microsoft/DeepSpeed

Pull Requests Commits

Update transformers ops unit tests to use `requried_torch_version` (#6884)

loadams committed 359 days ago

Verified 2f32966b

Fix --enable_each_rank_log when used with PDSH multi-node runner (#6863)

akeshet committed 359 days ago

Verified a964e435

Add MLP/lm_head tp grain size setting. (#6828)

Yejing-Lai committed 360 days ago

Verified da771ed4

Remove pin from transformers version and fix Processing/Threading issues in tests (#6822)

loadams committed 360 days ago

Verified 87c65068

Fix assertion for offloading states (#6855)

tohtana committed 360 days ago

Verified db98cc3a

Update real_accelerator.py (#6845)

keiwoo committed 363 days ago

Verified fc7c0700

Remove warnings from autodoc and sphinx (#6788)

loadams committed 363 days ago

Verified 6e3e13cb

Update TSC (#6867)

tjruwase committed 363 days ago

Verified 8efbcc49

add domino navigation (#6866)

GuanhuaWang committed 363 days ago

Verified b5e3fac6

Domino updates (#6861)

GuanhuaWang committed 363 days ago

Verified d7750c34

Fix xpu tests workflow failure by changing pip index url (#6864)

Liangliang-Ma committed 363 days ago

Verified 853a9764

Avoid poisoning process with CUDA calls as soon as importing (#6810)

HollowMan6 committed 364 days ago

Verified 91829476

Remove broken links to non-active site (#6854)

kaiksi-bb committed 1 year ago

Verified bd6fd50e

Fix nv-torch-nightly test by pinning transformers (#6849)

loadams committed 1 year ago

Verified 074d5c69

Fix uneven head sequence parallelism bug (#6774) (#6797)

Eugene29 committed 1 year ago

Verified ecb4bf32

Fix type error in `ZeROOrderedDict` (#6794)

oraluben committed 1 year ago

Verified 9e312528

Merge LoCo with Zero++ (#6730)

XingyuXie committed 1 year ago

Verified 1b58ba5e

Unpin pytest-subtests now that 0.14.1 is released (#6844)

loadams committed 1 year ago

Verified 06f1d360

Inference UTs check for trition support from accelerator (#6782)

raza-sikander committed 1 year ago

Verified 0c92c39d

Pin pytest-subtests version for accelerate tests (#6842)

loadams committed 1 year ago

Verified 08b907a2

Flops profiler support einops.einsum (#6755)

lvhoaa committed 1 year ago

Verified 9a41ccaf

Pin HPU tests (#6831)

loadams committed 1 year ago

Verified 9ca60160

Update version.txt after 0.16.1 release (#6826)

loadams committed 1 year ago

Verified a4499668

Update pre-commit version (#6821)

loadams committed 1 year ago

Verified 177832ed

Pin transformers version in cpu-torch-latest due to multiprocessing error. (#6823)

loadams committed 1 year ago

Verified 95ead2a0

Update README.md (#6825)

samadejacobs committed 1 year ago

Verified 2ea181f0

Update README.md (#6824)

samadejacobs committed 1 year ago

Verified 0e92f9b4

add FPDT tutorial (#6813)

samadejacobs committed 1 year ago

Verified 7b9fc8c7

Ulyssess offload blog (#6814)

samadejacobs committed 1 year ago

Verified 0b0fef3d

Pin transformers to avoid errors with latest version (#6820)

loadams committed 1 year ago

Verified b966e1f9

Newer Older