Debug calcfitmod by sschildhauer · Pull Request #240 · UCD-SERG/serodynamics

sschildhauer · 2026-05-27T22:19:32Z

Debugging calc_fit_mod in two ways:

Currently calc_fit_mod takes a filtered raw data set in run_mod that calculates the fitted and residuals for only a subset of the data. This PR changes this piece to take the full raw input data set and calculate fitted and residuals for the whole data set.
This PR also adds Stratification as a variable in the fitted_residuals output.
Linking to issues Fix comments on calc_fit_mod #226 and Bug: calc_fit_mod uses only the last stratum's data (R/Run_Mod.R:268–271) #215

…d input data for calc_fit_mod

github-actions · 2026-05-27T22:24:57Z

📕 Preview documentation for this PR has been cleaned up.

codecov · 2026-05-27T23:12:00Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ All tests successful. No failed tests found.

Files with missing lines	Coverage Δ
R/Run_Mod.R	`100.00% <100.00%> (ø)`
R/calc_fit_mod.R	`100.00% <100.00%> (ø)`

sschildhauer · 2026-05-28T00:12:54Z

@claude, please help me fix this PR to pass all reviews. Please keep all edits within this thread.

claude · 2026-05-28T00:16:08Z

Claude finished @sschildhauer's task in 12m 11s —— View job

Fixed: regenerated the missing Linux snapshots

Root cause: The 'by' must specify a uniquely valid column merge error shown in the codecov comment was stale — it came from a commit before the calc_fit_mod merge logic was fixed. I pulled the live R-CMD-check log for the latest commit (d086af6, ubuntu R 4.6.0) and confirmed all 8 failures were simply "Adding new file snapshot":

The PR deleted the non-darwin snapshot files under tests/testthat/_snaps/run_mod/ but never regenerated them.
macOS CI passed because its darwin snapshots were updated; Linux & Windows look for the root (non-darwin) files, which no longer existed → failure on every non-macOS runner.

Fix: Regenerated the 8 deleted Linux snapshots on a matching environment (Linux, R 4.6.0, serocalculator 1.4.0.9013, JAGS 4.3.2 — same as the CI ubuntu runner). They now include the full-dataset fitted/residual rows and the new Stratification column (e.g. Subject,Iso_type,Stratification,t,fitted,residual). The full test-run_mod.R suite passes locally (FAIL 0).

Diagnose failures (missing Linux snapshots; no live merge error)
Regenerate Linux (root) snapshots
Run full test-run_mod.R suite → all pass locally
Commit & push (63db4f3)

Notes

No code change to calc_fit_mod.R was needed beyond an incidental trailing-whitespace cleanup; the merge logic already works on R 4.6.0.
The reviewer's minor suggestions (Ss mcmc diagnostic functions #3/Create README.md #4 — documenting that Stratification = "None" in the unstratified case) are optional polish and not blocking. Let me know if you'd like those roxygen clarifications added.
⚠️ Snapshots are sensitive to platform/dependency versions. I generated these on a Linux runner matching CI; the new push will reverify on Windows + all Ubuntu jobs. If any platform still reports a diff, it would be a genuine numeric difference to review.
• Branch: debug_calcfitmod

The non-darwin (Linux/Windows) snapshots for run_mod were deleted but not regenerated, causing R-CMD-check and test-coverage to fail with "Adding new file snapshot" on every non-macOS runner. Regenerate them so they include the full-dataset fitted/residual values and the new Stratification column. Co-authored-by: Samuel Hisaji Schildhauer <sschildhauer@users.noreply.github.com>

The new strat parameter added in this PR was not documented with an @param tag, causing an undocumented-argument gap in calc_fit_mod.Rd. Co-authored-by: Samuel Hisaji Schildhauer <undefined@users.noreply.github.com>

sschildhauer · 2026-05-28T04:24:09Z

@claude please review

claude · 2026-05-28T04:27:29Z

Claude finished @sschildhauer's task in 1m 48s —— View job

Code Review: PR #240 — Debug calcfitmod

Read CLAUDE.md and repo guidelines (no root CLAUDE.md found)
Read changed files (R/calc_fit_mod.R, R/Run_Mod.R)
Review documentation changes (man/*.Rd, NEWS.md, WORDLIST)
Check CI status for the latest commit (9 ✅ / 1 ❌ → the failure is just the Claude Code Review workflow itself; all tests, lint, R-CMD-check, pkgdown, coverage pass)
Provide review feedback

