leleamol commented on issue #15252: FP16 training with 2bit gradient compression with train_imagenet.py URL: https://github.com/apache/incubator-mxnet/issues/15252#issuecomment-503744461 @Luo-Liang Thanks for raising this issue. I was able to reproduce it. However, the script worked for float32 datatype. I am adding relevant tags so that experts can comment on those. @rahul003 I noticed that the function Quantize2BitImpl() was added by you. Would you be able to comment? @mxnet-label-bot add [FP16]
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
