Commit Graph

216 Commits (1a6d2a740be33d769111ed03104bb5fa73b2ad50)

Author SHA1 Message Date
HELSON a9b8300d54
[zero] improve adaptability for not-shard parameters (#708)
* adapt post grad hooks for not-shard parameters
* adapt optimizer for not-shard parameters
* offload gradients for not-replicated parameters
2022-04-11 13:38:51 +08:00
アマデウス 3fc8a204dc
[]Corrected 3d vocab parallel embedding (#707) 2022-04-11 10:17:55 +08:00
HELSON b31daed4cf
fix bugs in CPU adam (#633)
* add cpu adam counter for all cpu adam

* fixed updating error in adam kernel
2022-04-02 17:04:05 +08:00
Liang Bowen 828e465622
[hotfix] Raise messages for indivisible batch sizes with tensor parallelism (#622) 2022-04-02 16:12:04 +08:00
アマデウス 77ad24bf94
[model checkpoint] updated saving/loading for 3d layers (#597) 2022-04-01 16:52:47 +08:00
アマデウス 93089ed708
[model checkpoint] updated saving/loading for 2.5d layers (#596) 2022-04-01 16:52:33 +08:00
アマデウス c50bfb807b
[model checkpoint] updated saving/loading for 1d layers (#594) 2022-04-01 16:51:52 +08:00
アマデウス 7636d518e1
[model checkpoint] updated saving/loading for 2d layers (#595) 2022-04-01 16:50:34 +08:00
アマデウス cd13b63832
[model checkpoint] reworked unified layers for ease of save/load states (#593) 2022-04-01 16:49:56 +08:00
Ziyue Jiang 1c40ee8749
[TP] add assert for tp1d (#621) 2022-04-01 16:44:23 +08:00
ver217 e619a651fb
polish optimizer docstring (#619) 2022-04-01 16:27:03 +08:00
ver217 8432dc7080
polish moe docsrting (#618) 2022-04-01 16:15:36 +08:00
ver217 104cbbb313
[hotfix] add hybrid adam to __init__ (#584) 2022-03-31 19:08:34 +08:00
HELSON e6d50ec107
[zero] adapt zero for unsharded parameters (#561)
* support existing sharded and unsharded parameters in zero

* add unitest for moe-zero model init

* polish moe gradient handler
2022-03-31 18:34:11 +08:00
Wesley 46c9ba33da update code format 2022-03-31 17:15:08 +08:00
Wesley 666cfd094a fix parallel_input flag for Linear1D_Col gather_output 2022-03-31 17:15:08 +08:00
Liang Bowen 2c45efc398
html refactor (#555) 2022-03-31 11:36:56 +08:00
LuGY c44d797072
[docs] updatad docs of hybrid adam and cpu adam (#552) 2022-03-30 18:14:59 +08:00
Ziyue Jiang 763dc325f1
[TP] Add gather_out arg to Linear (#541) 2022-03-30 09:35:46 +08:00
HELSON 8c90d4df54
[zero] add zero context manager to change config during initialization (#546) 2022-03-29 17:57:59 +08:00
Liang Bowen ec5086c49c Refactored docstring to google style 2022-03-29 17:17:47 +08:00
LuGY 105c5301c3
[zero]added hybrid adam, removed loss scale in adam (#527)
* [zero]added hybrid adam, removed loss scale of adam

* remove useless code
2022-03-25 18:03:54 +08:00
LuGY 6a3f9fda83
[cuda] modify the fused adam, support hybrid of fp16 and fp32 (#497) 2022-03-25 14:15:53 +08:00
Jiarui Fang a445e118cf
[polish] polish singleton and global context (#500) 2022-03-23 18:03:39 +08:00
ver217 9ec1ce6ab1
[zero] sharded model support the reuse of fp16 shard (#495)
* sharded model supports reuse fp16 shard

