this project can be trained on Google Colab,while you should turn your VPN on.
Clone the repository
git clone https://github.com/Andyato/Pulmonary-Nodule-Detection
The data is VOC format
├── VOCdevkit
│ ├── VOCdevkit_train
│ ├── Annotation
│ ├── JPEGImages
│ ├── VOCdevkit_test
│ ├── Annotation
│ ├── JPEGImages
cd $ROOT/data/io/
python convert_data_to_tfrecord.py --VOC_dir='***/VOCdevkit/VOCdevkit_train/' --save_name='train' --img_format='.jpg' --dataset='nodules'
1、Modify $ROOT/libs/lable_name_dict/***_dict.py, corresponding to the number of categories in the configuration file
2、download pretrain weight(resnet_v1_101_2016_08_28.tar.gz or resnet_v1_50_2016_08_28.tar.gz) from here, then extract to folder $ROOT/data/pretrained_weights
3、
cd $ROOT/tools
python train.py
cd $ROOT/tools
python test.py
cd $FPN_ROOT/tools
python eval.py --weights (your trained weights directory path) --image_num (eval images number)
tensorboard --logdir=$ROOT/output/res**_summary/