Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

AttributeError: module 'byteps.tensorflow' has no attribute 'allreduce' #8

Closed
ppwwyyxx opened this issue Jun 27, 2019 · 3 comments
Closed

AttributeError: module 'byteps.tensorflow' has no attribute 'allreduce' #8

ppwwyyxx opened this issue Jun 27, 2019 · 3 comments
Labels
documentation Improvements or additions to documentation

Comments

@ppwwyyxx
Copy link

Describe the bug
It was claimed in the README that:

If your tasks only rely on Horovod's allreduce and broadcast, you should be able to switch to BytePS in 1 minute. Simply replace import horovod.tensorflow as hvd by import byteps.tensorflow as bps, and then replace all hvd in your code by bps.

However, bps.allreduce does not exist.

To Reproduce
Steps to reproduce the behavior:

  1. import byteps.tensorflow as bps; bps.allreduce

Error:

AttributeError: module 'byteps.tensorflow' has no attribute 'allreduce'
@bobzhuyb
Copy link
Member

You are right. allreduce has been renamed to push_pull in BytePS. We will update the documentation. Please also let us know if bps.push_pull works for you.

@bobzhuyb bobzhuyb added the documentation Improvements or additions to documentation label Jun 27, 2019
@bobzhuyb
Copy link
Member

Added description in README.md. Closing this. Feel free to reopen if needed.

@boscotsang
Copy link

I used hps.allreduce and an error was raised
AttributeError: module 'byteps.torch' has no attribute 'allreduce'
However, I replace hvd.allreduce with hps.push_pull, there was alos an error
AttributeError: module 'byteps.torch' has no attribute 'push_pull'

@ghost ghost mentioned this issue Feb 25, 2020
Closed
pleasantrabbit pushed a commit that referenced this issue Jul 13, 2020
@jasperzhong
hotfix: use concurrent.futures.Timeout (#8)
f6fd608 
* hotfix: fix timeout typo

* hotfix: test timeout

* hotfix: set timeout 0.1s
@DeruiLiu DeruiLiu mentioned this issue Aug 5, 2020
Closed
pleasantrabbit pushed a commit that referenced this issue Nov 3, 2020
@ymjiang
rdma: make the mempool size configurable (#8)
35e3d13
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
documentation Improvements or additions to documentation
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@ppwwyyxx @bobzhuyb @boscotsang