Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10491891

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10491951

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10453584

1. Fix bugs in daily test
2. Fix a bug that the updating of lr is before the first time of updating of optimizer
    TODO this will still cause warnings when GA is above 1
3. Remove the judgement of mode in text-classification's preprocessor to fit the base trainer(Bug)
     Update some regression bins to fit the preprocessor
4. Update the regression tool to let outer code modify atol and rtol
5. Add the default metric for text-classification task
6. Remove the useless ckpt conversion method in bert to avoid the requirement of tf when loading modeling_bert
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10430764

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10421808

    * add face 2d keypoints & human wholebody keypoint finrtune test case

# Conflicts:
#	modelscope/metrics/__init__.py

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10427797

添加2d手部关键点检测finetune功能
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10371710

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10275823

1. 调整face_detection的文件层级(scrfd与其余新增face_detection方法平级);
2. 增加极大脸/旋转脸的检测方法，更新了新模型；
3. 支持读入数据集并finetune和eval；
4. 新增card_detection模型，支持读入datasethub数据集并finetune
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10244540

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10338265

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10111615

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10198228

    * [to #44847108] add sparsity hook (pst algorithm)

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10227608

# Conflicts:
#	modelscope/models/multi_modal/ofa_for_all_tasks.py
#	modelscope/msdatasets/ms_dataset.py
#	modelscope/trainers/utils/inference.py

# Conflicts:
#	modelscope/preprocessors/multi_modal.py
#	modelscope/trainers/trainer.py
#	tests/pipelines/test_ofa_tasks.py

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10185634

    * fix dist training

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10189613

1. Fix a bug in trainer's progress bar
2. Fix a bug that trainer does not support dataset in config file
3. Add feature: support go on training via checkpoint file
4. Add feature: support fixed filename when saving best checkpoint
5. Fix a bug that no id2label in config file after finetune of nlp models
6. Fix some other bugs
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10138906

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10142442

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10103248

    * update EasyCV MSDataset

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9856179

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10061562

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9935664

    * adapt to msdataset for EasyCV

加入space模型在banking数据集上的微调代码
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10006792

1. Fix random seed for trainer and init it at the first line of init
2. Add a regress test for fixed training
3. Change the dataset 'dureader_robust_qg' to 'DuReader_robust-QG'
4. Change some datasets from loading hf.datasets to loading msdataset.load
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10029509

1. 修复 palm，gpt3，mplug 模型存在的 finetune 后保存 checkpoint 与原有 checkpoint key 字段存在区别无法使用 from_pretrained 导入的问题
2. 调整 test_finetune_mplug.py 为只保存训练结束时的 checkpoint，减少 ci 耗时
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10016517

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/10011826

、

1. Support saving the best checkpoint for inference
2. Fix a bug that _max_iters field does not exist in trainer
3. Fix a bug that function in lambda_lr field cannot be saved to file
4. Fix a bug that save_pretrained would not be called by iterating
5. Fix a bug that interval is not passed from BestCkptHook's init
        Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9972765

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9970626

    * fix trainer unittest

# Conflicts:
#	modelscope/preprocessors/multi_modal.py

Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/9872869

* dataset json file add args