Skip to content

Bump macros to 92f54bf and use prederrf for prediction errors - #240

Merged
d-morrison merged 3 commits into
mainfrom
claude/project-thread-91flz2
Oct 22, 2026
Merged

d-morrison merged 3 commits into
mainfrom
claude/project-thread-91flz2

Conversation

@d-morrison

Copy link
Copy Markdown
Contributor

Requested by Ezra · project thread

Ezra asked on #238 whether \prederr should be a shorthand for \esterr. Morrison-Lab/macros#276 added \prederrf{g}, the prediction error of the predictor g, equal to \esterr{g}{\Outvar} with the label pred.

Changes

  • Bumps the latex-macros pin from 5f14c45 to 92f54bf (macros#270, #272, #274, #275, #276 and #277 on top).
  • Five subfiles replace the long forms in all 21 uses: \esterr{\predfnf{\Predvar}}{\Outvar} becomes \prederrf{\predfn}, \esterr{\truemeanf{\Predvar}}{\Outvar} becomes \prederrf{\truemean}, and \esterr{\predfn_1\paren{\Predvar}}{\Outvar} becomes \prederrf{\predfn_1}.
  • No other macro changed in that range is used in pds.

Not verified

The container has no LaTeX or the dobson R package, so expectation.qmd (the only page that includes these subfiles) was not rendered here. I will check the PR preview from gh-pages.


Generated by Claude Code

@d-morrison d-morrison self-assigned this Oct 22, 2026
@github-actions

This comment has been minimized.

@github-actions

This comment has been minimized.

@github-actions

This comment has been minimized.

Copy link
Copy Markdown
Contributor Author

review / require-clean-verdict is red on 5f984b7 because the Claude review was skipped with a 429 (account session limit, resets 1:20am UTC; gha#520), not because of a finding. The only change since the clean review of 7c7f305 is a merge of main (two workflow files from #239). Per the reviewer fallback in ai-config#4549, an independent separate-agent Claude review of this head is running and its verdict will be posted here; the automated review can be re-run after the reset.


Generated by Claude Code

Copy link
Copy Markdown
Contributor Author

Separate-agent Claude review of head 5f984b7 (reviewer fallback, ai-config#4549): CLEAN. Every \prederrf replacement denotes the same quantity as the \esterr form it replaces; \prederrf, \predfn and \truemean are defined in the pinned macros; no \esterr or \prederr use remains outside latex-macros; the diff is the five subfiles plus the submodule pointer. The automated Claude review of this head is still skipped by the 429 and can be re-run after the reset.
Posted by Claude Code (AI agent) --- not written by a human.


Generated by Claude Code

… devof for the gap

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Lup9bgjrsSr4RGUfSDfGjB

Copy link
Copy Markdown
Contributor Author

Separate-agent Claude review of head 595ed43 (reviewer fallback, ai-config#4549; the automated review is still skipped by the 429): CLEAN. Each aligned line of the risk-decomposition proof equals the previous one; \devof{a}{b} prints (a - b) in parentheses, so it matches the old \paren{...}; the multi-line block keeps one comment column per row; no \esterr or \prederr remains outside latex-macros.
Posted by Claude Code (AI agent) --- not written by a human.


Generated by Claude Code

@github-actions

Copy link
Copy Markdown
Contributor

Claude finished review — View run


The commit under review is 595ed43, and it changes only _subfiles/_thm-risk-decomposition.qmd. It still leaves no defects. This confirms my earlier review of 7c7f305 (2026-10-22 comment), which said Ready for merge; that verdict still stands.

  • What the commit does: it splits the long expansion lines into one term per line and swaps each \paren{a - b} for \devof{a}{b}. I did not render the page, so the math is checked by reading the source only.
  • Macro exists: latex-macros/macros.qmd:255 defines \providecommand{\devof}[2]{{\paren{#1 - #2}}}. It expands to the same parenthesised difference as before, so every \devof substitution reproduces the old expression.
  • Cross term: the cross term now reads \prederrf{\truemean} \cd \devof{...}, which adds an explicit \cd between the two factors. That is the same product with a visible multiplication sign.
  • Line breaks: the + lines begin the continuation inside one aligned row, with no \\ and no stray &. The line breaks are whitespace to LaTeX and MathJax, so the rendered equation is unchanged. The && annotation columns are untouched.
  • Anchors: no #id anchors or page paths changed, so links from course sites such as mlds are unaffected.
  • Merge commit: 5f984b7 matches git's automatic merge of its parents. It only brought in 32531cb from main, so there is nothing hand-resolved to check.

Verdict

Ready for merge

Structured Review Data (JSON)
{
  "schema_version": "1.1",
  "reviewer": "claude",
  "commit_sha": "595ed43693d1a9c63593d190953f63c3eec7a4dd",
  "verdict": "CLEAN",
  "findings": [],
  "detailed_assessment": "In _subfiles/_thm-risk-decomposition.qmd the devof macro exists at latex-macros/macros.qmd:255 and expands to the same parenthesised difference, and the one-term-per-line breaks stay inside a single aligned row.",
  "holistic_assessment": "The commit is a scoped formatting and notation change with no anchor or path changes, so cross-site links and regression risk are minimal; the merge commit matches git's automatic merge."
}

Reviewed commit: 595ed43

@github-actions

Copy link
Copy Markdown
Contributor

💰 Cost: $0.1564 (review) — run

@github-actions

github-actions Bot commented Oct 22, 2026 •

Copy link
Copy Markdown
Contributor
PR Preview Action v1.8.1
Preview removed because the pull request was closed.

2026-10-22 22:55 UTC

@d-morrison
d-morrison merged commit ba7401f into main Oct 22, 2026
29 checks passed
@d-morrison
d-morrison deleted the claude/project-thread-91flz2 branch October 22, 2026 22:55
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants