Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / Bluefog-Lib/bluefog issues and pull requests

#118 - Bump torch from 1.4.0 to 2.2.0

Pull Request - State: open - Opened by dependabot[bot] 2 months ago
Labels: dependencies

#117 - compressor

Pull Request - State: open - Opened by xuyufei-a 5 months ago

#116 - Work with newer torch

Issue - State: open - Opened by fecet over 1 year ago

#115 - Fixing verification of the hearbeat value

Pull Request - State: closed - Opened by dgumenyuk over 1 year ago - 1 comment

#114 - Argument "disable_heartbeat" does not exist

Issue - State: closed - Opened by dgumenyuk over 1 year ago - 1 comment

#113 - Is it possible to run more agents than the number of my CPU cores?

Issue - State: closed - Opened by 1qzhworld over 2 years ago - 2 comments

#112 - Error when calling push-sum optimizer

Issue - State: open - Opened by yangxuanfei over 2 years ago - 3 comments

#111 - Problems running decentralized trainning

Issue - State: closed - Opened by yangxuanfei over 2 years ago

#109 - Add ref

Pull Request - State: closed - Opened by kunyuan827 over 2 years ago

#108 - some error happened

Issue - State: closed - Opened by northhj over 2 years ago - 3 comments

#107 - when I Install Bluefog from Pip (GPU),some error happens

Issue - State: closed - Opened by lkzs over 2 years ago - 6 comments

#106 - Mypy

Pull Request - State: open - Opened by hanbinhu over 2 years ago

#105 - Update README.rst

Pull Request - State: open - Opened by ybc1991 over 2 years ago

#103 - Model trained by AWC style cannot be saved

Issue - State: open - Opened by kunyuan827 over 2 years ago

#101 - Fix the cuda stream creation in MPI

Pull Request - State: closed - Opened by Bluefog-Lib almost 3 years ago

#99 - Add -mca ^openib flag to test

Pull Request - State: closed - Opened by ybc1991 almost 3 years ago

#98 - add Troubleshooting & doc bugfix

Pull Request - State: closed - Opened by ymchen7 almost 3 years ago

#97 - Add BlueFog arxiv paper

Pull Request - State: closed - Opened by ybc1991 almost 3 years ago

#96 - Add hierarchical related ops test

Pull Request - State: closed - Opened by ybc1991 over 3 years ago - 1 comment

#95 - Allow to control local size by environment variable

Pull Request - State: closed - Opened by ybc1991 over 3 years ago

#94 - Docker release

Pull Request - State: closed - Opened by hanbinhu over 3 years ago
Labels: Automation Flow

#93 - Release flow

Pull Request - State: closed - Opened by hanbinhu over 3 years ago
Labels: Automation Flow

#92 - Condition variable

Pull Request - State: closed - Opened by ybc1991 over 3 years ago

#91 - Add deprecation args and fix the comments in neighbor_allreduce

Pull Request - State: closed - Opened by ybc1991 over 3 years ago

#89 - Disable heartbeat by default

Pull Request - State: closed - Opened by ybc1991 over 3 years ago

#88 - Add condition variable to control the loop

Pull Request - State: closed - Opened by ybc1991 over 3 years ago

#87 - Condition variable

Pull Request - State: closed - Opened by ybc1991 over 3 years ago - 1 comment

#86 - Topo service

Pull Request - State: closed - Opened by ybc1991 over 3 years ago

#85 - Revert "Topo service (#75)"

Pull Request - State: closed - Opened by ybc1991 over 3 years ago

#84 - No-op: format topo file only by Black

Pull Request - State: open - Opened by ybc1991 over 3 years ago

#83 - Better design pattern for data_weight synchronization

Issue - State: open - Opened by hanbinhu over 3 years ago

#81 - Add dst_weight for hierarchical neighbor allreduce

Issue - State: open - Opened by hanbinhu over 3 years ago

#80 - Context for CUDA and NCCL

Issue - State: open - Opened by hanbinhu over 3 years ago - 1 comment

#79 - Benchmark Example issue

Issue - State: open - Opened by hanbinhu over 3 years ago
Labels: bug

#78 - Improve neighbor allreduce

Pull Request - State: closed - Opened by hanbinhu over 3 years ago
Labels: enhancement

