stylable
stylable copied to clipboard
wix
Stylable - CSS for components
I wonder that how's the performance between byteps and butterfly? Can you offer more experiments of the performance? Thanks a lot.
After this PR, trainer.allreduce_grads will compute the average instead of the sum. Also, I think this line (https://github.com/bytedance/byteps/blob/master/byteps/mxnet/__init__.py#L322) is wrong, as it will ignore `self._scale = self._optimizer.rescale_grad` setting in the...
Can byteps support sync batch normalization while train with multi GPU?
Hi, I'm curious about why the REDUCE time is different with BROADCAST time.  And, the BROADCAST time of each partition is different with others.  Could the reason be...
When I train mnist with pytorch, I found the output accuracy and loss are werid. Then I tried to print it out before push_pull. 
**Describe the bug** I benchmarked BytePS and Horovod's performance using this [script](https://gist.github.com/azuresol/b7e4b332392d95578804dc34e9eaf78f) using 4VM * 8 V100 on TCP. It turned out that the performance I got from BytePS is...
**Describe the bug** Running all instances which are 2 workers, scheduler and server on one node with multiple gpus crashes when one of workers is asked to be run on...
Hi, My understanding is that Byteps uses ps-lite as a thirt party library. Is there any way the bounded delay (as described in ps-lite) could be changed using byteps? Thank...
