Add ImageRegistrationMethodDisplacement1/.Test2 baselines (Mattes metric) - #56
Draft
blowekamp wants to merge 1 commit into
Draft
Add ImageRegistrationMethodDisplacement1/.Test2 baselines (Mattes metric)#56blowekamp wants to merge 1 commit into
blowekamp wants to merge 1 commit into
Conversation
…ha (Mattes metric, single-threaded)
2 tasks
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
New primary baselines for SimpleITK's ImageRegistrationMethodDisplacement1 and ImageRegistrationMethodDisplacement1Test2 examples, generated after switching the example's stage-1 metric from JointHistogramMutualInformation to MattesMutualInformation.
Root cause context: JointHistogramMutualInformation's gradient was fixed in ITK (InsightSoftwareConsortium/ITK#6569), causing these examples to diverge from all existing baselines. Rather than chasing a moving target, the example was switched to MattesMutualInformation, which is unaffected by that fix and (verified) produces byte-identical output between the official SimpleITK v3.0.0b1 release and current ITK main, when run single-threaded.
Generated with ITK_GLOBAL_DEFAULT_NUMBER_OF_THREADS=1 for reproducibility: MattesMutualInformationImageToImageMetricv4's threaded derivative reduction (DerivativeBufferManager::ReduceBuffer) accumulates thread-local buffers into a shared array via mutex-protected += at OS-scheduling-dependent times, and floating-point addition is not associative -- so multi-threaded runs of the same build have small (~0.17-0.25 RMS) run-to-run variance that compounds over the ~74-95 iterations Displacement1Test2 takes to converge. Single-threaded execution eliminates this: confirmed deterministic across repeated runs (RMS 0.0) and identical between old (v3.0.0b1) and new (current main ITK) builds (RMS 0.0).