Commits

Breadcrumbs

History for

onmaster

User selector

All users

Datepicker

All time

Commit History

Commits on Jul 21, 2019

• TF-Keras support for TensorFlow 2.0 (horovod#1238)

  Show description for 5e69053
  alsrgv

  authored
  5e69053

  Copy full SHA for 5e69053

Commits on Jul 8, 2019

• Initial support for TensorFlow 2.0 (horovod#1193)

  Show description for 15692da
  alsrgv

  authored
  15692da

  Copy full SHA for 15692da

Commits on Jun 19, 2019

• fix imagenet example with synthetic input (horovod#1155)

  Show description for 50a3c01
  

  apeforest

  authored and

  alsrgv

  committed
  50a3c01

  Copy full SHA for 50a3c01

Commits on Apr 14, 2019

• shebang should come before comments (horovod#1013)

  Show description for da85af2
  

  afq984

  authored and

  alsrgv

  committed
  da85af2

  Copy full SHA for da85af2

Commits on Apr 3, 2019

• MXNet: Normalize rescale_grad in optimizer by Horovod size (horovod#976)

  Show description for af11495
  

  yuxihu

  authored and

  alsrgv

  committed
  af11495

  Copy full SHA for af11495

Commits on Mar 30, 2019

• Add broadcast_optimizer_state to pytorch_mnist.py (horovod#967)

  Show description for 3b41faa
  alsrgv

  authored
  3b41faa

  Copy full SHA for 3b41faa

Commits on Mar 29, 2019

• MXNet: add DistributedTrainer for Gluon (horovod#943)

  Show description for 5d8f80c
  

  yuxihu

  authored and

  alsrgv

  committed
  5d8f80c

  Copy full SHA for 5d8f80c

Commits on Mar 28, 2019

• MXNet: support broadcasting deferred initialization parameters in Gluon (horovod#915)

  Show description for 3a1083e
  

  yuxihu

  authored and

  alsrgv

  committed
  3a1083e

  Copy full SHA for 3a1083e

Commits on Mar 27, 2019

• Make Spark example Python3 compatible (horovod#959)

  Show description for 7f20fb7
  alsrgv

  authored
  7f20fb7

  Copy full SHA for 7f20fb7

Commits on Mar 8, 2019

• Simplify mnist example using Gluon API and update README (horovod#886)

  Show description for 91d5ae8
  

  apeforest

  authored and

  alsrgv

  committed
  91d5ae8

  Copy full SHA for 91d5ae8

Commits on Mar 2, 2019

• Update README.md (horovod#872)

  Show description for 9dbf8ff
  

  eric-haibin-lin

  authored and

  alsrgv

  committed
  9dbf8ff

  Copy full SHA for 9dbf8ff

Commits on Feb 22, 2019

• Bugfix PyTorch ImageNet example (horovod#853)

  Show description for ca3c1e3
  alsrgv

  authored
  ca3c1e3

  Copy full SHA for ca3c1e3

Commits on Feb 21, 2019

• Make tensorflow_mnist_eager.py Python2-friendly (horovod#846)

  Show description for 88b7825
  alsrgv

  authored
  88b7825

  Copy full SHA for 88b7825

Commits on Feb 7, 2019

• MXNet: support Gluon Trainer API (horovod#809)

  

  yuxihu

  authored and

  alsrgv

  committed
  f7a24e6

  Copy full SHA for f7a24e6

Commits on Feb 4, 2019

• Align broadcast_variables() API with broadcast() (horovod#807)

  alsrgv

  authored
  6643d6a

  Copy full SHA for 6643d6a

Commits on Jan 30, 2019

• MXNet: improve imagenet example performance (horovod#794)

  Show description for 1feb955
  

  yuxihu

  authored and

  alsrgv

  committed
  1feb955

  Copy full SHA for 1feb955

Commits on Jan 22, 2019

• Horovod support for MXNet framework (horovod#542)

