Fix processing tests #939
Fix processing tests #939
Conversation
|
Ah heck, so this PR is just to address the processing tests, not the others... Any ideas on how to do a PR that fixes only some of the tests? @ns-rse |
|
The processing tests pass on my machine, you might want to verify though |
…ng' into SylviaWhittle/800-tests-processing
The Continuous Integration tests will continue to fail until all tests pass, no way around that I can think of without switching to using Its useful to run the whole test suite in CI though and ensuring it runs on all pull requests was something I deliberately introduced the other day with #921 as had this been the case from the start (rather than just running on PRs that target |
…ng' into SylviaWhittle/800-tests-processing
SylviaWhittle
merged commit f890e6d
into
maxgamill-sheffield/800-better-tracing
| @@ -295,6 +300,41 @@ def remove_small_objects(self, image: np.array, **kwargs) -> npt.NDArray: | |||
| return small_objects_removed > 0.0 | |||
| return image | |||
|
|
|||
| def remove_objects_too_small_to_process( | |||
| self, image: npt.NDArray, minimum_size_px: int, minimum_bbox_size_px: int | |||
There was a problem hiding this comment.
Do we expect this method to be used independently of the processing?
If not then I don't see any advantage have minimum_size_px or minimum_bbox_size_px as arguments to the method as they are set as attributes to the class when it is initialised and available as self.minimum_grain_size_px and self.minimum_bbox_size_px.
The comment above on hardcoding suggests its not something that will ever be changed.
| bbox_width = region.bbox[2] - region.bbox[0] | ||
| bbox_height = region.bbox[3] - region.bbox[1] | ||
| # If the minimum dimension of the bounding box is less than the minimum dimension, remove the region | ||
| if min(bbox_width, bbox_height) < minimum_bbox_size_px: |
There was a problem hiding this comment.
However unlikely to get an almost straight line is it possible that there would ever be a bounding box with one dimension < 5 and the other say 400 pixels?
Would such an image not be processed by DNA tracing?
Could perhaps have such a dummy grain as a test scenario to check (but I appreciate this is additional work).
This PR (hopefully) fixes the tests in
test_processing.pyfor the #932