Skip to content

Copyedit the documentation, correct the pancreas data description, rbeam plot changes - #4

Merged
imallona merged 4 commits into
mainfrom
docs-copyedit-and-plots
Oct 7, 2026
Merged

imallona merged 4 commits into
mainfrom
docs-copyedit-and-plots

Conversation

@imallona

@imallona imallona commented Oct 7, 2026 •

Copy link
Copy Markdown
Owner

This is a copyedit of the documentation, with one correction to the data description and a few changes to the rbeam plots.

  • The README, site index, contributing guide, DESCRIPTION, explanations, how-tos, tutorial and vignettes have shorter sentences.
  • The metric card how-to now matches the code on the optional card fields, on the check for an inverted range, and on the OWL classes of a card.
  • Tran's Dataset 4 has five pancreas studies and the scIB task has six, four of them in common. The text said both used the same five. No numbers change.
  • beam_rank_dots replaces beam_rank_bars. Calls to the old name will fail.
  • The attribution plot uses the colours of the rank sensitivity plot.
  • beam_funky_heatmap(size_legend = TRUE) adds a key for the circle size.
  • The reliability plot labels its two reference lines.
  • The labels of the forest and critical difference plots fit inside the panel.
  • Axis labels say "fraction" where they said "share".
  • .gitignore lists the R build leftovers and the generated man pages.

pytest gives 893 passed and 3 skipped, and ruff check is clean. I did not run R CMD check locally.

…rence lines

Replace beam_rank_bars with beam_rank_dots, which draws one dot per
series joined by a line so that reversed axes do not make the longest
bar the worst method. Colour the attribution bars with the role palette
shared by the sensitivity and variance plots. Add a size key option to
the funky heatmap, label the group alpha and the bound in the
reliability plot, and keep the P-score and method labels of the forest
and critical difference plots inside the panel.
Plain sentences in the README, site index, contributing guide, package
description, explanations, how-tos, tutorial and vignettes. Correct
three statements that disagreed with the code: the optional card fields
in the metric card how-to, the range check (card consistency audit, not
the schema) and the OWL classes. Tran dataset 4 and the scIB pancreas
task share four of their studies; the data README, the loader docstring
and the cross-benchmark vignettes no longer call them the same data.
Ignore R build leftovers and the generated man pages.
Axis labels say fraction instead of share, the specification curve
names the method shown, the reliability plot has room above the group
alpha line and puts the bound label in the facet with the lowest bars.

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copilot review overview

🟡 Changes recommended

The inverted-range guidance contradicts runtime behavior, and the new size-legend branch lacks test coverage.

Review effort: Balanced
Findings: 1 Medium severity · 1 Low severity

Open (2)
What changed in this PR

This PR copyedits beam’s documentation, corrects the pancreas benchmark description, and updates rbeam visualizations.

Changes:

  • Clarifies documentation and the partial overlap between pancreas datasets.
  • Replaces beam_rank_bars with beam_rank_dots and improves plot labels, colors, legends, and spacing.
  • Updates tests, exports, changelog, and ignored R build artifacts.
File Description
src/​beam/​plot.py Renames variance proportions as fractions.
src/​beam/​datasets.py Corrects pancreas benchmark overlap documentation.
src/​beam/​data/​README.md Documents the actual pancreas studies.
README.md Reworks project and usage descriptions.
r/​beam/​vignettes/​ranking.Rmd Clarifies blinding terminology.
r/​beam/​vignettes/​metric-diagnostics.Rmd Clarifies blinding terminology.
r/​beam/​vignettes/​heterogeneity.Rmd Clarifies blinding terminology.
r/​beam/​vignettes/​cross-benchmark.Rmd Corrects pancreas overlap and related wording.
r/​beam/​vignettes/​cell-type-annotation.Rmd Clarifies blinding terminology.
r/​beam/​tests/​testthat/​test-plots.R Updates the rank-plot smoke test.
r/​beam/​R/​plot_kinds.R Improves labels, colors, annotations, and clipping.
r/​beam/​R/​plot_grouped.R Replaces grouped rank bars with dots and ranges.
r/​beam/​R/​plot_builders.R Adds custom stacked-plot palettes.
r/​beam/​R/​funky_heatmap.R Adds an optional circle-size legend.
r/​beam/​NAMESPACE Exports beam_rank_dots.
r/​beam/​DESCRIPTION Rewrites the package description.
index.qmd Reworks the documentation landing page.
examples/​transportation/​transportation.qmd Clarifies MEREC’s zero handling.
examples/​m4/​m4.qmd Updates variance terminology.
examples/​duo2018/​duo2018.qmd Clarifies pooling and missing-data handling.
examples/​cross_benchmark/​cross_benchmark.qmd Corrects pancreas overlap and terminology.
examples/​cell_type_annotation/​cell_type_annotation.qmd Clarifies dataset coverage and comparisons.
docs/​tutorials/​quickstart.md Copyedits introductory instructions.
docs/​README.md Clarifies the directory’s purpose.
docs/​metric-cards.qmd Documents optional card normalization.
docs/​how-to/​run.md Expands manifest documentation.
docs/​how-to/​run-heterogeneity-from-the-cli.md Clarifies required long-format input.
docs/​how-to/​install.md Clarifies R installation constraints.
docs/​how-to/​blind-an-analysis.md Clarifies method-label masking.
docs/​how-to/​add-a-metric-card.md Updates optional fields and validation guidance.
docs/​explanations/​weighting-schemes.md Copyedits CRITIC and MEREC descriptions.
docs/​explanations/​reference-levels.md Clarifies reference-level checks.
docs/​explanations/​rank-sensitivity.md Clarifies failure and missing-data behavior.
docs/​explanations/​pairwise-method-comparison.md Copyedits implementation attribution.
docs/​explanations/​ontology-mappings.md Clarifies URI requirements.
docs/​explanations/​normalization-and-scales.md Clarifies normalization errors and audits.
docs/​explanations/​missing-data.md Updates missing-cell terminology.
docs/​explanations/​method-by-dataset-heterogeneity.md Clarifies tied ranks.
docs/​explanations/​funky-heatmaps-and-robustness.md Clarifies caller-supplied panels.
docs/​explanations/​dataset-concordance-and-discrimination.md Copyedits ranking behavior.
docs/​explanations/​cards-and-pipeline.qmd Clarifies range-validation errors.
docs/​explanations/​card-data-consistency.md Clarifies consistency-audit behavior.
docs/​explanations/​attribution-synthesis.md Copyedits attribution descriptions.
docs/​explanations/​analysis-blinding.md Clarifies seal handling.
CONTRIBUTING.md Simplifies contribution guidance.
CHANGELOG.md Records the rbeam plotting changes.
.gitignore Ignores R build and generated documentation files.

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

Comment thread r/beam/R/funky_heatmap.R
Comment on lines +54 to +55
show_aggregation = TRUE, cliques = NULL,
size_legend = FALSE, ...) {
Comment thread docs/how-to/add-a-metric-card.md Outdated
```

The schema check runs against every card in the registry. A missing required field, a polarity that does not match the polarity enum, or an inverted range (lower > upper) raises an error.
The schema check runs against every card in the registry. A missing required field or a polarity outside the polarity enum raises an error. An inverted range (lower > upper) passes the schema check; `beam.rank` reports it as a violation in the [card-data consistency report](../explanations/card-data-consistency.md).
@imallona
imallona merged commit cc5893e into main Oct 7, 2026
10 checks passed
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