glauclsdrgrading
Science Score: 44.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (3.6%) to scientific vocabulary
Repository
Basic Info
- Host: GitHub
- Owner: JieGenius
- License: apache-2.0
- Language: Python
- Default Branch: main
- Size: 13.3 MB
Statistics
- Stars: 4
- Watchers: 1
- Forks: 2
- Open Issues: 16
- Releases: 0
Metadata Files
README.md
介绍
GlauClsDRGrading 是 眼科大模型 的Lagent子项目。
其主要内容为训练一个青光眼分类和DR分级模型，使用多任务的方式进行训练。
其中，青光眼分类使用Refuge数据集，DR分级使用Kaggle数据集
环境安装
conda create -n GlauClsDRGrading python=3.10
conda activate GlauClsDRGrading
pip install -r requirements.txt
pip install wandb
pip install openmim && mim install -e .
wandb login
数据准备
准备Refuge数据集
准备kaggle数据集
```bash pip install kaggle vim ~/.kaggle/kaggle.json # 设置key kaggle competitions download -c aptos2019-blindness-detection
unzip aptos2019-blindness-detection.zip
head -n 300 train.csv > val.csv tail -n 299 val.csv | cut -d "," -f 1 | xargs -i mv trainimages/{}.png valimages/
vim train.csv # 删除第1到第300行，因为这部分被划分到验证集了。 ```
生成多任务所需的数据集
bash
cd ~/GlauClsDRGrading
export PYTHONPATH="$(which python):$(pwd)"
python tools/dataset_converters/refuge_aptos.py
训练
bash
python tools/train.py confis/convnext/convnext-tiny_b32_refuge_aptos.py --amp --auto-scale-lr
部署
由于使用了多任务输出，mmdeploy暂时不支持，所以直接使用onnxruntime进行部署 python tools/export_onnx.py
模型上传到modescope
bash
curl -s https://packagecloud.io/install/repositories/github/git-lfs/script.deb.sh | bash
apt install git-lfs
git clone https://www.modelscope.cn/flyer123/GlauClsDRGrading.git
cd GlauClsDRGrading
git lfs track model.onnx
git add model.onnx
git config --global user.email "you@example.com"
git config --global user.name "Your Name"
git commit -m "add model"
git push
Owner
- Name: Yajie Yang
- Login: JieGenius
- Kind: user
- Location: Beijing, China
- Repositories: 1
- Profile: https://github.com/JieGenius
I am pursuing my master's degree in Renmin University of China
Citation (CITATION.cff)
cff-version: 1.2.0 message: "If you use this software, please cite it as below." title: "OpenMMLab's Pre-training Toolbox and Benchmark" authors: - name: "MMPreTrain Contributors" version: 0.15.0 date-released: 2023-04-06 repository-code: "https://github.com/open-mmlab/mmpretrain" license: Apache-2.0
GitHub Events
Total
- Watch event: 1
- Fork event: 1
Last Year
- Watch event: 1
- Fork event: 1
Dependencies
- actions/checkout v3 composite
- actions/setup-python v4 composite
- actions/checkout v3 composite
- actions/setup-python v4 composite
- codecov/codecov-action v1.0.14 composite
- actions/checkout v3 composite
- actions/setup-python v4 composite
- actions/checkout v3 composite
- actions/setup-python v4 composite
- pytorch/pytorch ${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel build
- pytorch/pytorch ${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel build
- pytorch/torchserve latest-gpu build
- docutils ==0.18.1
- modelindex *
- myst-parser *
- pytorch_sphinx_theme *
- sphinx ==6.1.3
- sphinx-copybutton *
- sphinx-notfound-page *
- sphinx-tabs *
- sphinxcontrib-jquery *
- tabulate *
- mmcv >=2.0.0,<2.4.0
- mmengine >=0.8.3,<1.0.0
- pycocotools *
- transformers >=4.28.0
- albumentations >=0.3.2
- grad-cam >=1.3.7,<1.5.0
- requests *
- scikit-learn *
- mmcv-lite >=2.0.0rc4
- mmengine *
- pycocotools *
- torch *
- torchvision *
- transformers *
- einops *
- importlib-metadata *
- mat4py *
- matplotlib *
- modelindex *
- numpy *
- rich *
- coverage * test
- interrogate * test
- pytest * test