* rename variable

* polish code

* polish code

* polish code
2022-03-23 14:59:59 +08:00
HELSON c9023d4078
[MOE] support PR-MOE (#488) 2022-03-22 16:48:22 +08:00
ver217 62b0a8d644
[zero] sharded optim support hybrid cpu adam (#486)
* sharded optim support hybrid cpu adam

* update unit test

* polish docstring
2022-03-22 14:56:59 +08:00
HELSON d7ea63992b
[MOE] add FP32LinearGate for MOE in NaiveAMP context (#480) 2022-03-22 10:50:20 +08:00
Jiarui Fang 65c0f380c2
[format] polish name format for MOE (#481) 2022-03-21 23:19:47 +08:00
HELSON 7544347145
[MOE] add unitest for MOE experts layout, gradient handler and kernel (#469) 2022-03-21 13:35:04 +08:00
HELSON aff9d354f7
[MOE] polish moe_env (#467) 2022-03-19 15:36:25 +08:00
HELSON bccbc15861
[MOE] changed parallelmode to dist process group (#460) 2022-03-19 13:46:29 +08:00
Jiarui Fang 0fcfb1e00d
[test] make zero engine test really work (#447) 2022-03-17 17:24:25 +08:00
Jiarui Fang 237d08e7ee
[zero] hybrid cpu adam (#445) 2022-03-17 15:05:41 +08:00
HELSON dbdc9a7783
added Multiply Jitter and capacity factor eval for MOE (#434) 2022-03-16 16:47:44 +08:00
HELSON 3f70a2b12f
removed noisy function during evaluation of MoE router (#419) 2022-03-15 12:06:09 +08:00
Jiang Zhuo 5a4a3b77d9 fix format (#376) 2022-03-11 15:50:28 +08:00
LuGY de46450461 Added activation offload (#331)
* Added activation offload

* Fixed the import bug, used the pytest
2022-03-11 15:50:28 +08:00
Kai Wang (Victor Kai) 53bb3bcc0a fix format (#362) 2022-03-11 15:50:28 +08:00
Yuer867 4a0f8c2c50 fix format parallel_2p5d (#357) 2022-03-11 15:50:28 +08:00
Liang Bowen 7eb87f516d flake8 style (#352) 2022-03-11 15:50:28 +08:00
xuqifan897 148207048e Qifan formated file ColossalAI\colossalai\nn\layer\parallel_1d\layers.py (#342) 2022-03-11 15:50:28 +08:00
DouJS cbb6436ff0 fix format for dir-[parallel_3d] (#333) 2022-03-11 15:50:28 +08:00
LuGY a3269de5c9 [zero] cpu adam kernel (#288)
* Added CPU Adam

* finished the cpu adam

* updated the license

* delete useless parameters, removed resnet

* modified the method off cpu adam unittest

* deleted some useless codes

* removed useless codes

Co-authored-by: ver217 <lhx0217@gmail.com>
Co-authored-by: Frank Lee <somerlee.9@gmail.com>
Co-authored-by: jiaruifang <fangjiarui123@gmail.com>
2022-03-11 15:50:28 +08:00
1SAA 82023779bb Added TPExpert for special situation 2022-03-11 15:50:28 +08:00
HELSON 36b8477228 Fixed parameter initialization in FFNExpert (#251) 2022-03-11 15:50:28 +08:00
アマデウス e13293bb4c fixed CI dataset directory; fixed import error of 2.5d accuracy (#255) 2022-03-11 15:50:28 +08:00
1SAA 219df6e685 Optimized MoE layer and fixed some bugs;
Decreased moe tests;

