microsoft/DeepSpeed

Pull Requests Commits

Merge branch 'master' into tohtana/consistent_zero_grad

tohtana committed 1 year ago

Verified 6e05d2c6

[XPU] Support DeepNVMe new code structure (#6532)

Liangliang-Ma committed 1 year ago

Verified d45cfd34

fix errors when setting zero3 leaf modules with torch.compile (#6564)

NirSonnenschein committed 1 year ago

Verified ba58682a

Fix gradient accumulation for Z2+offload (#6550)

tohtana committed 1 year ago

Verified c85c8703

[Accelerator] Cambricon MLU support (#6472)

Andy666G committed 1 year ago

Verified 0fbe96a5

DeepNVMe perf tuning (#6560)

tjruwase committed 1 year ago

Verified a5400974

Use msgpack for p2p comm (#6547)

tohtana committed 1 year ago

Verified 7622cd9e

Skip failing newly added tests in accelerate (#6574)

loadams committed 1 year ago

Verified 61de0171

fix for non-zero optimizer

Masahiro Tanaka committed 1 year ago

958dfc16

remove type hint

Masahiro Tanaka committed 1 year ago

4a59a2b4

remove type hint

Masahiro Tanaka committed 1 year ago

86da91da

remove type hint for python 3.7

Masahiro Tanaka committed 1 year ago

b1701960

fix zero_grad for z1/2 optimizer

Masahiro Tanaka committed 1 year ago

fea4811e

Merge branch 'tohtana/fix_grad_acc_z2_offload' into tohtana/consistent_zero_grad

Masahiro Tanaka committed 1 year ago

1b095d4c

improve consistency of zero_grad

Masahiro Tanaka committed 1 year ago

84ca923c

Merge branch 'master' into tohtana/fix_grad_acc_z2_offload

tjruwase committed 1 year ago

Verified 3a37b030

fix gradient accumulation for z2+offload

Masahiro Tanaka committed 1 year ago

18489835

Added Intel Gaudi to Accelerator Setup Guide (#6543)

ShifaAbu committed 1 year ago

Verified 2a56f533

Add conditional on torch version for scaled_dot_product_attention (#6517)

loadams committed 1 year ago

Verified 170b46e8

Avoid security issues of subprocess shell (#6498)

tjruwase committed 1 year ago

Verified 659f6be1

wrap include cuda_bf16.h with ifdef BF16_AVAILABLE (#6520)

oelayan7 committed 1 year ago

Verified c2748393

Revert "BF16 optimizer: Clear lp grads after updating hp grads in hook" (#6508)

nelyahu committed 1 year ago

Verified 8fa6b50b

fix environment variable export bug for MultiNodeRunner (#5878)

TideDra committed 1 year ago

Verified fc22d960

Fix the broken url link (#6500)

rogerxfeng8 committed 1 year ago

Verified 2a647c51

fix pipeline eval_batch micro_batches argument for schedule (#6484)

nelyahu committed 1 year ago

Verified 3b09d945

Op_builder->is_compatible quite warning (#6093)

terry-for-github committed 1 year ago

Verified 4f803852

HPU: add required ENV vars to acccelerator init (#6495)

nelyahu committed 1 year ago

Verified 857780a8

Update version.txt after 0.15.1 release (#6493)

loadams committed 1 year ago

Verified c210e601

Handle an edge case where `CUDA_HOME` is not defined on ROCm systems (#6488)

amorehead committed 1 year ago

Verified 10ba3dde

Safe usage of popen (#6490)

tjruwase committed 1 year ago

Verified 662a421b

Older