xray-lightning

Pytorch Lightning object detection pipeline for Chest X-ray data

Training

python train.py data.num_workers=10 trainer.max_epochs=3

python run.py logging.best_model_path=outputs/2020-07-06/22-47-46/best_model/model.pth data.batch_size=1 data.num_workers=1

This will generate predictions.csv that can be used for future analysis.

Note: To do predictions for random 100 images instead of entire dataset, also add testing.debug=True to command line arguments

Some of the results can be found in this notebook.

Name		Name	Last commit message	Last commit date
Latest commit History 14 Commits
data		data
notebooks		notebooks
requirements		requirements
utils		utils
.gitignore		.gitignore
.pre-commit-config.yaml		.pre-commit-config.yaml
LICENSE		LICENSE
README.md		README.md
config.yaml		config.yaml
dataset.py		dataset.py
logger.py		logger.py
model.py		model.py
pyproject.toml		pyproject.toml
run.py		run.py
setup.cfg		setup.cfg
train.py		train.py