Summary

The core change is sound and the intent matches the PR description:

Before	After
`calc_fit_mod()` received `dl_sub` (filtered to current stratum), producing fitted/residuals only for that subset on each iteration; the last iteration silently overwrote prior strata because `fit_res` was reassigned, not concatenated.	`calc_fit_mod()` now receives the full `data` once, together with `strat`, and computes fitted/residuals for every observation in a single merge keyed by `(Subject, Iso_type[, Stratification])`.

The new Stratification column in the output ("None" when unstratified) is well documented and the snapshots reflect it. Nice catch on the underlying bug.

Findings

🟢 Correctness — looks good

R/Run_Mod.R:201-203 — passing data (not dl_sub) and strat is the right fix.
R/calc_fit_mod.R:50-57 — merge(..., all.y = TRUE) correctly preserves every observation from original_data even if a (Subject, Iso_type, Stratification) tuple has no posterior estimate; those rows surface as NA fitted/residual, which is the expected behavior.
Snapshot files are consistent across the darwin/ and root variants and now include the Stratification column; nostrat uses "None" exactly as documented.

🟡 Suggestions (non-blocking)

strat has no default in calc_fit_mod() (R/calc_fit_mod.R:23-25). Since calc_fit_mod is @keywords internal and only called by run_mod, this is fine — but a default of strat = NA would make the no-stratification branch self-contained and reduce the chance of a future caller forgetting the argument:
```
calc_fit_mod <- function(modeled_dat, original_data, strat = NA)
```
NA values in the stratification column are silently dropped from modeling but preserved in original_data. In R/Run_Mod.R:105-107, dplyr::filter(.data[[strat]] == i) drops rows where the strat column is NA. But calc_fit_mod then receives the full data, so those NA-strat rows will land in matched_dat with Stratification = NA and no posterior to merge against → rows with NA fitted/residual will appear in fitted_residuals. If that's intentional (full-data output, NA where un-modeled), great; if not, consider filtering them out or documenting the behavior. Worth at least a sentence in the @returns doc.
Potential name collision on "Stratification" (R/calc_fit_mod.R:36). rename("Stratification" = strat) will fail if the input data already has a column literally named Stratification and the user passes a different column via strat. Unlikely in practice, but a dplyr::rename with the target name already present errors out. A one-line guard or an dplyr::any_of cleanup before the rename would harden this.
Unrelated WORDLIST change. inst/WORDLIST removes allowlist and adds calc. allowlist is no longer referenced in any spell-checked file (only in .github/workflows/claude.yml as a comment), so removing it is safe — but it's unrelated to this PR's purpose. Not a blocker; just noting for cleanliness.
Minor style nits in R/calc_fit_mod.R:
- Line 26: blank line inside the function body at the very top is unusual.
- Line 35: trailing whitespace after .data$result, (Fix this →).
NEWS wording (NEWS.md:96). Current: "Fixing calc_fit_mod to include all stratifications in fitted_residuals attributes (Debug calcfitmod #240)." This understates the change — it also (a) computes fitted/residuals over the full dataset rather than the last stratum's filtered subset, and (b) adds a Stratification column even in unstratified runs. Consider rewording to something like:
- run_mod()'s fitted_residuals attribute now covers all observations across all strata (previously only the last stratum was retained) and always includes a Stratification column ("None" when unstratified). (Debug calcfitmod #240)

🟢 Documentation

@param strat is now in both R/calc_fit_mod.R and man/calc_fit_mod.Rd ✅
@returns documents the new Stratification field and the "None" default ✅
man/run_mod.Rd regenerated consistently with the roxygen ✅

🟢 Tests / Snapshots

All R-CMD-check jobs (ubuntu/macOS/windows) pass on the latest commit (7c41296).
Both darwin/ and root snapshot variants are present and include the new Stratification column.
The run_mod.md deparse snapshots updated for the new attribute encoding.

