microsoft/DeepSpeed

Pull Requests Commits

Merge branch 'master' into tohtana/fix_sort_dp_univ_ckpt

tohtana committed 1 year ago

Verified b3d618b3

Parallel map step for `DistributedDataAnalyzer` map-reduce (#5291)

bm-synth committed 1 year ago

Verified 64defe65

Bug fix in `split_index` method (#5292)

bm-synth committed 1 year ago

Verified aaaf8bc5

Add required paths to trigger AMD tests on PRs (#5406)

loadams committed 1 year ago

Verified 3194fe85

[CPU] Support SHM based inference_all_reduce in TorchBackend (#5391)

delock committed 1 year ago

Verified b22706a7

Add more tests into XPU CI (#5427)

Liangliang-Ma committed 1 year ago

Verified 4d4c99d9

Update _sidebar.scss (#5293)

fasterinnerlooper committed 1 year ago

Verified a9cbd688

Update engine.py to avoid torch warning (#5408)

etiennebonnafoux committed 1 year ago

Verified bc0f7747

Update 'create-pr' action in release workflow to latest (#5415)

loadams committed 1 year ago

Verified 34969d20

Fix a convergence issues in TP topology caused by incorrect grad_norm. (#5411)

inkcherry committed 1 year ago

Verified 0896503e

[XPU] Use non_daemonic_proc by default on XPU device (#5412)

ys950902 committed 1 year ago

Verified 258e500f

Merge branch 'master' into tohtana/fix_sort_dp_univ_ckpt

tjruwase committed 1 year ago

Verified d1b13fe6

Remove dtype(fp16) condition check for residual_add unit test (#5329)

raza-sikander committed 1 year ago

Verified 89491053

Update version.txt after 0.14.1 release (#5413)

mrwyattii committed 1 year ago

Verified 2b3d31f9

Fix the FP6 kernels compilation problem on non-Ampere GPUs. (#5333)

JamesTheZ committed 1 year ago

Verified e3d873a0

Merge branch 'master' into tohtana/fix_sort_dp_univ_ckpt

tohtana committed 1 year ago

Verified fd683e01

stage3: efficient compute of scaled_global_grad_norm (#5256)

nelyahu committed 1 year ago

Verified 54c06872

fix pagable h2d memcpy (#5301)

GuanhuaWang committed 1 year ago

Verified 7b5b0660

Removal of cuda hardcoded string with get_device function (#5351)

raza-sikander committed 1 year ago

Verified f69f8840

Add custom reshaping for universal checkpoint (#5390)

tohtana committed 1 year ago

Verified 2c51aba0

Optimize the fp-dequantizer to get high memory-BW utilization (#5373)

RezaYazdaniAminabadi committed 1 year ago

Verified a8b82153

remove debug print

Masahiro Tanaka committed 1 year ago

8334b462

fix sorting of shard files

Masahiro Tanaka committed 1 year ago

c34cf7fe

Update README.md for intel XPU support (#5389)

Liangliang-Ma committed 1 year ago

Verified cc9e7b9c

Cleanup required_torch_version code and references. (#5370)

loadams committed 1 year ago

Verified 6dcced1d

Update checkout action on workflows on ubuntu 20.04 (#5387)

loadams committed 1 year ago

Verified 63029e8f

Revert "Update XPU workflow too"

loadams committed 1 year ago

62ab1810

Update XPU workflow too

loadams committed 1 year ago

eb216a83

Update path name on xpu-max1100.yml, add badge in README (#5386)

loadams committed 1 year ago

Verified 7e126d20

Add CI for Intel XPU/Max1100 (#5376)

Liangliang-Ma committed 1 year ago

Verified 87373889

Older