#77 - Create doc.yml

Pull Request - State: closed - Opened by ybc1991 over 3 years ago

#76 - Add github action

Pull Request - State: closed - Opened by ybc1991 over 3 years ago

#75 - Topo service

Pull Request - State: closed - Opened by ybc1991 over 3 years ago

#74 - Dynamic neighbor allgather

Pull Request - State: closed - Opened by ybc1991 over 3 years ago

#73 - Update intra communication from MPI window operations to shared memory

Issue - State: open - Opened by lucweichen over 3 years ago
Labels: enhancement

#72 - Add API to unregister window

Pull Request - State: closed - Opened by hanbinhu over 3 years ago - 1 comment
Labels: bug, enhancement

#71 - ATC multi-step case

Pull Request - State: closed - Opened by ybc1991 over 3 years ago

#70 - Optimizer num_step_per_communication behavior change and test

Pull Request - State: closed - Opened by hanbinhu over 3 years ago
Labels: bug, enhancement

#69 - Interactive bluefog

Pull Request - State: closed - Opened by kunyuan827 over 3 years ago

#68 - Interactive bluefog

Pull Request - State: closed - Opened by ybc1991 almost 4 years ago - 1 comment

#67 - Win put optimizer register name issue

Issue - State: closed - Opened by hanbinhu almost 4 years ago
Labels: bug, good first issue, investigation

#66 - ATC optimizers requires num_step_per_communication

Issue - State: open - Opened by hanbinhu almost 4 years ago
Labels: enhancement

#65 - Left-over data for BlueFog optimizers using num_step_per_communication

Issue - State: open - Opened by hanbinhu almost 4 years ago
Labels: investigation

#64 - Optimizer test

Pull Request - State: closed - Opened by hanbinhu almost 4 years ago

#63 - Atc

Pull Request - State: closed - Opened by ybc1991 almost 4 years ago

#62 - Static hier topo

Pull Request - State: closed - Opened by hanbinhu almost 4 years ago

#61 - Hier neighbor allreduce

Pull Request - State: closed - Opened by ybc1991 almost 4 years ago

#60 - Add static machine topology for hierarchical_neighbor_allreduce usage

Issue - State: closed - Opened by ybc1991 almost 4 years ago - 1 comment

#59 - Remove neighbor allreduce limitation

Pull Request - State: closed - Opened by hanbinhu almost 4 years ago

#58 - Hier neighbor allreduce

Pull Request - State: closed - Opened by ybc1991 almost 4 years ago

#57 - Add test for hierarchical operations

Issue - State: open - Opened by ybc1991 almost 4 years ago

#56 - Hierarchical dynamic graph

Pull Request - State: closed - Opened by ybc1991 almost 4 years ago

#55 - Update .travis.yml

Pull Request - State: closed - Opened by ybc1991 almost 4 years ago

#54 - Update optimizer for num_steps_per_communication

Pull Request - State: closed - Opened by hanbinhu almost 4 years ago

#53 - Version

Pull Request - State: closed - Opened by lucweichen almost 4 years ago

#52 - Better naming for API

Issue - State: open - Opened by ybc1991 almost 4 years ago
Labels: enhancement

#51 - Check if bf.barrier() is working properly

Issue - State: open - Opened by hanbinhu about 4 years ago
Labels: bug, investigation

#50 - Robust and readable code refactor for the order of neighbors in Neighbor_allreducce/allgather implementation

Issue - State: open - Opened by ybc1991 about 4 years ago
Labels: enhancement, help wanted

#49 - Add Tensor Fusion

Pull Request - State: closed - Opened by ybc1991 about 4 years ago

#48 - Use MCS lock instead of Spin Lock for more balance of getting mutex

Issue - State: open - Opened by ybc1991 about 4 years ago
Labels: enhancement

#47 - Add Negotiate Stage

Pull Request - State: closed - Opened by ybc1991 about 4 years ago
Labels: enhancement