Added FFNExperts and ViTMoE model
2022-03-11 15:50:28 +08:00
zbian 3dba070580 fixed padding index issue for vocab parallel embedding layers; updated 3D linear to be compatible with examples in the tutorial 2022-03-11 15:50:28 +08:00
アマデウス 9ee197d0e9 moved env variables to global variables; (#215)
added branch context;
added vocab parallel layers;
moved split_batch from load_batch to tensor parallel embedding layers;
updated gpt model;
updated unit test cases;
fixed few collective communicator bugs
2022-02-15 11:31:13 +08:00
HELSON 0f8c7f9804
Fixed docstring in colossalai (#171) 2022-01-21 10:44:30 +08:00
Frank Lee e2089c5c15
adapted for sequence parallel (#163) 2022-01-20 13:44:51 +08:00
ver217 f68eddfb3d
refactor kernel (#142) 2022-01-13 16:47:17 +08:00
BoxiangW 4a3d3446b0
Update layer integration documentations (#108)
Update the documentations of layer integration

Update _log_hook.py

Update _operation.py
2022-01-10 18:05:58 +08:00
HELSON dceae85195
Added MoE parallel (#127) 2022-01-07 15:08:36 +08:00
ver217 7904baf6e1
fix layers/schedule for hybrid parallelization (#111) (#112) 2022-01-04 20:52:31 +08:00
ver217 96780e6ee4
Optimize pipeline schedule (#94)
* add pipeline shared module wrapper and update load batch

* added model parallel process group for amp and clip grad (#86)

* added model parallel process group for amp and clip grad

* update amp and clip with model parallel process group

* remove pipeline_prev/next group (#88)

* micro batch offload

* optimize pipeline gpu memory usage

* pipeline can receive tensor shape (#93)

* optimize pipeline gpu memory usage

* fix grad accumulation step counter

* rename classes and functions

Co-authored-by: Frank Lee <somerlee.9@gmail.com>
2021-12-30 15:56:46 +08:00
アマデウス 01a80cd86d
Hotfix/Colossalai layers (#92)
* optimized 1d layer apis; reorganized nn.layer modules; fixed tests

* fixed 2.5d runtime issue

* reworked split batch, now called in trainer.schedule.load_batch

Co-authored-by: BoxiangW <45734921+BoxiangW@users.noreply.github.com>
2021-12-29 23:32:10 +08:00
アマデウス 0fedef4f3c
Layer integration (#83)
* integrated parallel layers for ease of building models

* integrated 2.5d layers

* cleaned codes and unit tests

* added log metric by step hook; updated imagenet benchmark; fixed some bugs

* reworked initialization; cleaned codes

Co-authored-by: BoxiangW <45734921+BoxiangW@users.noreply.github.com>
2021-12-27 15:04:32 +08:00
HELSON 632e622de8
overlap computation and communication in 2d operations (#75) 2021-12-16 16:05:15 +08:00
Frank Lee 35813ed3c4
update examples and sphnix docs for the new api (#63) 2021-12-13 22:07:01 +08:00
Frank Lee da01c234e1
Develop/experiments (#59)
* Add gradient accumulation, fix lr scheduler

* fix FP16 optimizer and adapted torch amp with tensor parallel (#18)

* fixed bugs in compatibility between torch amp and tensor parallel and performed some minor fixes

* fixed trainer

* Revert "fixed trainer"

This reverts commit 2e0b0b7699.

* improved consistency between trainer, engine and schedule (#23)

Co-authored-by: 1SAA <c2h214748@gmail.com>

* Split conv2d, class token, positional embedding in 2d, Fix random number in ddp
Fix convergence in cifar10, Imagenet1000

* Integrate 1d tensor parallel in Colossal-AI (#39)

* fixed 1D and 2D convergence (#38)

* optimized 2D operations

* fixed 1D ViT convergence problem

* Feature/ddp (#49)

* remove redundancy func in setup (#19) (#20)

* use env to control the language of doc (#24) (#25)

* Support TP-compatible Torch AMP and Update trainer API (#27)

* Add gradient accumulation, fix lr scheduler

* fix FP16 optimizer and adapted torch amp with tensor parallel (#18)

* fixed bugs in compatibility between torch amp and tensor parallel and performed some minor fixes

* fixed trainer

* Revert "fixed trainer"

This reverts commit 2e0b0b7699.

