BurnAnalysis

1. Environment

• Please prepare an environment with python=3.10|cudatoolkit=11.3.1, and then use the command "pip install -r requirements.txt" for the dependencies.

2. Prepare Burns datasets

3. Train model

• To train a multi-task network, you can enter the following command in the terminal interface.

python train_net.py --data-split='0', --batch-size=4, --max-epoch=150, --ckpt-dir='ckpt/tmpckpt'

4. Test model and evaluate

• To test a trained multi-task network, you can enter the following command in the terminal interface.

python test_eval.py --data-split='0', --batch-size=4, --ckpt-dir='ckpt/tmpckpt', --out-dir='output/tmp', --cndct-vis=False

BurnAI Application (Android OS version ≥10)

• You can download the installation pakage from the Baidu Netdisk (access code: gr67).