Tree:
266851bbbb
master
13 Commits (266851bbbbd2dbeb5b681172936c5b013b5bc144)
13 Commits (266851bbbbd2dbeb5b681172936c5b013b5bc144)
| Author | SHA1 | Message | Date |
|---|---|---|---|
 yzhao
|
3f75fcdb79 | fix bug | 2 years ago |
|
yuze.zyz
|
605cd7f44a |
[to #42322933] NLP 1030 Refactor
Features:
1. Refactor the directory structure of nlp models. All model files are placed into either the model folder or the task_model folder
2. Refactor all the comments to google style
3. Add detail comments to important tasks and nlp models, to list the description of the model, and its preprocessor&trainer
4. Model Exporting now supports a direct all to TorchModelExporter(no need to derive from it)
5. Refactor model save_pretrained method to support direct running(independent from trainer)
6. Remove the judgement of Model in the pipeline base class, to support outer register models running in our pipelines
7. Nlp trainer now has a NLPTrainingArguments class , user can pass arguments into the dataclass, and use it as a normal cfg_modify_fn, to simplify the operation of modify cfg.
8. Merge the BACKBONES and the MODELS, so user can get a backbone with the Model.from_pretrained call
9. Model.from_pretrained now support a task argument, so user can use a backbone and load it with a specific task class.
10. Support Preprocessor.from_pretrained method
11. Add standard return classes to important nlp tasks, so some of the pipelines and the models are independent now, the return values of the models will always be tensors, and the pipelines will take care of the conversion to numpy and the following stuffs.
12. Split the file of the nlp preprocessors, to make the dir structure more clear.
Bugs Fixing:
1. Fix a bug that lr_scheduler can be called earlier than the optimizer's step
2. Fix a bug that the direct call of Pipelines (not from pipeline(xxx)) throws error
3. Fix a bug that the trainer will not call the correct TaskDataset class
4. Fix a bug that the internal loading of dataset will throws error in the trainer class
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10490585
|
2 years ago |
 zhangzhicheng.zzc
|
d721fabb34 |
[to #42322933]bert with sequence classification / token classification/ fill mask refactor
1.新增支持原始bert模型（非easynlp的 backbone prefix版本）
2.支持bert的在sequence classification/fill mask /token classification上的backbone head形式
3.统一了sequence classification几个任务的pipeline到一个类
4.fill mask 支持backbone head形式
5.token classification的几个子任务（ner，word seg， part of speech）的preprocessor 统一到了一起TokenClassificationPreprocessor
6. sequence classification的几个子任务（single classification， pair classification）的preprocessor 统一到了一起SequenceClassificationPreprocessor
7. 改动register中 cls的group_key 赋值位置，之前的group_key在多个decorators的情况下，会被覆盖，obj_cls的group_key信息不正确
8. 基于backbone head形式将 原本group_key和 module同名的情况尝试做调整，如下在modelscope/pipelines/nlp/sequence_classification_pipeline.py 中
原本
@PIPELINES.register_module(
Tasks.sentiment_classification, module_name=Pipelines.sentiment_classification)
改成
@PIPELINES.register_module(
Tasks.text_classification, module_name=Pipelines.sentiment_classification)
相应的configuration.json也有改动，这样的改动更符合任务和pipline（子任务）的关系。
8. 其他相应改动为支持上述功能
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10041463
|
3 years ago |
 Yingda Chen
|
e0ef60ca9b | [to #42322933] skip demo test by default | 3 years ago |
 lingcai.wl
|
7a49fa1cc6 |
[to #44657982] add unittest for demo and demotest utils
unittest for demo service
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10006180
|
3 years ago |
|
yuze.zyz
|
fbde374659 |
[to #42322933] add regress tests
Add regression test for some unit tests.
Firstly, Run a baseline test to create a pickle file which contains the inputs and outputs of modules, then changes can be observed between
the latest version and the baseline file.
