AttributeError: module 'torch.distributed' has no attribute 'is_initialized'

See original GitHub issue

🐛 Bug

When running the webcam demo on macOS 10.14.1 as so:

python webcam.py --min-image-size 300 MODEL.DEVICE cpu

I get the error:

AttributeError: module 'torch.distributed' has no attribute 'is_initialized'

To Reproduce

Steps to reproduce the behavior:

  1. Followed install guide here: https://github.com/facebookresearch/maskrcnn-benchmark/blob/master/INSTALL.md
  2. Use the following command:

python webcam.py --min-image-size 300 MODEL.DEVICE cpu

Expected behavior

I expect this to run the webcam demo.

Environment

  • PyTorch Version (e.g., 1.0): 1.0 (nightly)
  • OS (e.g., Linux): macOS 10.14.1
  • How you installed PyTorch (conda, pip, source): Conda
  • Build command you used (if compiling from source):
  • Python version: 3.7
  • CUDA/cuDNN version: N/A
  • GPU models and configuration: N/A
  • Any other relevant information:

Additional context

I fixed this issue by changing all instances of is_initialized() to is_available() in maskrcnn-benchmark/maskrcnn_benchmark/utils/comm.py. I’m not sure if this is what is intended or if I am installing something incorrectly. After making that change, I was able to run the webcam demo.

Issue Analytics

  • State:closed
  • Created 5 years ago
  • Comments:9 (8 by maintainers)

github_iconTop GitHub Comments

1reaction
congve1commented, Dec 17, 2018

I’ve got the same error. You could check what torch.distributed.is_available() returns. If False, it’s obvious that your pytorch version doesn’t support distributing training, and the code may face some error

0reactions
fmassacommented, Feb 26, 2019
Read more comments on GitHub >

github_iconTop Results From Across the Web

module 'torch.distributed' has no attribute 'is_initialized' in ...
In order to solve this problem. Actually Window and Mac doesn't support distributed training facility. so this issue is occuring.
Read more >
Module 'torch.distributed' has no attribute 'is_initialized'
I am running inference using mmdetection (https://github.com/open-mmlab/mmdetection) and I get the above error for this piece of code; ...
Read more >
Distributed error. module 'torch.distributed' has no attribute ...
I've installed pytorch 1.0 on windows. When I try to use webcam demo provided by maskrcnn-benchmark. An error occured: Traceback (most ...
Read more >
Distributed RPC Framework — PyTorch 1.13 documentation
Not all features of the RPC package are yet compatible with CUDA support and thus their use ... torch.distributed.rpc.init_rpc(name, backend=None, rank=- 1, ...
Read more >
torchrun (Elastic Launch) — PyTorch 1.13 documentation
Transitioning from torch.distributed.launch to torchrun ... a Torch process group are provided to you by this module, no need for you to pass...
Read more >

github_iconTop Related Medium Post

No results found

github_iconTop Related StackOverflow Question

No results found

github_iconTroubleshoot Live Code

Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start Free

github_iconTop Related Reddit Thread

No results found

github_iconTop Related Hackernoon Post

No results found

github_iconTop Related Tweet

No results found

github_iconTop Related Dev.to Post

No results found

github_iconTop Related Hashnode Post

No results found