Thoughts/discussion: Interop with glow compiler

[saulshanabrook]

Glow compiler makes matrix math fast, by caring about cache locality: https://gist.github.com/nadavrot/5b35d44e8ba3dd718e595e40184d03f0

We could use compiled glow code as gumath kernels. they can compile aot: https://github.com/pytorch/glow/blob/master/docs/AOT.md (this would be a lot like our story with numba).

To create a glow network you either have to write C++ or compile from an onnyx model: https://github.com/pytorch/glow/blob/master/docs/Example.md https://github.com/pytorch/glow/blob/master/docs/IR.md#the-lifetime-of-a-glow-instruction

Could we have high level python APIs that compile to onnx spec? Like a lazy array/numpy library that builds up onnx graph as you interact with python objects? Then compiles that with glow and exposes gumath kernel for that operation?

If XND/gumath is the interop layer, then it could be used to combine tvm/glow/numba models. The underlying hypothesis is that the memory formats and computation could be expressed using xnd/gumath. I think the best way to answer this is to right code that attempts to interop and see where we stop.

[teoliphant]

This sounds like a promising line of investigation.

…

--- Travis Oliphant Quansight

On Tue, May 22, 2018 at 11:55 AM, Saul Shanabrook ***@***.***> wrote: Glow compiler makes matrix math fast, by caring about cache locality: https://gist.github.com/nadavrot/5b35d44e8ba3dd718e595e40184d03f0 We could use compiled glow code as gumath kernels. they can compile aot: https://github.com/pytorch/glow/blob/master/docs/AOT.md (this would be a lot like our story with numba). To create a glow network you either have to write C++ or compile from an onnyx model: https://github.com/pytorch/glow/blob/master/docs/Example.md https://github.com/pytorch/glow/blob/master/docs/IR.md# the-lifetime-of-a-glow-instruction Could we have high level python APIs that compile to onnx spec? Like a lazy array/numpy library that builds up onnx graph as you interact with python objects? Then compiles that with glow and exposes gumath kernel for that operation? If XND/gumath is the interop layer, then it could be used to combine tvm/glow/numba models. The underlying hypothesis is that the memory formats and computation could be expressed using xnd/gumath. I think the best way to answer this is to right code that attempts to interop and see where we stop. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#20>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAPjoPIazHUu186SJxvtIFqbHadB279pks5t1ELmgaJpZM4UJB7U> .