New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Image Experiments (quantitative): KernelSHAP #176

Open

2 tasks

elboyran opened this issue Feb 21, 2023 · 1 comment

Labels

blocked

Contributor

elboyran commented Feb 21, 2023

For the range of KernelSHAP method parameters defined in issue #174

Use a notebook (?) to compare pairs of 'neighboring' test images (as created in issue dianna-ai/dianna#470) and their KernelSHAP relevance maps .

Evaluate and display the p(average/mean/max/median?) relevancy vs the shape parameter scale for:

Circles. The 'neighbouring' images are defined as ones with close shape parameters (one variable, 2- fixed):
size and
gray-scale contrast
Triangles. The 'neighbouring' images are defined as ones with close shape parameters (one variable, 2- fixed):
size
rotation angle and
gray-scale contrast

Stems from dianna-ai/dianna#446 and #187.

elboyran added the blocked label

Contributor Author

elboyran commented Feb 21, 2023

Blocked by dianna-ai/dianna#470, #174 and the creation of research template notebook similar to the text case (dianna-ai/dianna#456).

elboyran mentioned this issue

Create concrete research tasks as issues: images dianna-ai/dianna#446

Closed

elboyran transferred this issue from dianna-ai/dianna

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment