microsoft/DeepSpeed

Pull Requests Commits

Skip test_zero_tensor_fragment.py tests that timeout

tohtana committed 28 days ago

Verified 35ec1e2f

Skip test_zero_leaf_module.py tests that timeout

tohtana committed 28 days ago

Verified a6fb0cb1

Skip test_ds_initialize.py tests that timeout

tohtana committed 28 days ago

Verified 0fd0a10d

Skip onebit tests that timeout with pipeline config

tohtana committed 28 days ago

Verified 02d5ed73

Run tests in parallel with -n 4 instead of sequential

tohtana committed 28 days ago

Verified a3a9f8e9

Skip test_multiple_models.py that timeouts

tohtana committed 28 days ago

Verified 5af1f37e

Skip pipeline checkpoint tests that timeout in CI

tohtana committed 28 days ago

Verified 70d90c50

Skip zenflow tests that timeout in CI

tohtana committed 29 days ago

Verified e0e1cabb

Skip CPU adam tests that timeout in CI

tohtana committed 29 days ago

Verified 29eff21a

Skip pipeline parallelism tests that timeout in CI

tohtana committed 29 days ago

Verified b77f9c80

Fix ulysses PEFT test to use mpu object instead of global groups

tohtana committed 29 days ago

Verified 0e52d3a8

Skip more hanging ops tests in CI

tohtana committed 29 days ago

Verified 9195771f

Skip aio tests that hang in CI environment

tohtana committed 29 days ago

Verified fa213d08

Fix bf16 checkpoint optimizer state and muon test

tohtana committed 29 days ago

Verified e8275323

Merge: set TORCH_CUDA_ARCH_LIST=8.9 and use -n 1 for debugging

tohtana committed 29 days ago

Verified 96d36748

Test with -n 1 to debug parallel execution issues

tohtana committed 29 days ago

Verified 9b556a5c

add workflow to run full tests

tohtana committed 29 days ago

Verified 7ff51dca

fix: skip compressed allreduce for empty tensors (#7769)

T1mn committed 30 days ago

Verified 97e74301

Fix multiprocessing testcase (#7743)

k-artem committed 30 days ago

Verified 7bbb6c77

Update CLA to refer to DCO (#7778)

loadams committed 31 days ago

Verified e31daa8c

[Docs] Fix `docs/_pages/config-json.md` format (#7779)

ooooo-create committed 31 days ago

Verified 79af89be

Remove cron/PR triggers for outdated V100 tests (#7777)

loadams committed 32 days ago

Verified 53c87ce8

Add timeout to test workflows (#7774)

tohtana committed 33 days ago

Verified 52361686

Update README with newer status badges for CI

loadams committed 33 days ago

b80f453f

fix: update Megatron-DeepSpeed tutorial to match current repo structure (#7761)

nathon-lee committed 35 days ago

Verified 8a9369d0

fix: delete using namespace std. (#7766)

nathon-lee committed 36 days ago

Verified 816e4aef

Udpate workflow trigger (#7768)

tohtana committed 36 days ago

Verified d988f6ca

Various fixes to run on mps (#7767)

jeffra committed 36 days ago

Verified e5b93809

Update version.txt after 0.18.4 release (#7765)

loadams committed 38 days ago

Verified 6b878203

fix: avoid IndexError in BF16_Optimizer.destroy() when using DummyOptim (#7763)

nathon-lee committed 39 days ago

Verified b35d9eb0

Newer Older