paraphrase-id-tensorflow icon indicating copy to clipboard operation
paraphrase-id-tensorflow copied to clipboard

Published 20 hours ago • verified publisher icon nelson-liu

Add MultiGPU (Data Parallelism)

Open nelson-liu opened this issue 7 years ago • 2 comments

Looks like the train time is pretty long on AWS instances with K80s. Adding MultiGPU data parallelism would be a good way to mitigate this (as done in https://www.tensorflow.org/tutorials/using_gpu#using_multiple_gpus)

nelson-liu avatar May 23 '17 02:05 nelson-liu

Does current version support multiple GPU ? I find this training time is pretty long on my training dataset which has almost 600,000 sentence pairs.

syxu828 avatar May 31 '18 17:05 syxu828

Unfortunately not, I haven’t been very good about maintaining this code since I switched to using PyTorch shortly after this. sorry!

nelson-liu avatar May 31 '18 18:05 nelson-liu