to_array to create a dimension as last axis

Is your feature request related to a problem?

I do ds.to_array(dim="variable").transpose("latitude", "longitude", "variable"). I would like to avoid the extra transpose call.

Describe the solution you’d like

ds.to_array(dim="variable", new_axis="last") where new_axis is Literal["first", "last"] = "first" or ds.to_array(dim="variable", new_axis=0) where new_axis is Literal[0, -1] = 0

code to change: https://github.com/pydata/xarray/blob/main/xarray/core/dataset.py#L5770

Describe alternatives you’ve considered

No response

Additional context

I imagine new_axis could be of type int to place the new axis where you would like but the proposal above may be a good first step.

For reference, i’m doing deep learning and want to shape the data as width (latitude), height (longitude), channel (feature)

[max-sixty]

What’s the advantage of this over the transpose call?

The transpose approach is more modular, since it takes advantage of the orthogonality of the methods — i.e. we have one method for all usages, rather than kwargs in lots of methods.

We’ve also managed to have axis as a hidden implentation detail, and this would start exposing it.

[max-sixty]

I would say very little. I’m a lazy programmer.

Haha, us both Ray!

I do think that part of laziness is not having to remember too much — so hopefully I can trade you the costs of an extra method call for more consistency!