huggingface/accelerate

Pull Requests Commits

device type helper (#3843)

kashif committed 53 days ago

Verified e3655b93

Deepspeed Ulysses/ALST integration (#3817)

stas00 committed 54 days ago

Verified c7e59dd7

use self hosted runner (#3841)

SunMarc committed 54 days ago

Verified e8f241f8

[Bug] Update torch.optim.Optimizer parameter states after tensor parallelism (#3835)

naomili0924 committed 55 days ago

Verified 00b1b181

Fix typo in broadcast_object_list docstring (#3823)

wsntxxn committed 61 days ago

Verified a73fd3a1

Change arxiv to hf papers (#3834)

qgallouedec committed 64 days ago

Verified a08fb323

update bnb quantisation 4bit flag (#3828)

hbraith committed 70 days ago

Verified 61bcdaa4

Remove warning for `cpu_ram_efficient_loading` (#3816)

SunMarc committed 83 days ago

Verified a12beee3

SunMarc committed 85 days ago

0a0c9895

refactor: nit change for get_parameters_from_modules (code debt) (#3815)

kmehant committed 91 days ago

Verified 5998f862

Fix tracking swanlab (#3810)

SunMarc committed 95 days ago

Verified f0313a64

Bump to python3.10 + update linter (#3809)

SunMarc committed 95 days ago

Verified df0c1870

SunMarc committed 97 days ago

Verified bc2478a4

fix (skip) cache flush when original device is `cpu` and offloaded to disk `meta` (#3796)

Qubitium committed 97 days ago

Verified 057edec2

Remove deprecated FindTiedParametersResult (#3786)

cyyever committed 116 days ago

Verified 14383311

Revert "fix: correct dictionary unpacking in recursively_apply function (#3766)" (#3787)

SunMarc committed 116 days ago

Verified a737437c

rm mlflow (#3783)

SunMarc committed 116 days ago

Verified 6997855a

Add optional typing (#3769)

cyyever committed 117 days ago

Verified 401075ff

refactor: Use `with` in Accelerator.autocast()instead of ` __enter__()` and` __exit__()` for more elegant style. (#3767)

EquationWalker committed 117 days ago

Verified 8031e24e

fix: correct dictionary unpacking in recursively_apply function (#3766)

qgallouedec committed 117 days ago

Verified 3db9fb69

switch XPU ccl backend to torch-builtin xccl in test_zero3_integration (#3773)

yao-matrix committed 117 days ago

Verified fe795fd3

Lower complexity of get_balanced_memory by adding a set (#3776)

SamuelBarryCS committed 118 days ago

Verified 409b356f

enable 2 model hook ut cases on XPU (#3774)

yao-matrix committed 119 days ago

Verified 1b50d939

Added Tip for better rendering (#3781)

sergiopaniego committed 120 days ago

Verified e79f3836

fix Muti node CUDA error: invalid device ordinal #3775 (#3779)

RicardoDominguez committed 122 days ago

Verified 0cb1a334

use reset_peak_memory_stats on xpu (#3772)

yao-matrix committed 123 days ago

Verified dfdc2190

fix FSDP2 test case failure on XPU (#3771)

yao-matrix committed 123 days ago

Verified 45959d7b

Fix: typo makes tests fail (#3765)

S1ro1 committed 126 days ago

Verified 8b493524

fix: torch_npu import error (#3764)

yanyongyu committed 126 days ago

Verified 9ead94e5

feat: allow mixed precision policy as dtype (#3751)

kmehant committed 127 days ago

Verified a0bc36e8

Newer Older