Proposed solution and baseline for CheXpert dataset, implemented in PyTorch. CheXpert is a large dataset of chest X-rays and competition for automated chest x-ray interpretation by Stanford ML Group.
The current baseline model has been trained on the low resolution version of the dataset and is based on ideas drawn from the paper presented along with the dataset.