Verdict

Approve-with-nits. The fix is correct, well tested, and CI is green on every platform. Items (1)–(3) and (6) are quality-of-life follow-ups; (5) is whitespace. None block merge.
• Branch: debug_calcfitmod

…namics into debug_calcfitmod

Merge commit '77dc07e6a01e267a2ad18533a359e630b83dc8e8' #Conflicts: # tests/testthat/_snaps/run_mod.md

sschildhauer · 2026-05-29T00:07:38Z

@claude, please implement all suggestions. Keep all edits within in this thread.

claude · 2026-05-29T00:10:49Z

claude · 2026-06-08T21:45:15Z

Claude finished @d-morrison's task in 1m 17s —— View job

Code Review: PR #240 — Debug calcfitmod

Read repo guidelines / CLAUDE.md (none at root)
Read current state of R/calc_fit_mod.R, R/Run_Mod.R
Review documentation, snapshots, WORDLIST
Check CI status for latest commit
Post final review

Summary

CI on a15dd19 is fully green on every substantive workflow — R-CMD-check, pkgdown, NEWS check, Version increment, Documentation, Spellcheck, Lint, check-readme, test-coverage all pass. The one reported "failure" is the Claude Code Review workflow itself (it is currently configured to fail on bot-authored commits, as @d-morrison noted — that's a CI-infrastructure issue, not a code issue, and unrelated to merge-readiness).

The two-part fix lands cleanly:

Before	After
`calc_fit_mod()` got `dl_sub` (per-stratum filtered); the last iteration silently overwrote prior strata in `fitted_residuals`	`calc_fit_mod()` gets full `data` + `strat`, merges once on `(Subject, Iso_type[, Stratification])`
No `Stratification` column in `fitted_residuals`	`Stratification` always present (`"None"` when unstratified)

Correctness — looks good

R/Run_Mod.R:237-239 — calc_fit_mod(modeled_dat = jags_out, original_data = data, strat = strat). Passing the full data (not the per-stratum filtered dl_sub) is the minimal, correct change. The per-stratum filter loop above only governs modeling; fitted/residual computation is now decoupled and runs once over everything.
R/calc_fit_mod.R:60-68 — merge(..., all.y = TRUE) on (Subject, Iso_type) or (Subject, Iso_type, Stratification) preserves every observation, surfacing un-modeled rows with NA fitted/residual. Documented at R/calc_fit_mod.R:24-27.
R/calc_fit_mod.R:44-47 — the if (strat != "Stratification") guard correctly handles the edge case where the user's strat column is already named Stratification.
Snapshots verified:
- strat-fitted_residuals.csv contains both typhi and paratyphi (no longer last-stratum-only) — confirmed via cut.
- sim-strat-fitted_residuals.csv contains both stratum 1 and stratum 2.
- nostrat-fitted_residuals.csv uses None exactly as documented.
- All snapshots have header Subject,Iso_type,Stratification,t,fitted,residual, matching R/calc_fit_mod.R:75-76.
- darwin/ and root variants are present with identical schemas and row counts (905 / 905 / 1134).

Documentation — looks good

@param strat present in both R/calc_fit_mod.R:10-11 and man/calc_fit_mod.Rd:15-16 ✅
@returns documents the Stratification field, the "None" default, and the NA-strat behavior ✅
R/Run_Mod.R:54 and R/Run_Mod.R:87 now correctly use `"None"` (quoted), matching R/calc_fit_mod.R:17 and the regenerated man/run_mod.Rd ✅
NEWS entry (NEWS.md:97) accurately captures both behavioral changes ✅
inst/WORDLIST: allowlist restored, calc retained ✅

Findings

No new blockers and no outstanding nits. Every prior review item — full-data fitted/residual computation, Stratification column in output, NA-strat documentation, name-collision guard, strat = NA default, `None` → `"None"` quoting harmonization, NEWS rewording, WORDLIST cleanup — has been addressed across the iterations of this thread.

Verdict

Approve. The fix is correct, well-tested, and CI is green across all platforms. Ready to merge.

