Skip to content

Issues: Bluefog-Lib/bluefog

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

27 Open 32 Closed
27 Open 32 Closed
Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Milestones
Filter by milestone
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

Work with newer torch
#116 opened Jun 24, 2023 by fecet
Error when calling push-sum optimizer
#112 opened Jun 15, 2022 by yangxuanfei
@bichengying
3
ImportError: /root/miniconda3/envs/bluefog/lib/python3.8/site-packages/bluefog/torch/mpi_lib.cpython-38-x86_64-linux-gnu.so: undefined symbol: _ZNK2at6Tensor6deviceEv
#110 opened May 13, 2022 by yangxuanfei
3
when run "Applying BlueFog on Deep Learning problem(High Level API Introduction)",some error happened
#104 opened Mar 16, 2022 by lkzs
1
Model trained by AWC style cannot be saved
#103 opened Jan 17, 2022 by kunyuan827
test_neighbor_allreduce_dst_weight_fusion failed with MPI CUDA Aware case
#102 opened Dec 13, 2021 by bichengying
@hanbinhu
CUDA initialized even when user didn't use CUDA at all in an environment with GPUs
#100 opened Dec 12, 2021 by hanbinhu
Check the topology is the same cross all agents when call set_topology
#90 opened Apr 18, 2021 by bichengying
@bichengying
Better design pattern for data_weight synchronization
#83 opened Mar 26, 2021 by hanbinhu
Symmetrical argument for self_weight, src_weights, dst_weights
#82 opened Mar 26, 2021 by hanbinhu
Add dst_weight for hierarchical neighbor allreduce
#81 opened Mar 26, 2021 by hanbinhu
Context for CUDA and NCCL
#80 opened Mar 26, 2021 by hanbinhu
1
Benchmark Example issue bug Something isn't working
#79 opened Mar 18, 2021 by hanbinhu
@hanbinhu @bichengying
Update intra communication from MPI window operations to shared memory enhancement New feature or request
#73 opened Jan 24, 2021 by lucweichen
@lucweichen
ATC optimizers requires num_step_per_communication enhancement New feature or request
#66 opened Dec 29, 2020 by hanbinhu
@bichengying
Left-over data for BlueFog optimizers using num_step_per_communication investigation This issues requires more investigation
#65 opened Dec 29, 2020 by hanbinhu
@hanbinhu @bichengying
Add test for hierarchical operations
#57 opened Oct 29, 2020 by bichengying
@bichengying
Better naming for API enhancement New feature or request
#52 opened Oct 7, 2020 by bichengying
@bichengying
Check if bf.barrier() is working properly bug Something isn't working investigation This issues requires more investigation
#51 opened Oct 2, 2020 by hanbinhu
@hanbinhu @bichengying
Robust and readable code refactor for the order of neighbors in Neighbor_allreducce/allgather implementation enhancement New feature or request help wanted Extra attention is needed
#50 opened Oct 1, 2020 by bichengying
@hanbinhu
Use MCS lock instead of Spin Lock for more balance of getting mutex enhancement New feature or request
#48 opened Sep 21, 2020 by bichengying
@bichengying
Mac + OpenMPI 4.0.5 Failed on Window test bug Something isn't working
#42 opened Aug 30, 2020 by Bluefog-Lib
@hanbinhu @bichengying
2
Infiniband Support Test enhancement New feature or request
#27 opened Jun 7, 2020 by Bluefog-Lib
@hanbinhu
Timeline Backward Tracking bug Something isn't working enhancement New feature or request
#24 opened May 19, 2020 by hanbinhu
@hanbinhu @bichengying
How to deal with the case that when one or some processes are much faster than others enhancement New feature or request investigation This issues requires more investigation
#21 opened Apr 14, 2020 by bichengying
@bichengying
1
ProTip! no:milestone will show everything without a milestone.