Re:Cloud TPUs in Beta
by vomjom on Monday February 12, 2018 @10:00AM
(#16358966)
Keep in mind that what you linked refers to TPUv1, which is built for quantized 8-bit inference. The TPUv2, which was announced in this blog post, is for general purpose training and uses 32-bit weights, activations, and gradients.
It will have very different performance characteristics.
It will have very different performance characteristics.