table-generation
Generate publication-quality LaTeX tables from experimental results. Convert JSON/CSV data to booktabs-styled tables with bold best results, multi-row layouts, and proper captions. Use when creating result tables, comparison tables, or ablation tables for papers.
Install
npx skills add https://github.com/lingzhi227/agent-research-skills --skill table-generationSKILL.md
Table Generation
Convert experimental results into publication-ready LaTeX tables.
Input
$0— Table type:comparison,ablation,descriptive,custom$1— Data source: JSON file, CSV file, or inline data
Scripts
Generate LaTeX table from JSON/CSV
python ~/.claude/skills/table-generation/scripts/results_to_table.py \
--input results.json --type comparison \
--bold-best max --caption "Performance comparison" \
--label tab:main_results
Supports: comparison, ablation, descriptive, multi-dataset table types.
Additional flags: --type multi-dataset for methods x datasets x metrics layout, --significance for p-value stars, --underline-second for second-best results.
References
- LaTeX table templates and examples:
~/.claude/skills/table-generation/references/table-templates.md
Table Types
comparison — Main results table
- Rows = methods (baselines + ours), Columns = metrics/datasets
- Bold the best result in each column
- Include mean +/- std when available
- Use
\multirowfor method categories (Supervised, Self-supervised, etc.)
ablation — Ablation study table
- Rows = variants (full model, minus component A, minus component B, ...)
- Columns = metrics
- Bold the full model result
- Use checkmarks for component presence
descriptive — Dataset/statistics table
- Dataset characteristics, hyperparameters, or summary statistics
- Clean formatting with proper units
custom — Free-form table
- User specifies layout and content
Required Packages
\usepackage{booktabs} % \toprule, \midrule, \bottomrule
\usepackage{multirow} % \multirow
\usepackage{multicol} % multi-column layouts
\usepackage{threeparttable} % table notes
Output Format
Always generate tables with:
booktabsrules (\toprule,\midrule,\bottomrule)\caption{}and\label{tab:...}- Bold best results using
\textbf{} - Table notes via
threeparttablewhen needed - Proper alignment (
lfor text,corrfor numbers)
Rules
- Only include numbers from actual experimental logs — never hallucinate results
- All numbers must match the data source exactly
- Use
$\pm$for standard deviations - Use
\underline{}for second-best results when appropriate - Keep tables compact — avoid unnecessary columns
- Use
table*for wide tables spanning two columns - Add glossary/notes for abbreviated column headers
Related Skills
- Upstream: data-analysis, experiment-code
- Downstream: paper-writing-section, paper-compilation
- See also: figure-generation
Related skills
researchmattpocock575KInvestigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.paper-context-resolverlllllllama451KRigor Paper Context helper for README-first deep learning repo reproduction. Use only when the README and repository files leave a narrow reproduction-critical gap and the task is to resolve a specific paper detail such as dataset split, preprocessing, evaluation protocol, checkpoint mapping, or runtime assumption from primary paper sources while recording conflicts. Do not use for general paper summary, repo scanning, environment setup, command execution, title-only paper lookup, or replacing Renv-and-assets-bootstraplllllllama450KRigor Setup skill for README-first deep learning repo reproduction. Use when the task is specifically to prepare a conservative conda-first environment, checkpoint and dataset path assumptions, cache location hints, and setup notes before any run on a README-documented repository. Do not use for repo scanning, full orchestration, paper interpretation, final run reporting, or generic environment setup that is not tied to a specific reproduction target.ai-research-explorelllllllama311KRigor Explore compatible skill slug for meaningful and potentially novel deep learning research candidates. Use when the researcher has chosen the task family, dataset, benchmark, evaluation method, provided SOTA references, and wants candidate-only exploration on top of `current_research` with auditable repo understanding, idea gating, fair comparison, and governed experiments written to `explore_outputs/`. Do not use for README-first trusted reproduction, open-ended direction finding, narrow c