issues
search
Bluefog-Lib
/
bluefog
Distributed and decentralized training framework for PyTorch over graph
https://bluefog-lib.github.io/bluefog/
Apache License 2.0
291
stars
71
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Interactive bluefog
#68
BichengYing
closed
3 years ago
1
Win put optimizer register name issue
#67
hanbinhu
closed
3 years ago
0
ATC optimizers requires num_step_per_communication
#66
hanbinhu
opened
3 years ago
0
Left-over data for BlueFog optimizers using num_step_per_communication
#65
hanbinhu
opened
3 years ago
0
Optimizer test
#64
hanbinhu
closed
3 years ago
0
Atc
#63
BichengYing
closed
3 years ago
0
Static hier topo
#62
hanbinhu
closed
3 years ago
0
Hier neighbor allreduce
#61
BichengYing
closed
3 years ago
0
Add static machine topology for hierarchical_neighbor_allreduce usage
#60
BichengYing
closed
3 years ago
1
Remove neighbor allreduce limitation
#59
hanbinhu
closed
3 years ago
0
Hier neighbor allreduce
#58
BichengYing
closed
3 years ago
0
Add test for hierarchical operations
#57
BichengYing
opened
3 years ago
0
Hierarchical dynamic graph
#56
BichengYing
closed
3 years ago
0
Update .travis.yml
#55
BichengYing
closed
3 years ago
0
Update optimizer for num_steps_per_communication
#54
hanbinhu
closed
3 years ago
0
Version
#53
lucweichen
closed
3 years ago
0
Better naming for API
#52
BichengYing
opened
3 years ago
0
Check if bf.barrier() is working properly
#51
hanbinhu
opened
3 years ago
0
Robust and readable code refactor for the order of neighbors in Neighbor_allreducce/allgather implementation
#50
BichengYing
opened
3 years ago
0
Add Tensor Fusion
#49
BichengYing
closed
3 years ago
0
Use MCS lock instead of Spin Lock for more balance of getting mutex
#48
BichengYing
opened
4 years ago
0
Add Negotiate Stage
#47
BichengYing
closed
3 years ago
0
Nccl win
#46
Bluefog-Lib
closed
4 years ago
0
Rename Power2 To Exponential 2 Network in codebase
#45
Bluefog-Lib
closed
3 years ago
1
NCCL an illegal memory access was encountered when running with 244*244*3 size dataset
#44
Bluefog-Lib
closed
3 years ago
2
Add Half tensor to MPI operations
#43
hanbinhu
closed
4 years ago
0
Mac + OpenMPI 4.0.5 Failed on Window test
#42
Bluefog-Lib
opened
4 years ago
2
Add Callback to wrap MPI operations
#41
hanbinhu
closed
4 years ago
0
Associate weight with p [For push-sum algorithm]
#40
Bluefog-Lib
closed
4 years ago
0
Allow an API that cancel other process's running communication.
#39
Bluefog-Lib
closed
3 years ago
1
Partial Neighbor Allreduce Implementation under NCCL
#38
BichengYing
closed
4 years ago
0
Failure on unit test with torch.cuda.DoubleTensor
#37
BichengYing
closed
4 years ago
2
Dynamic topo neighbor allreduce
#36
hanbinhu
closed
4 years ago
0
Neighbor Allreduce divided by zero error when -np 1
#35
BichengYing
closed
4 years ago
0
Add a simple Block Gossip routing
#34
BichengYing
closed
4 years ago
0
Move determining is_homogenenous function from mpi_context to mpi_con…
#33
hanbinhu
closed
4 years ago
0
Bluefog didn't throw an error when CUDA memory is not enough.
#32
hanbinhu
closed
4 years ago
3
NCCL issue with illegal memory access
#31
hanbinhu
closed
3 years ago
3
is_homogeneous in mpi_context causes double free memory issue
#30
hanbinhu
closed
4 years ago
1
Add NCCL Controller
#29
Bluefog-Lib
closed
4 years ago
0
Add Environment Variable Document
#28
Bluefog-Lib
closed
4 years ago
1
Infiniband Support Test
#27
Bluefog-Lib
opened
4 years ago
0
NaN Numerical Error in Neighbor_Allreduce
#26
Bluefog-Lib
closed
4 years ago
2
NCCL 2.7 Support Neighbor Ops
#25
Bluefog-Lib
closed
4 years ago
1
Timeline Backward Tracking
#24
hanbinhu
opened
4 years ago
0
Forward hook bluefog
#23
Bluefog-Lib
closed
4 years ago
0
neighbor_allreduce interface change
#22
hanbinhu
closed
4 years ago
0
How to deal with the case that when one or some processes are much faster than others
#21
BichengYing
opened
4 years ago
1
Proposal for local GPU communication merging
#20
BichengYing
opened
4 years ago
1
topology weight definition change on C side
#19
hanbinhu
closed
4 years ago
0
Previous
Next