Skip to content

Issues: pytorch/torchft

Beta
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

20 Open 14 Closed
20 Open 14 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

LighthouseClient: support heartbeats enhancement New feature or request good first issue Good for newcomers lighthouse Lighthouse and quorum related
#174 opened Apr 25, 2025 by d4l3k
1
support multiple quorums on the same LighthouseServer enhancement New feature or request lighthouse Lighthouse and quorum related
#173 opened Apr 25, 2025 by d4l3k
torchelastic Rendezvous Backend enhancement New feature or request
#172 opened Apr 25, 2025 by d4l3k
Towards Native Fault Tolerance for Semi-Synchronous Training enhancement New feature or request
#171 opened Apr 23, 2025 by WarrenZhu050413
1
[lighthouse] fast failure on missing heartbeat instead of timeout enhancement New feature or request lighthouse Lighthouse and quorum related
#164 opened Apr 15, 2025 by rualark
2
Explain quorum documentation Improvements or additions to documentation
#163 opened Apr 15, 2025 by rualark
3
Example train_ddp.py run command runs only one process, but there are two groups and it is not explained how to use both documentation Improvements or additions to documentation
#162 opened Apr 15, 2025 by rualark
3
README.md contains train.py, but it is not used and there is no explanation what that is documentation Improvements or additions to documentation
#161 opened Apr 15, 2025 by rualark
3
Add watchdog to _TimeoutManager+ProcessGroupNCCL to guarantee fast aborts enhancement New feature or request process_group related to ProcessGroups and collectives
#152 opened Mar 26, 2025 by d4l3k
2
Add profiling to Manager enhancement New feature or request manager process_group related to ProcessGroups and collectives
#137 opened Mar 17, 2025 by d4l3k
1
Support federated learning via Nebula
#131 opened Mar 11, 2025 by d4l3k
Add config sharing from Lighthouse
#130 opened Mar 11, 2025 by d4l3k
PGTransport in-place transfers checkpoint related to checkpointing/recovery/healing process_group related to ProcessGroups and collectives python
#118 opened Feb 22, 2025 by d4l3k
add profiling to ProcessGroupBaby process_group related to ProcessGroups and collectives python
#116 opened Feb 22, 2025 by d4l3k
@ekdnam
1
Use bucketized model averaging for LocalSGD enhancement New feature or request good first issue Good for newcomers
#66 opened Jan 10, 2025 by d4l3k
@Krishn1412
4
Dataloader question upon restart data Related to dataloading enhancement New feature or request question Further information is requested
#58 opened Jan 7, 2025 by cjolivier01
6
support DDP bucket rebuilding
#41 opened Dec 14, 2024 by d4l3k
LocalSGD / DiLoCo support enhancement New feature or request
#39 opened Dec 13, 2024 by d4l3k
5
[dataloader] dataloading improvement tracking issue data Related to dataloading enhancement New feature or request
#37 opened Dec 12, 2024 by d4l3k
3 tasks
2
[CheckpointServer] use streaming transfers enhancement New feature or request good first issue Good for newcomers
#36 opened Dec 12, 2024 by d4l3k
5
ProTip! Add no:assignee to see everything that’s not assigned.