Some baseline files are submitted in the data/test/regression folder
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9814693
|
3 years ago |
|
yingda.chen
|
45620dbc7f |
[to #42322933]clean up test level
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9662182 * clean up test level |
3 years ago |
|
yuze.zyz
|
21fa71baf0 |
[to #42322933] add/refactor nlp models source code and finetune
1. add sbert,veco,palm,space source code
2. support sbert sequence classification, token classification finetune
3. support veco sequence classification finetune
4. support palm nlg finetune
evaluation result: https://sheet.alibaba-inc.com/#/sheet/f7fdcc7f22bd5105 sheet:Maas
5. add ut for finetunes
6. add veco's taskdataset processor
7. add a common trainer for nlp, and a specific trainer for veco
8. merge some duplicate codes of models, preprocessors, pipelines
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9574105
* add basic class of hook&metrics
* pre-commit passed
* change some comments
* pre commit passed
* 1. remove accuracy's groups 2. remove useless hooks 3. simplify priorities
* pre-commit passed
* fix a comment
* Merge branch 'master' into finetune_hooks_metrics
# Conflicts:
# modelscope/metainfo.py
* pre-commit passed
* add basic class of hook&metrics
* pre-commit passed
* change some comments
* pre commit passed
* 1. remove accuracy's groups 2. remove useless hooks 3. simplify priorities
* pre-commit passed
* fix a comment
* Merge branch 'feat/finetune' of gitlab.alibaba-inc.com:Ali-MaaS/MaaS-lib into feat/finetune
* mv hooks related to modelscope/trainers/hooks
* mv priority back
* add torch mdoel base and test
* update hooks, trainer, import_util
* add torch epoch based trainer and dis utils
* add hooks
* fix warmup
* format code stype and fix warmup and add warmup unittest
* fix impls
* pre-commit check passed
* update hook and add EpochBasedTrainer
* add trainer unittest
* Merge branch 'feat/add_hooks' into feat/add_task
# Conflicts:
# modelscope/models/base_torch.py
# modelscope/trainers/hooks/hook.py
# modelscope/trainers/trainer.py
* update unittest name
* rewrite taskdataset to trainer
* fix trainer and add unittest
* add unittest
* code: run to forward
* run through... but ugly code
* arrange some cls
* fix some errs
* revert some mistakes
* init check in
* Merge branch 'feat/add_hooks' into feat/add_task
# Conflicts:
# modelscope/trainers/trainer.py
* test with bigger epoch and size
* add the default metrics class
* move build metrics code to a method
* merge add_task
* merge origin add_task
* add device initialization
* remove preprocessor arg for bool
* add task models
* move metric collect logic to metrics class
* pre-commit passed
* fix cr comments
* precommit passed
* add task models
* Merge remote-tracking branch 'origin/feat/add_task' into feat/backbone_head
* add comment
* change comment formats.
* fix comments
* fix ut bug
* fix comments
* add wrapper check
* fix comments
* pre commit passed
* fix cr comments
* solve a loop import problem
* fix ut bug
* fix ut errors
* change dummydataset to msdataset
* precommit passed
* merge add task
* backbone-head is build, model is not correctly loaded
* model load states matched
* result matched
* lint
* add veco/palm_v2 code
* merge master
* merge master success running
* add repr model name level
* Merge branch 'feat/veco_palm' into feat/finetune_sbert_veco
* model test for training
* add token-classification metric add formal ut
* fix running bug
* finetune and pipeline are working with backbone-head
* add nli
* add missing code
* finetune and pipeline are working with backbone-head
* Merge branch 'feat/backbone_head' of http://gitlab.alibaba-inc.com/Ali-MaaS/MaaS-lib into feat/backbone_head
* add a test repo for pr
* remove merge conflicted file
* remove merge conflicted file 1
* lint check
* import error
* none type bug fix
* forward input unpacking or dict bug
* move head into models, add build_backbone with registry, no base method
* merge master
* feat: 1. add interleave dataset method 2. support multiple dataset in trainer.build_dataset 3. support 3 sub tasks in sequence_classification task
* unfinished
* update the task model structure in NLP field
* merge master
* update by comments
* keep the default model id as current on production
* unfinished
* unfinished
* veco can run
* Merge remote-tracking branch 'origin/master' into feat/backbone_head
* add taskmodel for module management
* remove forward_input_is_dict
* unfinished
* token classification started
* update base model structure
* move space to backbone
* remove 'type' in build_from_cfg method
* test update
* bug fix
* on tesing, mess code
* Merge branch 'feat/backbone_head' into feat/refactor_nlp_730
# Conflicts:
# modelscope/metrics/builder.py
# modelscope/models/__init__.py
# modelscope/models/nlp/__init__.py
# modelscope/preprocessors/nlp.py
# modelscope/trainers/trainer.py
# requirements/multi-modal.txt
* add missing merge
* add sofa source code
* refactor
* add veco task dataset
* add veco task dataset
* pre-commit passed
* fix bug of log
* add some features
* merge master
* bug fix
* refine nlp models
* fix the training error
