News for hackers. Stuff that matters.
Re:PyTorch on Apple M1 MAX GPUs with SHARK – faster than Ten... by smoldesu on Tuesday February 22, 2022 @06:32PM (#30435371)
The M1 is (supposedly) specialized for ML calculations and on a decidedly denser node, though. I guess it's a testament to either CUDA's power or machine learning GPGPU optimization.
Re:PyTorch on Apple M1 MAX GPUs with SHARK – faster than Ten... by simondotau on Tuesday February 22, 2022 @07:00PM (#30435633)
> The M1 is (supposedly) specialized for ML calculations

It is, but they’re not using any of that. Only the M1 GPU at this time.

Re:PyTorch on Apple M1 MAX GPUs with SHARK – faster than Ten... by magic_at_nodai on Tuesday February 22, 2022 @07:07PM (#30435704)
Yeah the ANE and AMX on cpu are wrapped behind Accelerate Framework and CoreML. So you will have to use CoreML (which wasn't able to compile the latest TF BERT). ANE is also inference only. So if you want training you will have to use the GPU with Apple's Tensorflow-Metal.

//part of nod.ai / SHARK team.

Re:PyTorch on Apple M1 MAX GPUs with SHARK – faster than Ten... by yayr on Tuesday February 22, 2022 @06:40PM (#30435447)
I like this blog to understand NVidia perf stats (search for "Normalized GPU deep learning performance"):

https://timdettmers.com/2020/09/07/which-gpu-for-deep-learni...

It seems that their consumer GPUs also have a gap