Commits

Branch selector

 master

User selector

nv-dlasalle

Datepicker

All time

Commit History

Commits on Jan 28, 2022

• Fix race condition in creating directory (dmlc#3696)

  Show description for 268f617
  

  nv-dlasalle

  and

  BarclayII

  authored
  268f617

  Copy full SHA for 268f617

Commits on Nov 3, 2021

• Update cub for cuda 11.5 compatibility (dmlc#3468)

  nv-dlasalle

  authored
  f510214

  Copy full SHA for f510214

Commits on Oct 18, 2021

• [Fix] Split nccl sparse push into two groups (dmlc#3404)

  nv-dlasalle

  authored
  c560040

  Copy full SHA for c560040

Commits on Oct 14, 2021

• [PyTorch][Bugfix] Use uint8 instead of bool in pytorch to be compatible with nightly version (dmlc#3406)

  Show description for b81efb2
  

  nv-dlasalle

  and

  VoVAllen

  authored
  b81efb2

  Copy full SHA for b81efb2

Commits on Sep 20, 2021

• Enable faster validation for pytorch graphsage example (dmlc#3361)

  nv-dlasalle

  authored
  01a2214

  Copy full SHA for 01a2214

Commits on Sep 16, 2021

• [Performance][Feature] Add src_nodes paramter to to_block() to avoid cost running unique() when available. (dmlc#2973)

  Show description for 2647afc
  

  5 peopleauthored
  2647afc

  Copy full SHA for 2647afc

Commits on Aug 31, 2021

• [CPU][Sampling][Performance] Improve sampling on the CPU. (dmlc#3274)

  Show description for 8e525da
  

  nv-dlasalle

  and

  zheng-da

  authored
  8e525da

  Copy full SHA for 8e525da

• [Performance] Only copy and slice index's in subcolumn when data is used. (dmlc#2930)

  Show description for a53783c
  

  nv-dlasalle

  and

  BarclayII

  authored
  a53783c

  Copy full SHA for a53783c

Commits on Aug 20, 2021

• [Feature][DistDGL] Add NCCL support for range based partitions (dmlc#3213)

  Show description for 7f92793
  nv-dlasalle

  authored
  7f92793

  Copy full SHA for 7f92793

Commits on Aug 19, 2021

• [Performance][Feature] Implement edge excluding in EdgeDataLoader on GPU (dmlc#3226)

  Show description for f634950
  

  3 peopleauthored
  f634950

  Copy full SHA for f634950

• [bugfix] Link against pytorch library by name rather than path (attempting to fix linking issue). (dmlc#3246)

  Show description for fc6f0b9
  

  nv-dlasalle

  and

  BarclayII

  authored
  fc6f0b9

  Copy full SHA for fc6f0b9

Commits on Aug 8, 2021

• [bugfix] Only link the pytorch tensordispatcher against libtorch (dmlc#3225)

  Show description for 799c091
  nv-dlasalle

  authored
  799c091

  Copy full SHA for 799c091

Commits on Aug 6, 2021

• [Fix] FIx check for num_workers for using ScalarDataBatcher (dmlc#3219)

  Show description for 4d9beb3
  

  3 peopleauthored
  4d9beb3

  Copy full SHA for 4d9beb3

Commits on Aug 2, 2021

• [bugfix] Fix curand_init() calls in rowwise sampling (dmlc#3196)

  Show description for f7ce267
  

  3 peopleauthored
  f7ce267

  Copy full SHA for f7ce267

• Improve usage of pinned memory in sparse_optimizer (dmlc#3207)

  Show description for 6001001
  

  nv-dlasalle

  and

  classicsong

  authored
  6001001

  Copy full SHA for 6001001

Commits on Jul 2, 2021

• [Feature] Add dgl.utils.is_sorted_srcdst() (dmlc#2685)

  Show description for a0390dd
  

  3 peopleauthored
  a0390dd

  Copy full SHA for a0390dd

Commits on Jun 29, 2021

• [Example][Bugfix] Fix RGCN example datasplitting (dmlc#3037)

  Show description for 02e5d47
  

  nv-dlasalle

  and

  classicsong

  authored
  02e5d47

  Copy full SHA for 02e5d47

Commits on Jun 23, 2021

• [Bugfix] Handle case where process has no elements to update, in NCCL communicator (dmlc#3035)

  Show description for 7415eaa
  

  nv-dlasalle

  and

  jermainewang

  authored
  7415eaa

  Copy full SHA for 7415eaa

Commits on Jun 22, 2021

• [Feature][GPU] Add function for setting weights of a sparse embedding on multiple GPUs. (dmlc#3047)

