Denis Vieriu
Denis Vieriu
> 1.12.0 for macos (torch-nightly) @cauchy7203 thanks for the report. This issue should be already fixed in the latest pytorch nightly (**1.13.0**.dev20220924) - could you please give a try to...
Hi @FahimF - thanks for filling this issue. The fix for this issue just landed in PyTorch master (https://github.com/pytorch/pytorch/pull/94963) and next nightly build should have it. Please let me know...
> I check out this pr and run > > ``` > git submodule sync > git submodule update --init > ./backends/apple/mps/install_requirements.sh > python -m examples.models.llama2.export_llama -kv --mps > ```...
Hi @QuanyiLi - a fix was just merged in couple minutes ago. Next nightly version should pick up this fix and issue should be gone. Please let me know if...
Thanks @QuanyiLi for trying it out. Could you please share the nightly you are using? I've just downloaded latest nightly and this one contains the change: ``` $ pip list...
Thank you for the report @junukwon7 and for the repro code @pcuenca, @Birch-san. This is an issue on the MPS side - we are working on a fix.
Hey @devpacdd - this should be fixed in latest pytorch nightly (pip3 install --pre --force-reinstall torch --index-url https://download.pytorch.org/whl/nightly/cpu). Let me know if you still see any issues. Thanks
> The operator 'aten::repeat_interleave.self_int' is not currently supported on the MPS backend and will fall back to run on the CPU. This may have performance implications. (Triggered internally at /Users/runner/work/pytorch/pytorch/pytorch/aten/src/ATen/mps/MPSFallback.mm:11.)...
> Would like to add `aten::_local_scalar_dense` to the list. Also, is it possible to link to some examples in the top post on how we can implement these into Pytorch?...
> I also have the same issue, attempting to run OpenAI's Whisper with MPS support > > > /AppleInternal/Library/BuildRoots/c651a45f-806e-11ed-a221-7ef33c48bc85/Library/Caches/com.apple.xbs/Sources/MetalPerformanceShaders/MPSCore/Types/MPSNDArray.mm:794: failed assertion `[MPSNDArray, initWithBuffer:descriptor:] Error: buffer is not large enough. Must...