# DLTTA: Dynamic Learning Rate for Test-time Adaptation on Cross-domain Medical Images
Pytorch implementation for TMI paper DLTTA: Dynamic Learning Rate for Test-time Adaptation on Cross-domain Medical Images, by [Hongzheng Yang](https://github.com/HongZhengYang), [Chen Cheng](https://cchen-cc.github.io/), [Meirui Jiang](https://meiruijiang.github.io/MeiruiJiang/), [Quande Liu](https://liuquande.github.io/), [Jianfeng Cao](), [Pheng-Ann Heng](http://www.cse.cuhk.edu.hk/~pheng/), [Qi Dou](http://www.cse.cuhk.edu.hk/~qdou/).
## Abstract
![](assets/Figure.png)
## Files
In this repository, we provide the implementation of our dynamic learning rate method on OCT dataset. The ATTA and Tent implementation were adopted from their official implementation. ([Tent](https://github.com/DequanWang/tent), [ATTA]())
To reproduce results on Camelyon17 and Prostate datasets, please refer to the experiments folder.
## Datasets
The OCT dataset can downloaded from [here](http://iacl.ece.jhu.edu/index.php?title=Resources).
The Camelyon17 dataset can be downloaded from [here](https://wilds.stanford.edu/).
The Prostate dataset can be downloaded from [here](https://liuquande.github.io/SAML/).
## Usage
1. create conda environment
conda create -n DLTTA python=3.7
conda activate DLTTA
2. Install dependencies:
1. install pytorch==1.7.0 torchvision==0.9.0 (via conda, recommend)
3. download the dataset
4. download the pretrained model from [google drive](https://drive.google.com/drive/folders/1RyTLkoLf1w_7rgvQl_aEERd-8YKGc0Yt?usp=sharing)
5. modify the corresponding data path and model path in test.sh
6. run test.sh to adapt the model
## Citation
If this repository is useful for your research, please cite:
@article{2022DLTTA,
title={DLTTA: Dynamic Learning Rate for Test-time Adaptation on Cross-domain Medical Images},
author={Hongzheng Yang, Cheng Chen, Meirui Jiang, Quande Liu, Jianfeng Cao, Pheng Ann Heng, Qi Dou},
year={2022}
}
### Questions
Please feel free to contact 'hzyang05@gmail.com' if you have any questions.
没有合适的资源?快使用搜索试试~ 我知道了~
[IEEETMI'22]DLTTA:跨域医学图像测试时间适应的动.zip
共74个文件
py:70个
sh:2个
png:1个
需积分: 5 0 下载量 168 浏览量
2023-04-09
13:33:22
上传
评论
收藏 591KB ZIP 举报
温馨提示
[IEEETMI'22]DLTTA:跨域医学图像测试时间适应的动
资源推荐
资源详情
资源评论
收起资源包目录
[IEEETMI'22]DLTTA:跨域医学图像测试时间适应的动.zip (74个子文件)
DLTTA-main
memory.py 2KB
assets
Figure.png 480KB
experiments
camelyon17
utils.py 14KB
__init__.py 0B
losses.py 1KB
evaluate
mbtt.py 4KB
models.py 3KB
MbPA.py 2KB
conf.py 4KB
camelyon17_dataset.py 4KB
wilds_dataset.py 18KB
train.py 3KB
evaluate.py 10KB
networks
__init__.py 0B
models.py 4KB
densenet.py 10KB
scheduler.py 4KB
configs
utils.py 7KB
scheduler.py 570B
data_loader.py 229B
model.py 2KB
algorithm.py 3KB
datasets.py 18KB
supported.py 2KB
train_atta.py 14KB
run_expt.py 23KB
transforms.py 12KB
data_augmentation
__init__.py 0B
randaugment.py 3KB
autoencoder.py 707B
ttt.py 1KB
train_atta_ttt.py 14KB
models
initializer.py 10KB
__init__.py 0B
layers.py 280B
resnet_multispectral.py 9KB
train.py 10KB
train_ttt.py 14KB
optimizer.py 2KB
prostate
train
__init__.py 0B
center.py 1KB
test_dataset.py 2KB
api.py 3KB
configs.py 2KB
model_trainer.py 846B
model_trainer_segmentation.py 6KB
main.py 3KB
data
__init__.py 1B
generate_data_loader.py 390B
prostate
__init__.py 1B
dataset.py 2KB
generate_data.py 644B
utils
__init__.py 0B
loss.py 6KB
data_preprocess.py 2KB
nets
utils.py 11KB
__init__.py 0B
unet.py 6KB
buildingblocks.py 18KB
utils
__init__.py 0B
util.py 5KB
datasets
__init__.py 3KB
dataset.py 5KB
transform.py 8KB
models
__init__.py 485B
backends.py 5KB
segmodel.py 5KB
basemodel.py 21KB
adapmodel.py 5KB
train.py 4KB
README.md 2KB
config.py 8KB
scripts
train_oct.sh 1KB
test_oct.sh 876B
共 74 条
- 1
资源评论
快撑死的鱼
- 粉丝: 1w+
- 资源: 9154
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功