  Show description for 7359481
  

  nv-dlasalle

  and

  classicsong

  authored
  7359481

  Copy full SHA for 7359481

Commits on Jun 16, 2021

• Add Ampere support to cmake files (dmlc#3031)

  Show description for 7d069d6
  nv-dlasalle

  authored
  7d069d6

  Copy full SHA for 7d069d6

Commits on Jun 15, 2021

• [bugfix] Fix sparse_optim when the state is stored on the CPU (fixes dmlc#2760) (dmlc#3013)

  Show description for 8932285
  

  nv-dlasalle

  and

  classicsong

  authored
  8932285

  Copy full SHA for 8932285

Commits on Jun 13, 2021

• [Performance] Perform to_block on the GPU when the dataloader is created with a GPU device. (dmlc#3016)

  Show description for 8b64ae5
  

  nv-dlasalle

  and

  zheng-da

  authored
  8b64ae5

  Copy full SHA for 8b64ae5

• Fix example to test_acc is always computed at last epoch (dmlc#3014)

  nv-dlasalle

  authored
  79b057f

  Copy full SHA for 79b057f

Commits on Jun 11, 2021

• [Feature] Allow using NCCL for communication in dgl.NodeEmbedding and dgl.SparseOptimizer (dmlc#2824)

  Show description for 17d604b
  

  nv-dlasalle

  and

  classicsong

  authored
  17d604b

  Copy full SHA for 17d604b

Commits on Jun 10, 2021

• [Bugfix][Examples] Fix graphsage multigpu training example training set size (dmlc#3002)

  Show description for 9497a9b
  

  nv-dlasalle

  and

  BarclayII

  authored
  9497a9b

  Copy full SHA for 9497a9b

Commits on Jun 1, 2021

• [Testing] Enable C++ tests on GPU (dmlc#2945)

  Show description for 86ba233
  nv-dlasalle

  authored
  86ba233

  Copy full SHA for 86ba233

Commits on May 25, 2021

• Move pointer dereferencing in CDFSampler::draw() and AliasSampler::draw() to inside of conditional (dmlc#2943)

  Show description for 1db4ad4
  

  nv-dlasalle

  and

  BarclayII

  authored
  1db4ad4

  Copy full SHA for 1db4ad4

• [Bugfix] Include NCCL as a submodule (dmlc#2934)

  Show description for 66eb240
  

  nv-dlasalle

  and

  BarclayII

  authored
  66eb240

  Copy full SHA for 66eb240

Commits on May 20, 2021

• [Feature][Performance] Implement NCCL wrapper for communicating NodeEmbeddings and sparse gradients. (dmlc#2825)

  Show description for ae8dbe6
  

  nv-dlasalle

  and

  classicsong

  authored
  ae8dbe6

  Copy full SHA for ae8dbe6

Commits on Apr 16, 2021

• [Performance] Track sorted status of COO from creation (dmlc#2645)

  Show description for bbebde4
  

  nv-dlasalle

  and

  jermainewang

  authored
  bbebde4

  Copy full SHA for bbebde4

Commits on Apr 15, 2021

• [Performance][GPU] Enable GPU uniform edge sampling (dmlc#2716)

  Show description for e70138b
  

  nv-dlasalle

  and

  zheng-da

  authored
  e70138b

  Copy full SHA for e70138b

Commits on Mar 22, 2021

• [Bugfix] Wrap cub with CUB_NS_PREFIX and remove dependency on Thrust to linking issues with Torch 1.8 (dmlc#2758)

  Show description for 0ff7127
  

  3 peopleauthored
  0ff7127

  Copy full SHA for 0ff7127

Commits on Feb 8, 2021

• [Sampling] Implement dgl.to_block() for the GPU (dmlc#2339)

  Show description for bc3a532
  

  3 peopleauthored
  bc3a532

  Copy full SHA for bc3a532

Commits on Jan 27, 2021

• [Performance] Improve COO to CSR, and sort columns of CSR only when necessary. (dmlc#2391)

  Show description for 2576647
  nv-dlasalle

  authored
  2576647

  Copy full SHA for 2576647

Commits on Jan 24, 2021

• [Fix] Don't create extra formats in dataloader (dmlc#2551)

  Show description for a6abffe
  

  nv-dlasalle

  and

  jermainewang

  authored
  a6abffe

  Copy full SHA for a6abffe

Pagination

Previous

Next