Introducing ShortSeq192, the successor to ShortSeq128 #10
Conversation
This file contains 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
… logscale ticks thicker to emphasize that these plots do not use a linear scale like the others (affects test_hamming_distance() only)
… as it was when bases were being encoded into a uint128_t, so it effectively accommodates an extra 32 bases for free. Compared to the current object size, it is 8 bytes larger but with the benefit of keeping 33-96 nt sequences the same fixed size. This commit also fixes a bug that was introduced in the last PR. During equality checks, memcmp() was being called with an improper length (length in nucleotides rather than length in encoded bytes). This resulted in comparisons between memory regions that were much larger than necessary I still have no idea how this didn't segfault.
…eq marshalling methods. The logic for this task has been SIGNIFICANTLY refined and will be shared by the two array-type ShortSeqs: ShortSeq192 and ShortSeqVar. Benchmarks show a healthy improvement in encoding runtime and consistency. This commit also updates the benchmark plot for construction from PyBytes. I cleaned up my testing environment so the measurements have less variance. Curiously the Numpy task is consistently taking a couple microseconds longer and I don't yet have an explanation for why.
This commit also: - Fixes a bug in equality checks between two ShortSeqVars. Previously, the comparison sampled a memory region much larger than it should have been. I still have no idea how this didn't segfault. - Updates the constant MIN_VAR_NT - Removes the duplicate constant NT_PER_BLOCK
# Conflicts: # doc/plots/mem_by_length.svg
… a cleaner testing environment. Also making y-axis ticks thicker to emphasize the log scale
|
This failed test is because, for some reason, some of the macOS runners have a Xenon E5-1650 v2 CPU.
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
ShortSeq128has been superceded byShortSeq192, which adds an additional block to the fixed length array for storing packed nucleotides. The new object size is the same as it was when bases were being encoded into a uint128_t, so it effectively accommodates an extra 32 bases "for free." Compared to the current object size ofShortSeq128, it is 8 bytes larger but with the benefit of keeping 33-96 nt sequences the same fixed size.Additional changes in this PR:
ShortSeq192andShortSeqVar.ShortSeqVarobjects. Previously a larger than necessary memory region was used for comparison.