Note on the failing Claude Code Review workflow

To @d-morrison's point above: the auto-triggered Claude Code Review workflow has been failing on bot-authored commits in this PR. That's a workflow-config behavior, not a code-quality signal. Manually invoking @claude review (as you did just now) is the correct workaround. If you'd like the auto-review to succeed on bot commits too, the workflow file in .github/workflows/ would need to be updated to either (a) allowlist the bot author or (b) skip on bot-authored pushes rather than failing — happy to outline a patch if you want, but I can't modify .github/workflows/* directly due to App permissions.

Branch: debug_calcfitmod — Commit: a15dd19

d-morrison

looking good; please address inline comments

d-morrison · 2026-06-08T22:50:57Z

@sschildhauer PS - I just updated the main branch to hopefully catch tidyselect/data-masking errors going forward (#244). I made some changes to the main branch of calc_fit_mod in order to keep the main branch passing; sorry about the resulting merge conflict.

d-morrison · 2026-06-08T23:02:04Z

Heads-up about an incoming merge conflict: PR #244 (targeting main) de-deprecates the four .data-in-tidyselect sites in R/calc_fit_mod.R — converting the select(), .by, and pivot_wider() calls to all_of()/string selection — and turns on options(lifecycle_verbosity = "error") in the test suite so this class of soft deprecation fails CI going forward. Because this PR rewrites the same file, expect a conflict in calc_fit_mod.R when you next merge main in. The any_of()/join rewrite on this branch supersedes #244's minimal fix, so resolve the conflict in favor of this branch's version. (Posted from a Claude Code local session.)

Co-authored-by: Douglas Ezra Morrison <demorrison@ucdavis.edu>

Merge commit '0875741138c0195035f9a74ce8465b68f5fd6d58' #Conflicts: # R/calc_fit_mod.R

sschildhauer · 2026-06-09T22:28:53Z

@claude, please implement changes that will allow this PR to pass review.

claude · 2026-06-09T22:32:49Z

Claude finished @sschildhauer's task in 13m 55s —— View job

Implementing review changes — PR #240

Pushed as 15a66c2. All of d-morrison's inline review comments are addressed:

What the refactor looks like

calc_fit_mod <- function(modeled_dat,
                         original_data,
                         strat = NA) {
  strat_col <- if (is.na(strat)) character() else c(Stratification = strat)

  original_data <- original_data |>
    use_att_names() |>
    dplyr::select(
      dplyr::any_of(c("Subject", "Iso_type", "t", "result", strat_col))
    )

  modeled_dat <- modeled_dat |>
    dplyr::summarize(
      .by = dplyr::all_of(
        c("Parameter", "Iso_type", "Stratification", "Subject")
      ),
      med_value = stats::median(.data$value)
    ) |>
    tidyr::pivot_wider(names_from = "Parameter", values_from = "med_value")

  matched_dat <- modeled_dat |>
    dplyr::right_join(
      original_data,
      by = base::intersect(
        c("Subject", "Iso_type", "Stratification"),
        names(original_data)
      )
    )

  fitted_dat <- matched_dat |>
    dplyr::mutate(
      fitted = ab(.data$t, .data$y0, .data$y1, .data$t1,
                  .data$alpha, .data$shape),
      residual = .data$result - .data$fitted
    ) |>
    dplyr::select(
      dplyr::all_of(
        c("Subject", "Iso_type", "Stratification", "t", "fitted", "residual")
      )
    )

  return(fitted_dat)
}

Notes

I empirically verified that dplyr::any_of(c(..., "Stratification" = NA_character_)) raises "Selections can't have missing values" — so the named-vector trick has to be guarded with a one-line if/else on strat_col (the small assignment, not a branch around the select() call itself, which is what CLAUDE.md and d-morrison's comment target).
base::intersect() over the canonical key triplet lets right_join() pick the join columns from names(original_data), eliminating the if ("Stratification" %in% ...) branch entirely.
The Linux snapshot diff was row order only, not numeric drift — verified by sorting both and confirming all.equal() after sort. The darwin snapshots were already in the correct numerical order from 636200f/a1c35a9; the root variants had been regenerated in lex order by 63db4f3 and just hadn't caught up.
• Branch: debug_calcfitmod • Commit: 15a66c2

Combine the rename + select into a single select(any_of(...)) call using a named-vector entry for the optional Stratification column, switch the .by argument and final select to all_of(), and replace the by_vars if/else + right_join with right_join(by = intersect(...)) so the join key picks up Stratification only when it exists in original_data. Adds explicit return(), dplyr:: prefixes on the final mutate/select, and the safe strat = NA default. Also regenerate the stale Linux (non-darwin) fitted_residuals snapshots, which were in lex Subject order from an earlier regeneration; the right_join now produces the same numerical Subject order as the darwin snapshots, so the previous-pass test failures resolve. Co-authored-by: Samuel Hisaji Schildhauer <sschildhauer@users.noreply.github.com>

sschildhauer · 2026-06-09T22:59:02Z

@claude, review.

claude · 2026-06-09T23:02:23Z

Claude finished @sschildhauer's task in 1m 57s —— View job

Code Review: PR #240 — Debug calcfitmod

Read CLAUDE.md and current state of changed files
Check CI status and diff vs origin/main
Verify d-morrison's inline comments are addressed
Post review findings

Summary

The refactor in 15a66c2 lands cleanly on top of the recent main (which now contains #244). CI is fully green on the latest commit — 9/9 substantive workflows passing (R-CMD-check, pkgdown, NEWS check, Version increment, Documentation, Spellcheck, Lint, check-readme, test-coverage). The 10th workflow (Claude Code Review) is skipped (bot-author skip rule), not a failure.

Every one of d-morrison's five inline comments is addressed:

Comment	Resolution
Use `any_of()` with named-vector rename instead of `if`/`else`	`R/calc_fit_mod.R:32-38` — `strat_col <- if (is.na(strat)) character() else c(Stratification = strat)` + `any_of(c(..., strat_col))` ✅
Do the rename inside `select()`	Same — rename + select collapsed to one step ✅
`.by` is a tidy-selection context	`R/calc_fit_mod.R:43` — `.by = dplyr::all_of(...)` ✅
Use `dplyr::*_join` + tidyselect to avoid `if`/`else` on join keys	`R/calc_fit_mod.R:51-58` — `right_join(by = base::intersect(c("Subject","Iso_type","Stratification"), names(original_data)))` ✅
"ditto" on `pivot_wider`	`R/calc_fit_mod.R:48` — `names_from = "Parameter", values_from = "med_value"` (string literals) ✅

Correctness — looks good

R/Run_Mod.R:237-239 — calc_fit_mod(modeled_dat = jags_out, original_data = data, strat = strat). Full dataset (not per-stratum dl_sub) ➜ no more last-stratum-overwrite. Correct.
R/calc_fit_mod.R:32 — strat_col construction is sound across all three cases I checked: NA → character() (no-op in select); "Stratification" → c(Stratification = "Stratification") (degenerate rename, harmless); arbitrary name → c(Stratification = <name>) (rename to canonical). Cleanly handles the name-collision case without an explicit guard.
R/calc_fit_mod.R:51-58 — base::intersect(c("Subject","Iso_type","Stratification"), names(original_data)) selects join keys based on what's actually in original_data, which itself only contains Stratification when strat was non-NA. Branch eliminated entirely. Good.
R/calc_fit_mod.R:68 — right_join (vs prior merge(…, all.y = TRUE)) preserves every observation in original_data; NA-strat rows surface with NA fitted/residual. Documented at lines 24-27.
Snapshots verified:
- strat-fitted_residuals.csv contains both typhi and paratyphi (905 rows).
- sim-strat-fitted_residuals.csv contains both stratum 1 and stratum 2 (1134 rows).
- nostrat-fitted_residuals.csv uses None exactly as documented (905 rows).
- Header Subject,Iso_type,Stratification,t,fitted,residual matches R/calc_fit_mod.R:67-70.
- darwin/ and root variants have identical schemas and row counts; numeric drift between platforms is normal JAGS variance.

