Enhanced U-Net: A Feature Enhancement Network for Polyp Segmentation

Requirements

torch
torchvision
scipy
PIL
numpy
tqdm

Data

train

Images \
Masks \ test \
Images \
Masks \ Valid \
Images \
Masks

1. Training

python train.py  --mode train  --train_data_dir /path-to-train_data  --valid_data_dir  /path-to-valid_data

2. Testing

bash python test.py --mode test --load_ckpt checkpoint --test_data_dir /path-to-test_data