#46 - Nccl win

Pull Request - State: closed - Opened by Bluefog-Lib about 4 years ago

#45 - Rename Power2 To Exponential 2 Network in codebase

Issue - State: closed - Opened by Bluefog-Lib about 4 years ago - 1 comment

#44 - NCCL an illegal memory access was encountered when running with 244*244*3 size dataset

Issue - State: closed - Opened by Bluefog-Lib about 4 years ago - 2 comments
Labels: bug

#43 - Add Half tensor to MPI operations

Pull Request - State: closed - Opened by hanbinhu about 4 years ago
Labels: enhancement

#42 - Mac + OpenMPI 4.0.5 Failed on Window test

Issue - State: open - Opened by Bluefog-Lib about 4 years ago - 2 comments
Labels: bug

#41 - Add Callback to wrap MPI operations

Pull Request - State: closed - Opened by hanbinhu about 4 years ago
Labels: enhancement

#40 - Associate weight with p [For push-sum algorithm]

Pull Request - State: closed - Opened by Bluefog-Lib about 4 years ago
Labels: enhancement

#39 - Allow an API that cancel other process's running communication.

Issue - State: closed - Opened by Bluefog-Lib about 4 years ago - 1 comment
Labels: enhancement

#38 - Partial Neighbor Allreduce Implementation under NCCL

Pull Request - State: closed - Opened by ybc1991 about 4 years ago
Labels: enhancement

#37 - Failure on unit test with torch.cuda.DoubleTensor

Issue - State: closed - Opened by ybc1991 about 4 years ago - 2 comments
Labels: bug

#36 - Dynamic topo neighbor allreduce

Pull Request - State: closed - Opened by hanbinhu about 4 years ago
Labels: enhancement

#35 - Neighbor Allreduce divided by zero error when -np 1

Issue - State: closed - Opened by ybc1991 about 4 years ago

#34 - Add a simple Block Gossip routing

Pull Request - State: closed - Opened by ybc1991 about 4 years ago
Labels: enhancement

#33 - Move determining is_homogenenous function from mpi_context to mpi_con…

Pull Request - State: closed - Opened by hanbinhu about 4 years ago
Labels: bug

#32 - Bluefog didn't throw an error when CUDA memory is not enough.

Issue - State: closed - Opened by hanbinhu about 4 years ago - 3 comments
Labels: bug

#31 - NCCL issue with illegal memory access

Issue - State: closed - Opened by hanbinhu about 4 years ago - 3 comments
Labels: bug

#30 - is_homogeneous in mpi_context causes double free memory issue

Issue - State: closed - Opened by hanbinhu about 4 years ago - 1 comment
Labels: bug

#29 - Add NCCL Controller

Pull Request - State: closed - Opened by Bluefog-Lib over 4 years ago

#28 - Add Environment Variable Document

Issue - State: closed - Opened by Bluefog-Lib over 4 years ago - 1 comment
Labels: documentation

#27 - Infiniband Support Test

Issue - State: open - Opened by Bluefog-Lib over 4 years ago
Labels: enhancement

#26 - NaN Numerical Error in Neighbor_Allreduce

Issue - State: closed - Opened by Bluefog-Lib over 4 years ago - 2 comments
Labels: bug

#25 - NCCL 2.7 Support Neighbor Ops

Issue - State: closed - Opened by Bluefog-Lib over 4 years ago - 1 comment
Labels: enhancement

#24 - Timeline Backward Tracking

Issue - State: open - Opened by hanbinhu over 4 years ago
Labels: bug, enhancement

#23 - Forward hook bluefog

Pull Request - State: closed - Opened by Bluefog-Lib over 4 years ago

#22 - neighbor_allreduce interface change

Pull Request - State: closed - Opened by hanbinhu over 4 years ago
Labels: enhancement

#20 - Proposal for local GPU communication merging

Issue - State: open - Opened by ybc1991 over 4 years ago - 1 comment

#15 - Mysterious behavior of requiring setting CUDA device explicitly in OpenMPI 1.10.7

Issue - State: closed - Opened by hanbinhu over 4 years ago - 2 comments
Labels: bug, investigation

#14 - Unifying weight definition

Issue - State: closed - Opened by hanbinhu over 4 years ago
Labels: enhancement