Skip to content

rajpurkarlab/disease-seg-baseline

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

42 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

disease-seg-baseline

Instructions

Evaluating Disease Segmentation

After running pip3 install -r requirements.txt, run

python3 run_seg.py [model] [test_set] [visualize] [method]

where model is the name of the model to use (e.g, BioViL), test_set is the name of the test set to be used for evaluation (e.g, CheXlocalize), visualize denotes whether or not to output plots of imags and masks (e.g., yes or no), and method is what phrase grounding method to use (naive, grad_cam, cocoa).

Example command:

python3 run_seg.py BioViL CheXlocalize no grad_cam

Finding Best Prompts

Run

python3 find_prompts.py [model] [validation_set] [corpus_set] [method]

where model is the name of the model to use (e.g, BioViL), validation_set is the name of the validation set to be used (e.g., CheXlocalize), corpus_set is the set of all report phrases to search over (e.g, MIMIC-CXR or MS-CXR), and method is what phrase grounding method to use (naive, grad_cam, cocoa).

Example command:

python3 find_prompts.py BioViL CheXlocalize MS-CXR naive

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages