More resilient regex in utils.code_utils.extract_diffs and removed re…#3
Merged
codelion merged 2 commits intoalgorithmicsuperintelligence:mainfrom May 19, 2025
Conversation
…dundant implementation in apply_diff() function
Member
|
@jvm123 thanks for your contributions and the PRs. Can you please merge from main once and resolve these conflicts so that I can accept this PR. |
shubham3-ucb
pushed a commit
to mert-cemri/autoevolve-dev
that referenced
this pull request
Nov 26, 2025
…/improved-whitespace-handling More resilient regex in utils.code_utils.extract_diffs and removed re…
shubham3-ucb
pushed a commit
to mert-cemri/autoevolve-dev
that referenced
this pull request
Nov 26, 2025
merge from upstream
toslali-ibm
added a commit
to toslali-ibm/openevolve
that referenced
this pull request
Feb 17, 2026
1. Fix scoring formula documentation mismatch (Issue algorithmicsuperintelligence#3) - Update config.yaml to match actual implementation: score = -avg_latency - Clarify that higher score (less negative) = better 2. Fix baseline weight mismatch (Issue algorithmicsuperintelligence#12) - Update routing_policy.yaml to match documented baseline - Change from cache=0/load=0.9 to cache=0.6/load=0.4 3. Add robust cluster metric parsing (Issue algorithmicsuperintelligence#5) - Validate number of e2e_mean_ms matches (expect num_instances + 1) - Handle single-instance case - Warn if unexpected number of matches These fixes ensure: - LLM understands correct optimization direction - Evolution optimizes around correct baseline weights - Metric parsing is resilient to BLIS output changes Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
toslali-ibm
added a commit
to toslali-ibm/openevolve
that referenced
this pull request
Mar 3, 2026
Bug #1 (CRITICAL): program.diversity was never set on Program objects in _calculate_feature_coords — only computed as a local for bin indexing. This caused the rescue selection diversity filter to always see 0.0 for all programs, making diversity_median=0 and the filter a no-op. ~50% of programs got rescue-tuned instead of the intended ~32%. Bug algorithmicsuperintelligence#2 (MODERATE): _polish_elites bypassed best-program tracking by writing directly to self.database.programs[id] without calling _update_best_program(). If final polish improved a non-best program to surpass the current best, get_best_program() would return the old best. Bug algorithmicsuperintelligence#3 (MODERATE): The rescue tuning exception handler lost tracking data (tuning_stats_dict stayed None) and discarded pre_tune_metrics even when already computed, causing a redundant evaluation and gaps in the tuning tracker's total_iterations count. Bug algorithmicsuperintelligence#4 (MINOR): annotations_stripped counter used .count("@tune") which also matches "@tuned" as a substring. Replaced with word-boundary regex _TUNE_WORD_RE = re.compile(r"@tune\b(?!D)"). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
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
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.
…dundant implementation in apply_diff() function
The more resilient white-space handling fixes unit test failures of the two functions.