Commits · faa23b9d9abe9a547423aa1e2b408b78b2df800c · OpenDAS / ColossalAI

"docs/source/zh-Hans/vscode:/vscode.git/clone" did not exist on "8264cd7ef11d1b40d1fc6069bd8c1b20dc78bd08"

14 Sep, 2022 3 commits

[autoparallel] add reshape handler (#1594) · faa23b9d
YuliangLiu0306 authored Sep 14, 2022
```
* [autoparallel] add reshape handler

* polish code
```
faa23b9d
Automated submodule synchronization (#1596) · c938dda0
github-actions[bot] authored Sep 14, 2022
```
Co-authored-by: github-actions <github-actions@github.com>
```
c938dda0

[fx] provide an accurate estimation of memory. (#1587) · 5c494d45

Super Daniel authored Sep 14, 2022

* [fx] add some comment and docstrings.

* [fx] add dataflow analysis for an autograd graph.

* add intepretation for graph analysis.

* [fx] before doing save_tensor_hooks.

* [fx] provide an accurate estimation of memory except for GPT-2.

* [fx] provide an accurate estimation of memory except for GPT-2.

* [fx] provide an accurate estimation of memory except for GPT-2.

* [fx] a very accurate version on GPT-2.

* [fx] refactor code.

* [fx] remove redundant inplace=True.

* [fx] refactor code.

* [fx] refactor code.

* [fx] refactor code.

* [fx] dive into backward memory.

5c494d45

13 Sep, 2022 8 commits
- [autoparallel] refactored shape consistency to remove redundancy (#1591) · 27fe8af6
  Frank Lee authored Sep 13, 2022
```
* [autoparallel] refactored shape consistency to remove redundancy

* polish code

* polish code

* polish code
```
  27fe8af6
- [autoparallel] add resnet autoparallel unit test and add backward weight communication cost (#1589) · d164449d
  YuliangLiu0306 authored Sep 13, 2022
  
  d164449d
- [autoparallel] added generate_sharding_spec to utils (#1590) · 7c18a588
  Frank Lee authored Sep 13, 2022
  
  7c18a588
- [fx] Improve linearize and rotor solver (#1586) · 49ccf8b5
  Boyuan Yao authored Sep 13, 2022
```
* [fx] add nested activation_checkpoint codegen

* undo algorithms commits

* solver

* undo some commits

* [fx] torch11 add nested activation checkpoint codegen

* remove some imports

* [fx] add some comments in activation codegen

* [fx] codegen instance error fix

* [fx] imporve linearize and rotor solver

* [fx] some comments and format modification
```
  49ccf8b5
- [autoparallel] added solver option dataclass (#1588) · 219f66c5
  Frank Lee authored Sep 13, 2022
  
  219f66c5
- [autoparallel] adapt solver with resnet (#1583) · 82d4376c
  YuliangLiu0306 authored Sep 13, 2022
```
* [autoparallel]adapt solver with resnet

* polish code

* polish code
```
  82d4376c
- [embeddings] add already_split_along_rank flag for tablewise mode (#1584) · f3403ff9
  CsRic authored Sep 13, 2022
  
  f3403ff9
- Automated submodule synchronization (#1550) · 77399dc9
  github-actions[bot] authored Sep 13, 2022
```
Co-authored-by: github-actions <github-actions@github.com>
```
  77399dc9
12 Sep, 2022 1 commit

[fx] Add nested checkpoint in activation checkpoint codegen (#1585) · f3687e4e

Boyuan Yao authored Sep 12, 2022

* [fx] add nested activation_checkpoint codegen

* undo algorithms commits

* solver

* undo some commits

* [fx] torch11 add nested activation checkpoint codegen

* remove some imports

* [fx] add some comments in activation codegen

* [fx] codegen instance error fix

f3687e4e

09 Sep, 2022 1 commit
- [NFC] add OPT serving (#1581) · 1c9ec327
  binmakeswell authored Sep 09, 2022
  
  1c9ec327
08 Sep, 2022 25 commits
- [NFC] polish ./colossalai/trainer/hooks/_lr_scheduler_hook.py code style (#1576) · 20e46652
  Boyuan Yao authored Sep 08, 2022
  
  20e46652
- [NFC] polish colossalai/zero/sharded_model/reduce_scatter.py code style (#1554) · 06dccdde
  Fazzie-Maqianli authored Sep 08, 2022
  
  06dccdde
- [NFC] polish utils/tensor_detector/__init__.py code style (#1573) · 2ac46f7b
  CsRic authored Sep 08, 2022
```
Co-authored-by: ric <mkkt_bkkt@mail.ustc.edu.cn>
```
  2ac46f7b
- [NFC] polish colossalai/nn/lr_scheduler/multistep.py code style (#1572) · 2144cbae
  Sze-qq authored Sep 08, 2022
  
