Aren't the Tensor Cores mostly for inference, not training?
Re:A new GPU backend for the TVM stack
by dragontamer on Monday October 30, 2017 @04:17PM
(#15588500)
The Volta Tensor Cores aren't released yet, and I haven't played with Google's TPUs at all.
But NVidia markets Tensor cores as:
> New Tensor Cores designed specifically for deep learning deliver up to 12x higher peak TFLOP/ss for training, and 6x higher peak TFLOP/s for inference
https://devblogs.nvidia.com/parallelforall/cuda-9-features-r...
I wouldn't be surprised if they were inflating the numbers slightly, as is common in a lot of marketing material.
You can use Tensor Cores on AWS p3 instances (v100) now. Just doesn't seem to be a whole lot of ML framework support for mixed precision training with CUDA 9 yet
https://devblogs.nvidia.com/parallelforall/programming-tenso...
Google's TPUs also do low precision training (with some special version of TF). https://cloud.google.com/tpu/
Vega 64 can in theory do 25 TFLOPs half precision.
But as you say there's a large price difference too.
For a market segment that needs 1-8 GPU rigs for ML on a low budget AMD could kill it if they invested in software support and kernel optimisation.
For servers and large scale training, unless AMD has some ML specialised cores in the pipeline, Nvidia Volta and Google TPUs have a serious lead.