Re:Benchmarking Google’s new TPUv2
by boulos on Friday February 23, 2018 @02:45PM
(#16448792)
Disclosure: I work on Google Cloud.
Oh! You should definitely say that. It's semi-reasonable then to choose the batch size that is optimal for the part. It'd be good to make sure this isn't why your LSTM didn't converge though...
Oh! You should definitely say that. It's semi-reasonable then to choose the batch size that is optimal for the part. It'd be good to make sure this isn't why your LSTM didn't converge though...