Hacker News Reader

Question to the community: At some point, I may want to work towards implementing a small tensor-core (or a vertex/pixel shader as in "GPU implementation of neural networks," Oh&Jung, 2004) for a toy ml project. Do you have some materials to recommend that would teach me how I should/could proceed?

Thanks a lot.

View on HN

One good approach could be to base the architecture on the TPU v1 from [1]. There are also open-source accelerators you could get inspiration from, for example [2][3]. If you want to do less work/not hand code the RTL yourself then you could look into methods for automatically mapping OpenCL to an FPGA accelerator architecture (or a service like [3] provides pre-designed architectures for multiple FPGAs).

[1] https://arxiv.org/abs/1704.04760

[2] https://github.com/jofrfu/tinyTPU

[3] https://github.com/tensil-ai/tensil

Thanks!

Tensor cores are simply multiplying matrices. Relatively easy to do with a compute shader.

Here's an example of such shader in D3D11: https://github.com/Const-me/Whisper/blob/master/ComputeShade...

Reply on news.ycombinator.com