4 ms·
Is anyone doing machine learning on AMD hardware right now? What's the preferred pipeline to go from Tensorflow to GPU?
by ipsum2 8y ago
Is anyone doing machine learning on AMD hardware right now? What's the preferred pipeline to go from Tensorflow to GPU?
- bitL 8y agoOnly adventurous people (i.e. framework developers). Everybody else runs on NVidia and TPU.
- currymj 8y agoAMD is scrambling to build a CUDA-esque thing called ROCm and get various frameworks including TF ported over, but it’s not even in “early adopter” state yet, so i doubt anyone is using it. However they seem to be making progress.
- singhrac 8y agoIt's not that far away. PyTorch merged a build script semi-recently, though I don't know if it's fully tested or supported. https://github.com/pytorch/pytorch/blob/master/tools/amd_build/build_pytorch_amd.py https://github.com/pytorch/pytorch/blob/master/tools/amd_bui...
- nl 8y agoThat “disabled_feautures.json” file indicates it will be a while before it is useful though.
- rdlecler1 8y agoCould it really be that hard? They’ve had at least a few years heads up and this feels like something that could be accomplished relatively quickly with 5-10 10x engineers. As a high profile project that shouldn’t have difficulty finding the talent.
- maeln 8y agoAs I remember, AMD was supporting OpenCL (and now SPIR-V Computer Shader) fairly well. Are those technologies not well regarded in ML ?
- tanilama 8y agoNot really. No small task really. ML models are not even resistant enough to same framework upgrade, let alone hardware switch.