Swin-Transformer训练自己的数据集

环境配置

#创建conda环境
conda create -n SwinTransformerObjectDetection -y python=3.8
conda activate SwinTransformerObjectDetection
conda install pytorch=1.8.0 torchvision cudatoolkit=10.2 -c pytorch -y

# mmcv安装
cd ~/software/
rm -rf mmcv
git clone https://github.com/open-mmlab/mmcv.git
cd mmcv
MMCV_WITH_OPS=1 pip install -e .

# 安装apex
git clone https://github.com/NVIDIA/apex
cd apex
pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./

# mmdetection安装
git clone https://github.com/open-mmlab/mmdetection.git
cd mmdetection
pip install -r requirements/build.txt
pip install -r requirements.txt
pip install -v -e . 

cd /home/zengxh/workspace/Swin-Transformer-Object-Detection
pip install -r requirements.txt

测试是否安装成功

python demo/image_demo.py demo/demo.jpg configs/swin/mask_rcnn_swin_tiny_patch4_window7_mstrain_480-800_adamw_3x_coco.py mask_rcnn_swin_tiny_patch4_window7.pth

下载文件

下载changemaskrcnn.py文件

https://github.com/carlsummer/python_developer_tools/tree/main/python_developer_tools/cv/Transformer/Swin-Transformer-Object-Detectionhttps://github.com/carlsummer/python_developer_tools/tree/main/python_developer_tools/cv/Transformer/Swin-Transformer-Object-Detection

训练自己的数据

1. 准备coco数据集中的(数据集准备部分)
2. 修改changemaskrcnn.py中num_class并运行，将num_class改成自己的数据集的种类数目

3. 修改configs\_base_\models\mask_rcnn_swin_fpn.py中num_classes
4. 修改configs\_base_\default_runtime.py中interval,load_from
5. 修改configs\swin\mask_rcnn_swin_tiny_patch4_window7_mstrain_480-800_adamw_3x_coco.py中的max_epochs、lr
6. 修改configs\_base_\datasets\coco_instance.py中samples_per_gpu和workers_per_gpu
7. 修改mmdet\datasets\coco.py中CLASSES
8. python tools/train.py configs/swin/mask_rcnn_swin_tiny_patch4_window7_mstrain_480-800_adamw_3x_coco.py