microsoft/DeepSpeed

Pull Requests Commits

Merge branch 'master' into staging-deepnvme-gds-v1

jomayeri committed 1 year ago

Verified b746c41f

Enable dynamic shapes for pipeline parallel engine inputs (#5481)

tohtana committed 1 year ago

Verified 1ab1928d

Merge branch 'master' into staging-deepnvme-gds-v1

tjruwase committed 1 year ago

Verified a179e2be

Usability improvements

tjruwase committed 1 year ago

9a36d8be

moving torch check to aio only

Ubuntu committed 1 year ago

a75010b1

cpp std based on torch version

Ubuntu committed 1 year ago

3b400423

loadams committed 1 year ago

4833c03a

Merge branch 'master' into staging-deepnvme-gds-v1

loadams committed 1 year ago

Verified 874c1f4d

Move inf_or_nan_tracker to cpu for cpu offload (#5826)

BacharL committed 1 year ago

Verified 4d4ff0ed

Merge branch 'staging-deepnvme-gds-v1' of github.com:microsoft/DeepSpeed into staging-deepnvme-gds-v1

Ubuntu committed 1 year ago

a89ac6cb

updating sweep script with gds

Ubuntu committed 1 year ago

5043f43a

Merge branch 'master' into staging-deepnvme-gds-v1

tjruwase committed 1 year ago

Verified 60194ad1

Merge branch 'master' into staging-deepnvme-gds-v1

tjruwase committed 1 year ago

Verified 4c3b7282

tjruwase committed 1 year ago

ccc7d183

Merge branch 'master' into staging-deepnvme-gds-v1

tjruwase committed 1 year ago

Verified 1f87f1be

Merge branch 'master' into staging-deepnvme-gds-v1

tjruwase committed 1 year ago

Verified c79d7025

Update version.txt after 0.14.5 release (#5982)

loadams committed 1 year ago

Verified 3f6df9a2

Merge branch 'master' into staging-deepnvme-gds-v1

tjruwase committed 1 year ago

Verified c17a47cc

add moe topk(k>2) gate support (#5881)

inkcherry committed 1 year ago

Verified 9a3ede70

Merge branch 'staging-deepnvme-gds-v1' of github.com:microsoft/DeepSpeed into staging-deepnvme-gds-v1

Ubuntu committed 1 year ago

7ff5c45f

move pynvml install to setup.py (#5840)

Rohan138 committed 1 year ago

Verified 30428d03

Ubuntu committed 1 year ago

01f74571

Create abstract base class

tjruwase committed 1 year ago

d0a222cd

Merge branch 'staging-deepnvme-gds-v1' of github.com:microsoft/DeepSpeed into staging-deepnvme-gds-v1

tjruwase committed 1 year ago

606938e7

Avoid passing class member num_threads

tjruwase committed 1 year ago

6c464031

Allow accelerator to instantiate the device (#5255)

nelyahu committed 1 year ago

Verified eb07d41f

Update xpu-max1100.yml with new config and add some tests (#5668)

Liangliang-Ma committed 1 year ago

Verified 4ba49dda

Z3: optimizations for grad norm calculation and gradient clipping (#5504)

nelyahu committed 1 year ago

Verified 6eed634e

Add accelerator setup guides (#5827)

delock committed 1 year ago

Verified 19b01e1d

HPUAccelerator: remove support in set_visible_devices_envs (#5929)

nelyahu committed 1 year ago

Verified a8d1b44d

Older