Documentation — looks good

@param strat documented at R/calc_fit_mod.R:10-11 and man/calc_fit_mod.Rd ✅
@returns covers the Stratification field, the "None" default, and NA-strat row behavior ✅
R/Run_Mod.R:54 / :87 use the harmonized `"None"` quoting; man/run_mod.Rd regenerated cleanly ✅
NEWS entry (NEWS.md:116) accurately captures both behavioral changes ✅
DESCRIPTION bumped to 0.0.0.9058 ✅
inst/WORDLIST: allowlist restored, calc retained — both warranted ✅

Style — looks good

Lab-manual / tidyverse-style: dplyr:: prefixes on all dplyr calls, all_of()/any_of() in every tidy-selection context, no .data pronoun in selection contexts, explicit return(fitted_dat), no if/else whose sole purpose is to vary columns.
The refactor exactly matches the patterns required by CLAUDE.md (data-masking in tidy-selection, branching-that-only-varies-columns, base merge() → dplyr joins, near-duplicate stratified/unstratified paths).

Findings

No blockers and no nits. All prior items across the review thread are now resolved.

Verdict

Approve. The fix is correct, well-documented, well-tested, and CI is green across every substantive workflow. d-morrison's inline comments are all addressed in the most recent refactor. Ready to merge.

Branch: debug_calcfitmod — Commit: 15a66c2

sschildhauer · 2026-06-09T23:06:55Z

@d-morrison , Thank you for calling out those changes. I will keep track of the new dplyr features in the future.

claude · 2026-06-14T01:18:45Z

Claude finished @d-morrison's task in 2m 5s —— View job

Code Review: PR #240 — Debug calcfitmod

Read CLAUDE.md and copilot-instructions.md
Read current state of R/calc_fit_mod.R and R/Run_Mod.R
Diff vs origin/main
Review documentation and snapshots
Post review findings

Summary

The two-part fix lands correctly and the refactor is now fully idiomatic per CLAUDE.md and d-morrison's inline review:

Before	After
`calc_fit_mod()` got `dl_sub` (per-stratum filtered); last iteration silently overwrote prior strata in `fitted_residuals`	`calc_fit_mod()` gets full `data` + `strat`, merges once on `(Subject, Iso_type[, Stratification])`
No `Stratification` column in `fitted_residuals`	`Stratification` always present (`"None"` when unstratified)

CI is 5/5 complete workflows passing (NEWS check, Documentation check, Spellcheck, check-readme, lint-changed-files); 4 workflows still in progress (R-CMD-check, test-coverage, version-check, pkgdown).

Correctness — looks good

R/Run_Mod.R:237-239 — calc_fit_mod(modeled_dat = jags_out, original_data = data, strat = strat) passes the full dataset, decoupling fitted/residual computation from the per-stratum modeling loop. The core bug fix is correct.

R/calc_fit_mod.R:32 — strat_col construction handles all three cases cleanly:

strat = NA → character() → no rename in any_of(), no Stratification key in the join
strat = "Stratification" → c(Stratification = "Stratification") → degenerate rename, harmless
strat = "foo" → c(Stratification = "foo") → renames foo → Stratification in one step

R/calc_fit_mod.R:51-58 — base::intersect(c("Subject", "Iso_type", "Stratification"), names(original_data)) cleanly eliminates the if ("Stratification" %in% ...) branch. Since original_data only gains a Stratification column when strat is non-NA (controlled by any_of(strat_col) on line 37), intersect() does exactly the right thing in both cases.

R/calc_fit_mod.R:67-71 — dplyr::select(dplyr::all_of(c("Subject", "Iso_type", "Stratification", "t", "fitted", "residual"))) works for both stratified and unstratified cases: in the unstratified path, Stratification = "None" comes from modeled_dat via the right join. Correct.

CLAUDE.md idioms — all addressed

CLAUDE.md requirement	Status
`.by` uses `all_of()` (tidy-selection context)	`R/calc_fit_mod.R:43` — `.by = dplyr::all_of(...)` ✅
`any_of()` with named vector for rename-in-select	`R/calc_fit_mod.R:36-38` — `any_of(c(..., strat_col))` ✅
No `if/else` whose sole purpose is to vary columns	`strat_col` named-vector + `intersect()` eliminates both branches ✅
`dplyr::` joins over `base::merge()`	`dplyr::right_join()` ✅
Consistent `dplyr::` namespace prefixes	All dplyr calls prefixed ✅
Explicit `return()`	`return(fitted_dat)` at `R/calc_fit_mod.R:73` ✅
`strat = NA` default (not self-referential)	`R/calc_fit_mod.R:31` ✅

Documentation — looks good

@param strat documented at R/calc_fit_mod.R:10-11 and man/calc_fit_mod.Rd:15-16 ✅
@returns covers the Stratification field, the "None" default, and NA-strat row behavior ✅
R/Run_Mod.R:54 / :87 use harmonized `"None"` quoting; man/run_mod.Rd regenerated cleanly ✅
NEWS entry (NEWS.md:116) accurately captures both behavioral changes ✅
DESCRIPTION bumped to 0.0.0.9058 ✅
inst/WORDLIST: allowlist restored, calc retained (both warranted) ✅

Findings

🟡 Minor — unstratified unmatched rows get `Stratification = NA`, not `"None"` (edge case, non-blocking)

In the unstratified path, modeled_dat carries Stratification = "None". After the right_join, rows that are in original_data but have no matching posterior estimate (if any) will get Stratification = NA (from the missing left-side value) rather than "None". In practice this cannot happen when run_mod() runs successfully — every subject is modeled — so this is a theoretical edge case. The documented behavior at R/calc_fit_mod.R:24-27 covers NA-strat rows, and the snapshots confirm correct output. Not a blocker.

Verdict

Approve. The fix is correct, well-documented, and fully idiomatic per CLAUDE.md. All d-morrison's inline review comments from the previous round are resolved. Lint, spellcheck, docs, and NEWS all pass; R-CMD-check is still running but the code is clean. Ready to merge once CI completes.

Branch: debug_calcfitmod •

sschildhauer added 2 commits May 27, 2026 15:10

Updating calc_fit_mod to include stratification. Fixing bug in run_mo…

7d419ac

…d input data for calc_fit_mod

Updating snaps for the new fitted attributes that includes all ids

03aab15

This comment has been minimized.

Sign in to view

Increment version number to 0.0.0.9056

e186e72

sschildhauer added 2 commits May 27, 2026 16:04

Updating calc_fit_mod to merge based on Stratification

2c91b0c

Updating wordlist

ef3429e

sschildhauer added 2 commits May 27, 2026 16:28

Debugging calc_fit_mod to match on start only if is.na()==F

3b65173

Redoing snaps

d086af6

github-actions Bot and others added 3 commits May 28, 2026 00:27

Specifying documentation regarding stratification

d087e8a

Add missing @param strat documentation to calc_fit_mod

7c41296

The new strat parameter added in this PR was not documented with an @param tag, causing an undocumented-argument gap in calc_fit_mod.Rd. Co-authored-by: Samuel Hisaji Schildhauer <undefined@users.noreply.github.com>

sschildhauer requested a review from Kwan-Jenny May 28, 2026 01:05

This was linked to issues May 28, 2026

Bug: calc_fit_mod uses only the last stratum's data (R/Run_Mod.R:268–271) #215

Closed

Fix comments on calc_fit_mod #226

Closed

github-actions Bot removed the request for review from Kwan-Jenny May 28, 2026 04:27

github-actions Bot requested a review from Kwan-Jenny May 28, 2026 04:29

sschildhauer added 3 commits May 28, 2026 16:47

Adding default NA for strat

525ac51

Merge branch 'debug_calcfitmod' of https://github.com/UCD-SERG/serody…

a54a129

…namics into debug_calcfitmod

Merging with main

b8311d7

Merge commit '77dc07e6a01e267a2ad18533a359e630b83dc8e8' #Conflicts: # tests/testthat/_snaps/run_mod.md

github-actions Bot removed the request for review from Kwan-Jenny May 29, 2026 00:06

github-actions Bot requested review from Kwan-Jenny and removed request for Kwan-Jenny May 29, 2026 00:10

github-actions Bot requested a review from d-morrison June 8, 2026 21:46

d-morrison requested changes Jun 8, 2026

View reviewed changes

Comment thread R/calc_fit_mod.R Outdated

Comment thread R/calc_fit_mod.R Outdated

Comment thread R/calc_fit_mod.R Outdated

Comment thread R/calc_fit_mod.R Outdated

Comment thread R/calc_fit_mod.R Outdated

d-morrison mentioned this pull request Jun 8, 2026

ci+fix: strict lifecycle deprecations, idiom review guidance, bot-skip review #244