* unfinished
* refactor pipeline
* Merge branch 'feat/backbone_head' into feat/refactor_nlp_730
# Conflicts:
# modelscope/metrics/builder.py
# modelscope/models/nlp/__init__.py
# modelscope/models/nlp/backbones/structbert/modeling_sbert.py
# modelscope/models/nlp/palm_v2/palm_for_text_generation.py
# modelscope/preprocessors/base.py
# modelscope/preprocessors/nlp.py
# modelscope/trainers/trainer.py
* Merge commit 'ab04ceafc5453ce7daa9aa09e37a55f703072a10' into feat/refactor_nlp_730
# Conflicts:
# modelscope/metainfo.py
# modelscope/metrics/builder.py
# modelscope/models/__init__.py
# modelscope/models/base/base_torch_model.py
# modelscope/models/nlp/__init__.py
# modelscope/models/nlp/backbones/space/model/intent_unified_transformer.py
# modelscope/models/nlp/backbones/space/model/model_base.py
# modelscope/models/nlp/palm_v2/palm_for_text_generation.py
# modelscope/models/nlp/sbert_for_sequence_classification.py
# modelscope/models/nlp/sequence_classification.py
# modelscope/models/nlp/space/__init__.py
# modelscope/models/nlp/space_for_dialog_intent_prediction.py
# modelscope/models/nlp/space_for_dialog_modeling.py
# modelscope/models/nlp/space_for_dialog_state_tracking.py
# modelscope/models/nlp/task_model.py
# modelscope/pipelines/nlp/sentiment_classification_pipeline.py
# modelscope/preprocessors/base.py
# modelscope/preprocessors/nlp.py
# modelscope/trainers/trainer.py
* revert changes
* unify sentnece classification postprocess
* revert some changes, move some model files
* pipeline first case run through
* ws pipeline passed
* Merge branch 'feat/refactor_nlp_730' into feat/finetune_sbert_veco
* finetune
* revert code
* revert some code
* ws finetune started, only the accuracy is weird
* Merge branch 'feat/veco_taskdataset' into feat/finetune_sbert_veco
# Conflicts:
# modelscope/task_datasets/veco_dataset.py
# tests/taskdataset/test_veco_dataset.py
* veco+nli finetune started
* Merge branch 'master' into feat/finetune_sbert_veco
# Conflicts:
# modelscope/models/nlp/sbert_for_sequence_classification.py
# modelscope/models/nlp/sbert_for_token_classification.py
# modelscope/models/nlp/sbert_for_zero_shot_classification.py
# modelscope/models/nlp/space/space_for_dialog_intent_prediction.py
# modelscope/models/nlp/space/space_for_dialog_modeling.py
# modelscope/trainers/trainer.py
* add trainer for nlp
* trainer: dataset params passed into preprocessor
* test passed by nlptrainer
* fix some bugs
* fix some bugs
* add backbone/head subclass
* fix regression bugs
* fix bug in token-cls finetune
* support cfg modification
* fix bug
* fix bug
* update requirements
* add some comments and fix some t
* add some comments and revert a argument
* split to two test files
* revert code
* fixbug in precessor
(cherry picked from commit
|
3 years ago |
 wenmeng.zwm
|
d55525bfb6 | [to #43112771] requirements check and lazy import support | 3 years ago |
|
zhangzhicheng.zzc
|
cf194ef6cd |
[to #42322933] nlp preprocessor refactor
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9269314 * init * token to ids * add model * model forward ready * add intent * intent preprocessor ready * intent success * merge master * test with model hub * add flake8 * update * update * update * Merge branch 'master' into nlp/space/gen * delete file about gen * init * fix flake8 bug * [to #42322933] init * bug fix * [to #42322933] init * update pipeline registry info * Merge remote-tracking branch 'origin/master' into feat/nli * [to #42322933] init * [to #42322933] init * modify forward * [to #42322933] init * generation ready * init * Merge branch 'master' into feat/zero_shot_classification # Conflicts: # modelscope/preprocessors/__init__.py * [to #42322933] bugfix * [to #42322933] pre commit fix * fill mask * registry multi models on model and pipeline * add tests * test level >= 0 * local gen ready * merge with master * dialog modeling ready * fix comments: rename and refactor AliceMindMLM; adjust pipeline * space intent and modeling(generation) are ready * bug fix * add dep * add dep * support dst data processor * merge with nlp/space/dst * merge with master * Merge remote-tracking branch 'origin' into feat/fill_mask Conflicts: modelscope/models/nlp/__init__.py modelscope/pipelines/builder.py modelscope/pipelines/outputs.py modelscope/preprocessors/nlp.py requirements/nlp.txt * merge with master * merge with master 2/2 * fix comments * fix isort for pre-commit check * allow params pass to pipeline's __call__ method * Merge remote-tracking branch 'origin/master' into feat/zero_shot_classification * merge with nli task * merge with sentiment_classification * merge with zero_shot_classfication * merge with fill_mask * merge with space * merge with master head * Merge remote-tracking branch 'origin' into feat/fill_mask Conflicts: modelscope/utils/constant.py * fix: pipeline module_name from model_type to 'fill_mask' & fix merge bug * unfiinished change * fix bug * unfinished * unfinished * revise modelhub dependency * Merge branch 'feat/nlp_refactor' of http://gitlab.alibaba-inc.com/Ali-MaaS/MaaS-lib into feat/nlp_refactor * add eval() to pipeline call * add test level * ut run passed * add default args * tmp * merge master * all ut passed * remove an useless enum * revert a mis modification * revert a mis modification * Merge commit 'ace8af92465f7d772f035aebe98967726655f12c' into feat/nlp * commit 'ace8af92465f7d772f035aebe98967726655f12c': [to #42322933] Add cv-action-recongnition-pipeline to maas lib [to #42463204] support Pil.Image for image_captioning_pipeline [to #42670107] restore pydataset test [to #42322933] add create if not exist and add(back) create model example Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9130661 [to #41474818]fix: fix errors in task name definition # Conflicts: # modelscope/pipelines/builder.py # modelscope/utils/constant.py * Merge branch 'feat/nlp' into feat/nlp_refactor * feat/nlp: [to #42322933] Add cv-action-recongnition-pipeline to maas lib [to #42463204] support Pil.Image for image_captioning_pipeline [to #42670107] restore pydataset test [to #42322933] add create if not exist and add(back) create model example Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9130661 [to #41474818]fix: fix errors in task name definition # Conflicts: # modelscope/pipelines/builder.py * fix compile bug * refactor space * Merge branch 'feat/nlp_refactor' of http://gitlab.alibaba-inc.com/Ali-MaaS/MaaS-lib into feat/nlp_refactor * Merge remote-tracking branch 'origin' into feat/fill_mask * fix * pre-commit lint * lint file * lint file * lint file * update modelhub dependency * lint file * ignore dst_processor temporary * solve comment: 1. change MaskedLMModelBase to MaskedLanguageModelBase 2. remove a useless import * recommit * remove MaskedLanguageModel from __all__ * Merge commit '1a0d4af55a2eee69d89633874890f50eda8f8700' into feat/nlp_refactor * commit '1a0d4af55a2eee69d89633874890f50eda8f8700': [to #42322933] test level check Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9143809 [to #42322933] update nlp models name in metainfo Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9134657 # Conflicts: # modelscope/metainfo.py * update * revert pipeline params update * remove zeroshot * update sequence classfication outpus * merge with fill mask * Merge remote-tracking branch 'origin' into feat/fill_mask * fix * fix flake8 warning of dst * Merge remote-tracking branch 'origin/feat/fill_mask' into feat/nlp * merge with master * remove useless test.py * Merge remote-tracking branch 'origin/master' into feat/nlp * remove unformatted space trainer * revise based on comment except chinease comment * skip ci blocking * translation pipeline * csanmt model for translation pipeline * update * update * update builder.py * change Chinese notes of space3.0 into English * translate chinese comment to english * add space to metainfo * update casnmt_translation * update csanmt transformer * merge with master * update csanmt translation * update lint * update metainfo.py * Update translation_pipeline.py * Update builder.py * fix: 1. make csanmt derived from Model 2. add kwargs to prevent from call error * pre-commit check * temp exclue flake8 * temp ignore translation files * fix bug * pre-commit passed * fixbug * fixbug * revert pre commit ignorance * pre-commit passed * fix bug * merge with master * add missing setting * merge with master * add outputs * modify test level * modify chinese comment * remove useless doc * space outputs normalization * Merge remote-tracking branch 'origin/master' into nlp/translation * update translation_pipeline.py * Merge remote-tracking branch 'origin/master' into feat/nlp * Merge remote-tracking branch 'origin/master' into nlp/translation * add new __init__ method * add new __init__ method * update output format * Merge remote-tracking branch 'origin/master' into feat/nlp * update output format * merge with master * merge with nlp/translate * update the translation comment * update the translation comment * Merge branch 'nlp/translation' into feat/nlp * Merge remote-tracking branch 'origin/master' into feat/nlp * Merge remote-tracking branch 'origin/master' into feat/nlp * nlp preprocessor refactor * add get_model_type in util.hub * update the default preprocessor args * update the fill mask preprocessor * bug typo fixed |
3 years ago |
|
yingda.chen
|
1a0d4af55a |
[to #42322933] test level check
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9143809 |
3 years ago |
 mulin.lyh
|
76c6ff6329 |
[to #42675838]merge model hub code
合并model hub 代码
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9096493
|
3 years ago |
|
zhangzhicheng.zzc
|
ba471d4492 |
[to #42322933]sentence-similarity
Adding the new task of sentence_similarity, in which the model is the sofa version of structbert
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9016402
* sbert-sentence-similarity
* [to #42322933] pip8
* merge with master for file dirs update
* add test cases
* pre-commit lint check
* remove useless file
* download models again~
* skip time consuming test case
* update for pr reviews
* merge with master
* add test level
* reset test level to env level
* [to #42322933] init
* [to #42322933] init
* adding purge logic in test
* merge with head
* change test level
* using sequence classification processor for similarity
|
3 years ago |