Prompt-based Grouping Transformer for Nucleus Detection and Classification

-python=3.8

-pytorch=1.12.1+cu102

Install mmcv using mim

pip install -U openmim
mim install mmcv-full==1.6.1

Git clone acformer

git clone https://github.com/LL3RD/PGT.git

Install

cd PGT
cd thirdparty/mmdetection
python -m pip install -e .
cd ../..
python -m pip install -e .

Your can download 20x CoNSeP Dataset and Checkpoint from here.

Your can download BRCA Dataset and Checkpoint from here.

Your can download Original Lizard from the official website or Preprocessed Lizard and Checkpoint that is converted to hovernet consep format and split into patches.

The training contains two phases, In the pre-tuning phase, modify your dataset path in configs/Swin_BRCA.py and run

CUDA_VISIBLE_DEVICES=0 bash tools/dist_train.sh configs/Swin_BRCA.py 1 --work-dir=Path_to_your_work_dir

Then modify the load_from path in configs/Swin_BRCA_PGT.py and run

CUDA_VISIBLE_DEVICES=0 bash tools/dist_train.sh configs/Swin_BRCA_PGT.py 1 --work-dir=Path_to_your_work_dir

Modify your dataset path and checkpoint path in tools/inference and run

python tools/inference/inference_CoNSeP_group.py

python tools/inference/inference_BRCA_group.py

python tools/inference/inference_Lizard_group.py

• PGT is built based on SoftTeacher and MMDetection.