Implement vmap

[JuanPedroGHM]

Related Provide a clean way to apply NumPy or torch functions to the local arrays, without having to manually get the local array, transform it to torch/NumPy, apply the function and then create the DNDArray again.

Feature functionality The user should be able to apply a function from torch or NumPy in a single command, as long as the result is independent of the global array and the result shape of the local array does not change.

[mrfh92]

Is this intended to work similar as, e.g., torch.vmap in the sense that you can do something like:

fun = ht.heatmap(torchfun) 
y = fun(x)

with a function torchfun defined on torch.Tensors?

Or do you even want to go further and include torch.vmap somehow if torchfun can also be parallelized on each process?

[github-actions[bot]]

This issue is stale because it has been open for 60 days with no activity.

[github-actions[bot]]

Branch features/1383-Implement_vmap created!