  

  ctcyang

  authored and

  alsrgv

  committed
  c36e964

  Copy full SHA for c36e964

Commits on Jan 15, 2019

• Fix CI: use per-rank dataset file for TF eager (horovod#756)

  alsrgv

  authored
  de4cc08

  Copy full SHA for de4cc08

Commits on Jan 14, 2019

• PyTorch MNIST example improvements (horovod#754)

  alsrgv

  authored
  41d6dc8

  Copy full SHA for 41d6dc8

• Make PyTorch MNIST example more documented (horovod#753)

  alsrgv

  authored
  3628655

  Copy full SHA for 3628655

Commits on Jan 9, 2019

• Support for TensorFlow GradientTape (horovod#670)

  

  kuroko1t

  authored and

  alsrgv

  committed
  65bb358

  Copy full SHA for 65bb358

Commits on Dec 31, 2018

• Horovod in PySpark (horovod#606)

  alsrgv

  authored
  3cc0f1e

  Copy full SHA for 3cc0f1e

Commits on Dec 14, 2018

• Added parallelization of allreduce during eager execution and synthetic benchmark for TensorFlow (horovod#704)

  Show description for af0153c
  tgaddair

  authored
  af0153c

  Copy full SHA for af0153c

Commits on Dec 12, 2018

• Enable local gradient accumulation (horovod#546)

  

  andfoy

  authored and

  alsrgv

  committed
  9081ba3

  Copy full SHA for 9081ba3

Commits on Dec 10, 2018

• Removed depreciated tensorflow dataset APIs (horovod#680)

  

  abditag2

  authored and

  alsrgv

  committed
  7d90bd1

  Copy full SHA for 7d90bd1

Commits on Nov 22, 2018

• Fix memory leak in pytorch_imagenet_resnet50.py example (horovod#641)

  

  Julieta

  authored and

  alsrgv

  committed
  6bac3dd

  Copy full SHA for 6bac3dd

Commits on Oct 8, 2018

• PyTorch Synthetic Benchmark (horovod#545)

  alsrgv

  authored
  983a06e

  Copy full SHA for 983a06e

Commits on Sep 29, 2018

• Added fp16 to pytorch_imagenet_resnet50 example (horovod#535)

  tgaddair

  authored
  17459d8

  Copy full SHA for 17459d8

Commits on Sep 28, 2018

• FP16 support for GPU tensors in all frameworks (horovod#529)

  Show description for b2e6c06
  tgaddair

  authored
  b2e6c06

  Copy full SHA for b2e6c06

Commits on Sep 26, 2018

• PyTorch 1.0 readiness updates (horovod#523)

  Show description for 67c7f88
  alsrgv

  authored
  67c7f88

  Copy full SHA for 67c7f88

Commits on Aug 2, 2018

• Added command line args to Keras ImageNet examples (horovod#419)

  Show description for ed18e31
  tgaddair

  authored
  ed18e31

  Copy full SHA for ed18e31

Commits on Jul 14, 2018

• Replace hvd.broadcast_object() with hvd.broadcast_optimizer_state() (horovod#379)

  alsrgv

  authored
  6104a70

  Copy full SHA for 6104a70

Commits on Jul 13, 2018

• Added custom load_model function to wrap the model optimizer with a Horovod DistributedOptimizer (horovod#359)

  Show description for 60d8c14
  tgaddair

  authored
  60d8c14

  Copy full SHA for 60d8c14

• Added PyTorch support for restoring optimizer state on model load and broadcast (horovod#371)

  Show description for 85b09a9
  tgaddair

  authored
  85b09a9

  Copy full SHA for 85b09a9

Commits on Jul 3, 2018

• Made synchronous pure functional Horovod ops differentiable in PyTorch (horovod#338)

  Show description for 7ebf116
  tgaddair

  authored
  7ebf116

  Copy full SHA for 7ebf116

Pagination

Previous

Next