Does it train just as fast on gpus? | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		jiggunjer on July 17, 2020 \| parent \| context \| favorite \| on: TensorFlow, Keras and deep learning, without a PhD Does it train just as fast on gpus?

comicjk on July 17, 2020 | [–]

In my experience, not quite as fast for fully-tuned code, but the difference is small - and given the same project deadline, the PyTorch version will probably be faster.

m0zg on July 17, 2020 | [–]

Properly tuned (DistributedDataParallel + mixed precision) it will train faster, and consume a lot less RAM, allowing you to use larger batches, and higher LR.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact