Closed GalyaZalesskaya closed 2 months ago
Thanks for the nice feature, could you please post some screenshots of how it looks like in jupyter and python?
Regarding juputer, we might want to update the openvino notebooks with this feature (where applicable) after the release. This might be a good check to create a draft pull request with your current branch and see how it works in current notebooks
Attention: Patch coverage is 97.77778%
with 1 line
in your changes missing coverage. Please review.
Project coverage is 92.17%. Comparing base (
00ebd2e
) to head (a6bc688
).
Files | Patch % | Lines |
---|---|---|
openvino_xai/explainer/explanation.py | 97.77% | 1 Missing :warning: |
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
@negvet Added pretty images to the description :)
Use matplotlib (recommended for Jupyter) - default backend
explanation.plot(targets=[7, 11], backend="matplotlib") explanation.plot(targets=-1, backend="matplotlib") # plot saliency maps for all classes explanation.plot(targets=['cat', 'dog'], backend="matplotlib")
Use OpenCV (recommended for Python) - will open new windows with maps
explanation.plot(targets=[7, 11], backend="cv") explanation.plot(targets=-1, backend="cv") # plot saliency maps for all classes explanation.plot(targets=['cat', 'dog'], backend="cv")