microsoft/DeepSpeed

Pull Requests Commits

Fix memory leak for pipelined optimizer swapper (#5700)

mauryaavinash95 committed 1 year ago

Verified db5a875b

Unit Test: Add error handling for rate limit exceeded in model list (#5715)

HeyangQin committed 1 year ago

Verified 83aa1843

Add Windows scripts (deepspeed, ds_report). (#5699)

costin-eseanu committed 1 year ago

Verified 74f3dcab

[INF] Enable torch compile for inference (#5612)

oelayan7 committed 1 year ago

Verified 7b1ea225

Update checkout action for nv-human-eval workflow (#5757)

loadams committed 1 year ago

Verified 2105976e

fix the missing argument in test and typo (#5730)

xylian86 committed 1 year ago

Verified 774b8977

Update node16 check on self-hosted runners and remove python 3.6 (#5756)

loadams committed 1 year ago

Verified 84118165

Fix tutorial links (#5714)

samadejacobs committed 1 year ago

Verified 3d347276

UCP Chinese Blog (#5713)

HeyangQin committed 1 year ago

Verified dd7a5be5

DeepSpeed Universal Checkpointing: Blog and Tutorial (#5711)

samadejacobs committed 1 year ago

Verified 121efdbd

update xpu fusedadam opbuilder for pytorch 2.3 (#5702)

baodii committed 1 year ago

Verified e3922967

Update XPU docker version (#5712)

loadams committed 1 year ago

Verified df58a784

Add additional paths to trigger xpu tests (#5707)

loadams committed 1 year ago

Verified aecfec7f

Change source of CPUAdam for xpu accelerator (#5703)

Liangliang-Ma committed 1 year ago

Verified 4b8a4a07

Add an argument to enable the injection of missing state during the conversion of universal checkpoints (#5608)

xylian86 committed 1 year ago

Verified f0e3f01d

Disable nvtx decorator to avoid graph break (#5697)

tohtana committed 1 year ago

Verified b421e8c8

ENV var added for recaching in INF Unit tests (#5688)

raza-sikander committed 1 year ago

Verified e9ffe028

inference unit test injectionPolicy split world_size to multiple tests (#5687)

oelayan7 committed 1 year ago

Verified 4f8531bc

Universal checkpoint for zero stage 3 (#5475)

xylian86 committed 1 year ago

Verified d2b1d7fc

[CPU] add fp16 support to shm inference_all_reduce (#5669)

delock committed 1 year ago

Verified 19da95f7

Fix memory leak from _hp_mapping (#5643)

chiragjn committed 1 year ago

Verified 88b2ef71

Bug fix for the "Link bit16 and fp32 parameters in partition" (#5681)

U-rara committed 1 year ago

Verified 224a05c3

Fixed Windows inference build. (#5609)

costin-eseanu committed 1 year ago

Verified b3767d01

Update version.txt after 0.14.4 release (#5694)

mrwyattii committed 1 year ago

Verified d89e8cdf

[XPU] support op builder from intel_extension_for_pytorch kernel path (#5425)

YizhouZ committed 1 year ago

Verified d254d75e

Update BUFSIZE to come from autotuner's constants.py, not numpy (#5686)

loadams committed 1 year ago

Verified 0c979d67

Fix numpy upgrade to 2.0.0 BUFSIZE import error (#5680)

Yejing-Lai committed 1 year ago

Verified 4000cee8

Fix latest pytorch '_get_socket_with_port' import error (#5654)

Yejing-Lai committed 1 year ago

Verified a4cd5504

enable yuan autotp & add conv tp (#5428)

Yejing-Lai committed 1 year ago

Verified 8ea995ee

Fixing the reshape bug in sequence parallel alltoall, which corrupted all QKV data (#5664)

YJHMITWEB committed 1 year ago

Verified 3bdd187e

Newer Older