* improved consistency between trainer, engine and schedule (#23)

Co-authored-by: 1SAA <c2h214748@gmail.com>

Co-authored-by: 1SAA <c2h214748@gmail.com>
Co-authored-by: ver217 <lhx0217@gmail.com>

* add an example of ViT-B/16 and remove w_norm clipping in LAMB (#29)

* add explanation for ViT example (#35) (#36)

* support torch ddp

* fix loss accumulation

* add log for ddp

* change seed

* modify timing hook

Co-authored-by: Frank Lee <somerlee.9@gmail.com>
Co-authored-by: 1SAA <c2h214748@gmail.com>
Co-authored-by: binmakeswell <binmakeswell@gmail.com>

* Feature/pipeline (#40)

* remove redundancy func in setup (#19) (#20)

* use env to control the language of doc (#24) (#25)

* Support TP-compatible Torch AMP and Update trainer API (#27)

* Add gradient accumulation, fix lr scheduler

* fix FP16 optimizer and adapted torch amp with tensor parallel (#18)

* fixed bugs in compatibility between torch amp and tensor parallel and performed some minor fixes

* fixed trainer

* Revert "fixed trainer"

This reverts commit 2e0b0b7699.

* improved consistency between trainer, engine and schedule (#23)

Co-authored-by: 1SAA <c2h214748@gmail.com>

Co-authored-by: 1SAA <c2h214748@gmail.com>
Co-authored-by: ver217 <lhx0217@gmail.com>

* add an example of ViT-B/16 and remove w_norm clipping in LAMB (#29)

* add explanation for ViT example (#35) (#36)

* optimize communication of pipeline parallel

* fix grad clip for pipeline

Co-authored-by: Frank Lee <somerlee.9@gmail.com>
Co-authored-by: 1SAA <c2h214748@gmail.com>
Co-authored-by: binmakeswell <binmakeswell@gmail.com>

* optimized 3d layer to fix slow computation ; tested imagenet performance with 3d; reworked lr_scheduler config definition; fixed launch args; fixed some printing issues; simplified apis of 3d layers (#51)

* Update 2.5d layer code to get a similar accuracy on imagenet-1k dataset

* update api for better usability (#58)

update api for better usability

Co-authored-by: 1SAA <c2h214748@gmail.com>
Co-authored-by: ver217 <lhx0217@gmail.com>
Co-authored-by: puck_WCR <46049915+WANG-CR@users.noreply.github.com>
Co-authored-by: binmakeswell <binmakeswell@gmail.com>
Co-authored-by: アマデウス <kurisusnowdeng@users.noreply.github.com>
Co-authored-by: BoxiangW <45734921+BoxiangW@users.noreply.github.com>
2021-12-09 15:08:29 +08:00
ver217 dbe62c67b8
add an example of ViT-B/16 and remove w_norm clipping in LAMB (#29) 2021-11-18 23:45:09 +08:00
Frank Lee 3defa32aee
Support TP-compatible Torch AMP and Update trainer API (#27)
* Add gradient accumulation, fix lr scheduler

* fix FP16 optimizer and adapted torch amp with tensor parallel (#18)

* fixed bugs in compatibility between torch amp and tensor parallel and performed some minor fixes

* fixed trainer

* Revert "fixed trainer"

This reverts commit 2e0b0b7699.

* improved consistency between trainer, engine and schedule (#23)

Co-authored-by: 1SAA <c2h214748@gmail.com>

Co-authored-by: 1SAA <c2h214748@gmail.com>
Co-authored-by: ver217 <lhx0217@gmail.com>
2021-11-18 19:45:06 +08:00
ver217 3c7604ba30 update documentation 2021-10-29 09:29:20 +08:00
zbian 404ecbdcc6 Migrated project 2021-10-28 18:21:23 +02:00