Frank Lee
725a39f4bd
update github CI with the current workflow ( #441 )
3 years ago
Frank Lee
5a1e33b97f
update contributing.md with the current workflow ( #440 )
3 years ago
Jiarui Fang
17b8274f8a
[unitest] polish zero config in unittest ( #438 )
3 years ago
Jiarui Fang
640a6cd304
[refactory] refactory the initialize method for new zero design ( #431 )
3 years ago
Frank Lee
4f85b687cf
[misc] replace codebeat with codefactor on readme ( #436 )
3 years ago
Frank Lee
bffd85bf34
added testing module ( #435 )
3 years ago
HELSON
dbdc9a7783
added Multiply Jitter and capacity factor eval for MOE ( #434 )
3 years ago
Frank Lee
b03b3ae99c
fixed mem monitor device ( #433 )
...
fixed mem monitor device
3 years ago
Frank Lee
14a7094243
fixed fp16 optimizer none grad bug ( #432 )
3 years ago
ver217
fce9432f08
sync before creating empty grad
3 years ago
ver217
ea6905a898
free param.grad
3 years ago
ver217
9506a8beb2
use double buffer to handle grad
3 years ago
Frank Lee
0f5f5dd556
fixed gpt attention mask in pipeline ( #430 )
3 years ago
Jiarui Fang
f9c762df85
[test] merge zero optim tests ( #428 )
3 years ago
Frank Lee
f0d6e2208b
[polish] add license meta to setup.py ( #427 )
3 years ago
Jiarui Fang
5d7dc3525b
[hotfix] run cpu adam unittest in pytest ( #424 )
3 years ago
Jiarui Fang
54229cd33e
[log] better logging display with rich ( #426 )
...
* better logger using rich
* remove deepspeed in zero requirements
3 years ago
HELSON
3f70a2b12f
removed noisy function during evaluation of MoE router ( #419 )
3 years ago
Jiarui Fang
adebb3e041
[zero] cuda margin space for OS ( #418 )
3 years ago
Jiarui Fang
56bb412e72
[polish] use GLOBAL_MODEL_DATA_TRACER ( #417 )
3 years ago
Jiarui Fang
23ba3fc450
[zero] refactory ShardedOptimV2 init method ( #416 )
3 years ago
Frank Lee
e79ea44247
[fp16] refactored fp16 optimizer ( #392 )
3 years ago
Frank Lee
f8a0e7fb01
Merge pull request #412 from hpcaitech/develop
...
merge develop to main
3 years ago
Jiarui Fang
21dc54e019
[zero] memtracer to record cuda memory usage of model data and overall system ( #395 )
3 years ago
Jiarui Fang
a37bf1bc42
[hotfix] rm test_tensor_detector.py ( #413 )
3 years ago
Jiarui Fang
370f567e7d
[zero] new interface for ShardedOptimv2 ( #406 )
3 years ago
LuGY
a9c27be42e
Added tensor detector ( #393 )
...
* Added tensor detector
* Added the - states
* Allowed change include_cpu when detect()
3 years ago
Frank Lee
32296cf462
Merge pull request #409 from 1SAA/develop
...
[hotfix] fixed error when no collective communication in CommProfiler
3 years ago
1SAA
907ac4a2dc
fixed error when no collective communication in CommProfiler
3 years ago
Frank Lee
62b08acc72
update hf badge link ( #410 )
3 years ago
Frank Lee
2fe68b359a
Merge pull request #403 from ver217/feature/shard-strategy
...
[zero] Add bucket tensor shard strategy
3 years ago
Frank Lee
cf92a779dc
added huggingface badge ( #407 )
3 years ago
HELSON
dfd0363f68
polished output format for communication profiler and pcie profiler ( #404 )
...
fixed typing error
3 years ago
ver217
63469c0f91
polish code
3 years ago
ver217
54fd37f0e0
polish unit test
3 years ago
ver217
88804aee49
add bucket tensor shard strategy
3 years ago
Frank Lee
aaead33cfe
Merge pull request #397 from hpcaitech/create-pull-request/patch-sync-submodule
...
[Bot] Synchronize Submodule References
3 years ago
github-actions
6098bc4cce
Automated submodule synchronization
3 years ago
Frank Lee
6937f85004
Merge pull request #402 from oikosohn/oikosohn-patch-1
...
fix typo in CHANGE_LOG.md
3 years ago
sohn
ff4f5d7231
fix typo in CHANGE_LOG.md
...
- fix typo, `Unifed` -> `Unified` below Added
3 years ago
Frank Lee
fc5101f24c
Merge pull request #401 from hpcaitech/develop
3 years ago
Frank Lee
fc2fd0abe5
Merge pull request #400 from hpcaitech/hotfix/readme
...
fixed broken badge link
3 years ago
Frank Lee
6d3a4f51bf
fixed broken badge link
3 years ago
HELSON
7c079d9c33
[hotfix] fixed bugs in ShardStrategy and PcieProfiler ( #394 )
3 years ago
Frank Lee
1e4bf85cdb
fixed bug in activation checkpointing test ( #387 )
3 years ago
Jiarui Fang
3af13a2c3e
[zero] polish ShardedOptimV2 unittest ( #385 )
...
* place params on cpu after zero init context
* polish code
* bucketzed cpu gpu tensor transter
* find a bug in sharded optim unittest
* add offload unittest for ShardedOptimV2.
* polish code and make it more robust
3 years ago
binmakeswell
ce7b2c9ae3
update README and images path ( #384 )
3 years ago
ScalableEKNN
2fcd4f38ee
fix format ( #379 )
3 years ago
Jiang Zhuo
5a4a3b77d9
fix format ( #376 )
3 years ago
lucasliunju
ce886a9062
fix format ( #374 )
3 years ago