microsoft/DeepSpeed

Pull Requests Commits

commit 2c47618c1a282f925446506d53108dc6e82d9ef0

loadams committed 1 year ago

00fccf53

Update to 9a94dfe1239ddfb8010a654aa1e677d56c01eee0

loadams committed 1 year ago

a0fd0666

loadams committed 1 year ago

b6d5d6e2

Update to midday

loadams committed 1 year ago

f0bd1125

loadams committed 1 year ago

c2ec60a3

Update midday 12/18

loadams committed 1 year ago

7a3c30bc

Add earlier comment

loadams committed 1 year ago

b8d0e981

loadams committed 1 year ago

610f7b24

loadams committed 1 year ago

253246a0

loadams committed 1 year ago

0aa45d0b

Update to EOD 12/18

loadams committed 1 year ago

41e461e8

Identify breaking commit 12/17

loadams committed 1 year ago

ac0fd0ec

Commit EOD 12/24

loadams committed 1 year ago

53df5200

Test with 4.47.1

loadams committed 1 year ago

53b3166d

Pin transformers on a6000 tests

loadams committed 1 year ago

cd228d3b

Fix windows blog examples (#6934)

loadams committed 1 year ago

Verified 53fb5795

Fix building on Windows with presence of Triton (#6749)

woct0rdho committed 1 year ago

Verified b62c84d8

Update python version classifiers (#6933)

loadams committed 1 year ago

Verified 6628127a

Use `torch.log1p` (#6930)

kit1980 committed 1 year ago

Verified c41b0c28

inference: remove unused _validate_args function (#5505)

nelyahu committed 1 year ago

Verified c7f30322

Check transformers version in BLOOM for inference v1 (#6766)

lekurile committed 1 year ago

Verified f2cc8090

Cleanup ops/transformer/inference tests (#6925)

loadams committed 1 year ago

Verified c348c5b1

Reduce the device bubble introduced by heavy loop synchronization in coalesced fetch/release(z3_leaf_module) (#6694)

inkcherry committed 1 year ago

Verified b0040b6c

Add fp8_gemm fallback for non-triton systems (#6916)

oelayan7 committed 1 year ago

Verified c5e48f49

[BUG FIX]:fix get torch.version.cuda error when cuda is None in rocm (#6909)

kairos-yu committed 1 year ago

Verified f8c9f314

Fix `checkpointable_layers` Logic (#6881)

Quentin-Anthony committed 1 year ago

Verified 0dbbb70b

Cleanup ops/transformer/inference tests (#6830)

loadams committed 1 year ago

Verified a8ede3a9

Stage3: Use new torch grad accumulation hooks API (#6773)

deepcharm committed 1 year ago

Verified 456c9ac6

Change compile for pipeline module torch.compile (#6478)

NirSonnenschein committed 1 year ago

Verified 3573858e

Update Gaudi2 jobs to latest 1.19 build (#6905)

raza-sikander committed 1 year ago

Verified cc03c76d

Older