time) and three dimensions in the codomain (temperature, precipitation, wind). In this case you can use the `VectorValued` basis...

Sorry, but in your example I think that it is working as intended. Why do you think that the parameters are inconsistent? The domain dimension is 2, as intended, and...