Repository navigation
Review fixes: drop a redundant re-parse and align stale strategy counts - #16
Merged
Merged
Conversation
Apply the surviving findings from a max-effort review of the rule-engine ablation diff: - rule_engine.next_params: read the previous iteration's factors/iterations straight from its params dict instead of re-parsing a whole ALSParams (whose full field validation is redundant here) only to read two numbers. Removes the derived `prev` and collapses the duplicate None-guard. - config.py: the "All three ... continuous hull" comment and the MAX_ITERATIONS cap comment undercounted the sharers; the diff added RULE_SPACE directly below the former and rule loops the same budget. Reword to match. - CLAUDE.md: the project overview and the "Comparability is fair by design" section still said four strategies / listed only RANDOM/BAYES/HOLMES_SPACE; add rule and RULE_SPACE so the authoritative guide matches the README and the updated TestComparabilityInvariants guardrail. Behavior-preserving. The rule-engine decision logic's other review candidates were all documented inherent limitations, test-locked intended behavior, or consequences of the deliberately stateless ablation design, so left as-is. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VNYwpwcwSLW5ySf8dcQbpr
Code reviewNo issues found. Checked for bugs and CLAUDE.md compliance. |
A second review pass over the previous review-fix commit surfaced two things:
- rule_engine.next_params: the prev_params local became a do-nothing alias of
prev_entry["params"] once the ALSParams re-parse was dropped; inline it so all
three lookback flags share the one prev_entry None-guard shape.
- The strategy-count/roster drift class ran wider than the spots patched last
time: stale "three strategies" / "grid/random/bayes" / "Bayes and HOLMES"
claims remained in config.py (GRID_SPACE comment, _grid_hull docstring),
CLAUDE.md (Data scale), holmes.py (bounds docstring and error message,
TrajectoryEntry.k, run_iteration budget note), SKILL.md (frontmatter, budget
and ranges notes), dispatch.py --strategies help, and test docstrings
(TestComparabilityInvariants, TestWriteSearchOutput, budget/cutoff tests).
Reword them count-free ("every strategy", "the other strategies", "all of
them") so the next strategy addition cannot re-create the drift, and fix the
test comment claiming the ALSParams defaults are not all inside HOLMES_SPACE
(they are — the rule-engine guardrail test locks exactly that).
Behavior-preserving; no test pins any of the reworded text.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VNYwpwcwSLW5ySf8dcQbpr
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
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.
Apply the surviving findings from a max-effort review of the rule-engine
ablation diff:
straight from its params dict instead of re-parsing a whole ALSParams (whose
full field validation is redundant here) only to read two numbers. Removes
the derived
prevand collapses the duplicate None-guard.cap comment undercounted the sharers; the diff added RULE_SPACE directly below
the former and rule loops the same budget. Reword to match.
section still said four strategies / listed only RANDOM/BAYES/HOLMES_SPACE;
add rule and RULE_SPACE so the authoritative guide matches the README and the
updated TestComparabilityInvariants guardrail.
Behavior-preserving. The rule-engine decision logic's other review candidates
were all documented inherent limitations, test-locked intended behavior, or
consequences of the deliberately stateless ablation design, so left as-is.
Co-Authored-By: Claude Opus 4.8 noreply@anthropic.com
Claude-Session: https://claude.ai/code/session_01VNYwpwcwSLW5ySf8dcQbpr