Merged

d-morrison mentioned this pull request Jun 8, 2026

docs(data-masking): add tidy-selection vs data-masking section UCD-SERG/lab-manual#297

Open

sschildhauer and others added 7 commits June 9, 2026 09:47

Update R/calc_fit_mod.R

75d4675

Co-authored-by: Douglas Ezra Morrison <demorrison@ucdavis.edu>

Merging with main and dplyr changes

9c00db6

Merge commit '0875741138c0195035f9a74ce8465b68f5fd6d58' #Conflicts: # R/calc_fit_mod.R

Fixing calc_fit_mod dplyr references

d31e17d

Reordering output fitted_residuals

a1c35a9

Fitted_residuals now orderds IDs correctly

636200f

Increment version number to 0.0.0.9057

84ec519

Fixing lint in calc_fit_mod

73bb0fe

sschildhauer requested a review from d-morrison June 9, 2026 23:07

Merge branch 'main' into debug_calcfitmod

5f0871b

github-actions Bot removed the request for review from d-morrison June 14, 2026 01:18

github-actions Bot requested a review from d-morrison June 14, 2026 01:21

claude Bot mentioned this pull request Jun 14, 2026

Document required relationship arg on dplyr joins #248

Merged

d-morrison approved these changes Jun 15, 2026

View reviewed changes

d-morrison merged commit 89b70c6 into main Jun 15, 2026
17 checks passed

d-morrison deleted the debug_calcfitmod branch June 15, 2026 07:12

claude Bot mentioned this pull request Jun 16, 2026

Prepare v0.1.0 for first CRAN submission #207

Merged

15 tasks

Conversation

sschildhauer commented May 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

This comment has been minimized.

github-actions Bot commented May 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

codecov Bot commented May 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

sschildhauer commented May 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

claude Bot commented May 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Fixed: regenerated the missing Linux snapshots

Notes

Uh oh!

sschildhauer commented May 28, 2026

Uh oh!

claude Bot commented May 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Code Review: PR #240 — Debug calcfitmod

Summary

Findings

🟢 Correctness — looks good

🟡 Suggestions (non-blocking)

🟢 Documentation

🟢 Tests / Snapshots

Verdict

Uh oh!

sschildhauer commented May 29, 2026

Uh oh!

claude Bot commented May 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review suggestions implemented — pushed as 97a76da

Notes

Uh oh!

claude Bot commented Jun 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Code Review: PR #240 — Debug calcfitmod

Summary

Correctness — looks good

Documentation — looks good

Findings

Verdict

Note on the failing Claude Code Review workflow

Uh oh!

d-morrison left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

d-morrison commented Jun 8, 2026

Uh oh!

d-morrison commented Jun 8, 2026

Uh oh!

sschildhauer commented Jun 9, 2026

Uh oh!

claude Bot commented Jun 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Implementing review changes — PR #240

What the refactor looks like

Notes

Uh oh!

sschildhauer commented Jun 9, 2026

Uh oh!

claude Bot commented Jun 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Code Review: PR #240 — Debug calcfitmod

Summary

Correctness — looks good

Documentation — looks good

Style — looks good

Findings

sschildhauer commented May 27, 2026 •

edited

Loading

github-actions Bot commented May 27, 2026 •

edited

Loading

codecov Bot commented May 27, 2026 •

edited

Loading

sschildhauer commented May 28, 2026 •

edited

Loading

claude Bot commented May 28, 2026 •

edited

Loading

claude Bot commented May 28, 2026 •

edited

Loading

claude Bot commented May 29, 2026 •

edited

Loading

Review suggestions implemented — pushed as `97a76da`

claude Bot commented Jun 8, 2026 •

edited

Loading

claude Bot commented Jun 9, 2026 •

edited

Loading

claude Bot commented Jun 9, 2026 •

edited

Loading

claude Bot commented Jun 14, 2026 •

edited

Loading

🟡 Minor — unstratified unmatched rows get `Stratification = NA`, not `"None"` (edge case, non-blocking)