  2144cbae
- [NFC] polish colossalai/nn/lr_scheduler/torch.py code style (#1571) · e4bf7ae6
  superhao1995 authored Sep 08, 2022
```
Co-authored-by: Research <research@soccf-snr3-017.comp.nus.edu.sg>
```
  e4bf7ae6
- [NFC] polish colossalai/nn/parallel/data_parallel.py code style (#1570) · 3263cdf5
  Jiatong Han authored Sep 08, 2022
```
Co-authored-by: JThh <jiatong.han@u.nus.edu>
```
  3263cdf5
- [NFC] polish colossalai/pipeline/utils.py code style (#1562) · f566c9b9
  Zirui Zhu authored Sep 08, 2022
  
  f566c9b9
- [NFC] polish colossalai/fx/tracer/meta_patch/patched_module/convolution.py code style (#1563) · e070ca45
  Xue Fuzhao authored Sep 08, 2022
  
  e070ca45
- [NFC] polish colossalai/gemini/update/chunkv2.py code style (#1565) · 9823cbf2
  Zangwei Zheng authored Sep 08, 2022
  
  9823cbf2
- [NFC] polish colossalai/nn/layer/colossalai_layer/dropout.py code style (#1568) · f586887a
  DouJS authored Sep 08, 2022
  
  f586887a
- [NFC] polish colossalai/utils/tensor_detector/tensor_detector.py code style (#1566) · c7d49329
  LuGY authored Sep 08, 2022
  
  c7d49329
- [NFC] polish colossalai/nn/_ops/embedding.py code style (#1561) · 0c4c9aa6
  BigOneLiXiaoMing authored Sep 08, 2022
  
  0c4c9aa6
- [NFC] polish colossalai/builder/__init__.py code style (#1560) · 08815f0e
  Ziheng Qin authored Sep 08, 2022
```
Co-authored-by: henryqin1997 <henryqin1997@gamil.com>
```
  08815f0e
- [NFC] polish colossalai/testing/comparison.py code style. (#1558) · 83289173
  Super Daniel authored Sep 08, 2022
  
  83289173
- [NFC] polish colossalai/nn/layer/colossalai_layer/linear.py (#1556) · 7cc052f6
  Ofey Chan authored Sep 08, 2022
  
  7cc052f6
- [NFC] polish code colossalai/gemini/update/search_utils.py (#1557) · 46931e3c
  Kai Wang (Victor Kai) authored Sep 08, 2022
  
  46931e3c
- [NFC] polish colossalai/nn/_ops/layernorm.py code style (#1555) · 413f9c19
  yuxuan-lou authored Sep 08, 2022
  
  413f9c19
- [NFC] polish colossalai/nn/loss/loss_2p5d.py code style (#1553) · 8edb777c
  shenggan authored Sep 08, 2022
  
  8edb777c
- [NFC] polish colossalai/nn/_ops/embedding_bag.py code style (#1552) · bd2d7898
  Maruyama_Aya authored Sep 08, 2022
  
  bd2d7898
- [NFC] polish colossalai/nn/lr_scheduler/cosine.py code style · 73e9eb13
  binmakeswell authored Sep 08, 2022
  
  73e9eb13
- [NFC] polish colossalai/utils/multi_tensor_apply/multi_tensor_apply.py code style (#1559) · 318fbf11
  Kirigaya Kazuto authored Sep 08, 2022
  
  318fbf11
- update version (#1574) · b0f4c0bd
  ver217 authored Sep 08, 2022
  
  b0f4c0bd
- [embedding] cache_embedding small improvement (#1564) · a389ac4e
  CsRic authored Sep 08, 2022
  
  a389ac4e
- add gather_output for VocabParallelClassifier1D (#1569) · 10dd8226
  ver217 authored Sep 08, 2022
  
  10dd8226
- [NFC] polish test component gpt code style (#1567) · e615cfc3
  アマデウス authored Sep 08, 2022
  
  e615cfc3
07 Sep, 2022 2 commits

[pipeline/tuning] improve dispatch performance both time and space cost (#1544) · 6159d454
Kirigaya Kazuto authored Sep 07, 2022

6159d454

[fx] provide a stable but not accurate enough version of profiler. (#1547) · 4f596932

Super Daniel authored Sep 07, 2022

* [fx] compute memory stat and flop count for MetaInfoProp.

* [fx] modify node attribute.

* [fx] modify ckpt_chen.

* [fx] fix compatibility.

* [fx] fix import error.

* [fx] skip test for MetaInfoProp.

* [fx] skip test for MetaInfoProp.

* [fx] skip test for MetaInfoProp.

* [fx] skip test for MetaInfoProp.

* [fx] skip if torch 1.11.0.

* [fx] recover MetaInfoProp support for PyTorch 1.11.

* [fx] provide a stable but not accurate enough version of profiler.

* [fx] provide a stable but not accurate enough version of profiler.

* [fx] fix compatibility in tests.

* [fx] fix compatibility in tests.

* [fx] fix compatibility in tests.

* [fx] fix compatibility in tests.

* [fx] fix compatibility in tests.

* [fx] fix compatibility in tests.

* [fx] fix compatibility in tests.

* [fx] fix compatibility in tests.

* [fx] fix compatibility in tests.

* [fx] fix compatibility in tests.

* [fx] fix import error.

4f596932