jxtps

If we want the upsampling to work like in the tensorflow version, then we'd have to swap the order of the `conv2d` and `upsampling` steps in `SegmentationHead`: ```python class SegmentationHead(nn.Sequential):...

There's also the "final stage" `Conv3x3BnReLU` between the Decoder and the Head in the tensorflow version.

Personal pet peeve: MarkJeronimus' implementation selects `BufferedImage.TYPE_3/4BYTE_BGR` over `BufferedImage.TYPE_INT_A/RGB`, whereas I much prefer the latter. I know it's possible to force the format used when reading (as you so graciously...

The snapshot with those additional settings gets stuck on: ``` Debug: Loading library libopenblas ``` Running `MyTest.main()` standalone does not get stuck (but errors our since I don't have the...

Ok, this is interesting, while digging up the stack traces it all of a sudden started working! I had to revert to sbt 1.3.13 due to [issues with incremental compilation...

Well, when I do that it does try mkl_rt first, but since I don't actually have those libs it falls back on openblas, and then freezes... What I find really...

Ok, so there's a smattering of resources out there referencing various builds for various ARM architectures: 1. https://discuss.pytorch.org/t/crosscompilation-for-arm64/9885/9 2. Leads to: https://gist.github.com/soumith/45f61db3379db67414814ad9d5f2a427 - a gist, "Install PyTorch on NVIDIA AGX...

It looks like PyTorch is publishing ARM wheels and have been for some time: https://github.com/pytorch/pytorch/issues/53357 If you search for "aarch64" on https://download.pytorch.org/whl/torch_stable.html it looks like recent versions are all there....

Does that include GPU support, e.g. for use with the AWS G5g instance type?

Nevermind, now that I look closer at https://download.pytorch.org/whl/torch_stable.html they don't have anything cross-compiled for cuda (cu113, cu100 etc), it's all just for cpu. I take it that's not helpful here?