eickenberg

asv actually doesn't run for me. It's can't create the environment it wants to create. Because it can't find pytorch. Do you not get this error? It might be a...

Yep, so `conda-channels` was commented. I uncommented, which gives access to conda-forge, and it still fails. Let me see if I can add the pytorch channel there

Change the environment engine from `conda` to `virtualenv` and now it looks like it's running

```The command "if [[ "$TEST_MODE" == "PEP8" ]]; then PYTHONPATH=$PWD:$PYTHONPATH py.test --pep8 -m pep8 -n0; else PYTHONPATH=$PWD:$PYTHONPATH py.test tests/; fi" exited with 2.``` Seems this line has different exit codes...

huh that is curious ... the 2d images are produced by the 2d example here: https://www.kymat.io/gallery_2d/plot_filters.html#sphx-glr-gallery-2d-plot-filters-py Indeed, there may be some confusion at the sphinx level where some images end...

these should all be aligned and the backend should regularize everything to numpy convention (which for stack means axis=0 by default) On Tue, Oct 31, 2023 at 4:12 AM Cyrus...

yeah I'm describing the desired state ;) I'd like to go much further and expose basic numpy functions on all backends with the same function signatures. this would unblock a...

Looks like plans are destroyed always https://github.com/locuslab/pytorch_fft/blob/2ca24e3a1300bf0bfdaeeae29b082138a043a4b3/pytorch_fft/src/generic/th_fft_cuda.c#L84 Any plans to make it possible to keep them?

It would have to be persisted in a pytorch module or something (like e.g. nn.Conv2d stores its own weights) and initialized when the object is created, with a specific shape....