mirror of
https://github.com/modelscope/modelscope.git
synced 2026-02-24 04:01:10 +01:00
support vision efficient tuning finetune
## 查看改动点 ↓↓↓
### vision efficient tuning finetune
- Model模块改造成适配训练的
- Model模块在支持训练同时向下兼容之前发布的modecard
- Pipline兼容modelcard加载的preprocessor或直接定义的
- 添加 ImageClassificationPreprocessor (非mmcv版本)
- 添加 VisionEfficientTuningTrainer
- ~~添加 opencv_transforms==0.0.6~~ (以源代码引入必要)
### Modelcard
- test pipeline和trainer合并到一起
- 新增3个模型的test
- 新增demo service
### 公共组件
- ms_dataset.py: fix warning, [UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or xxx]
- preprocessor添加common:ToNumpy、Rename、Identity
- preprocessor common对于dict进行key判断再取值。
- ~~修复learning rate在iter级别变化的逻辑。~~ (本次不做了)
- ~~修复非dist状态下train data没有进行shuffle的bug。~~ (Master已有人改了)
- 修复训练时调用util中非cv包的异常 zhconv。
### 其他
- 为防止新引入的preprocessor模块在config中被原代码加载,导致在其他人做CI时会报错;所以暂时没有添加新的tag,等CR完成后,会进行打tag再rerun CI。
Link: https://code.alibaba-inc.com/Ali-MaaS/MaaS-lib/codereview/11762108
* support vision efficient tuning finetune
* update test case
* update shuffle on IterableDataset
* update bitfit & sidetuning
* compatible with base trainer
This commit is contained in:
committed by
wenmeng.zwm
parent
8298a3c31d
commit
bf3a2b6c09
3
data/test/images/vision_efficient_tuning_test_apple.jpg
Normal file
3
data/test/images/vision_efficient_tuning_test_apple.jpg
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:407d70db9f01bc7a6f34377e36c3f2f5eefdfca8bd3c578226bf5b31b73325dc
|
||||
size 127213
|
||||
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:9c67733db75dc7fd773561a5091329fd5ee919b2268a3a65718261722607698f
|
||||
size 226882
|
||||
Reference in New Issue
Block a user