DFT shift fix for new scikit, speed improvement for find_nearest #638
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
… using writeto built in overwrite
# Conflicts: # vip_hci/fm/negfc_simplex.py # vip_hci/fm/utils_negfc.py
Contributor
|
Thanks @IainHammond ! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This request is mainly to address the upgrade from scikit-image 0.18.3 to 0.23.1, which released today.
Using
phase_cross_correlationin the_shift_dftfunction would not give a backwards compatible answer for versions>0.18.3 without providingnormalization=None. There is now a check on the version of scikit to handle the case where a user has scikit 0.18.3 and this parameter doesn't exist, and two outputs are returned from the function (compared to three in 0.22 ->). The version requirement has been removed from requirements.txt and a standard installation of VIP will grab scikit-image 0.23.1.I've also sped up find_nearest with some boolean operations and numpy functions (not that it needed it).
fakecomp.py and fit_2d.py have had a small tidy up relating to the import for centroid_com from photutils.centroids as the old method for importing this function is 8 years old. Avoiding this check makes the initial start up faster by avoiding the entire package being imported, which can be about 160MB