Design and execute evaluation benchmarks with statistical tests and ablation studies.
npx skills add https://github.com/Tnemo65/template --skill evaluation-tnemo65
Please help me install this Agent Skill. Skill: evaluation Source: https://github.com/Tnemo65/template/tree/main/.cursor/skills/07-ml/evaluation Command: npx skills add https://github.com/Tnemo65/template --skill evaluation-tnemo65
Evaluation planning and execution for research systems to assess performance, reliability, and significance.
Formulate hypotheses, select appropriate tests, run evaluations, and generate a report ready for publication.
High-intent search queries and answers about installing and using this skill.
Design statistical tests for research evaluation by formulating hypotheses and selecting appropriate tests like Wilcoxon, t-test, or Friedman. This Skill helps you execute rigorous benchmarks and generate publication-ready reports with clear confidence intervals and effect sizes.
Run ablation studies by applying design templates that measure individual component contributions to your research system. This Skill identifies and executes ablation benchmarks to isolate performance impacts and generates reproducible result interpretations.
Calculate bootstrap confidence intervals for benchmark results by executing reproducible evaluation runs across experiments. This Skill supports bootstrap confidence intervals to quantify result reliability and produces publication-ready reporting.
Generate publication-ready reports with effect sizes and statistical significance by running evaluations and interpreting results. This Skill formats benchmark outputs with clear confidence intervals and effect sizes suitable for publication.
Use the Friedman test instead of a t-test for evaluation when comparing multiple systems across multiple datasets. This Skill supports selecting appropriate statistical tests based on your hypothesis testing requirements.
This evaluation approach works for reproducibility in research benchmarks by applying standardized benchmark execution and bootstrap confidence intervals. It ensures reliable result reporting with clear confidence intervals and effect sizes.
ksopylaGuide hypothesis selection, baselines, and decision criteria for Concept Encoder research.
debug-zhuweijianAutomate research results into publication-grade figures and tables with pubfig/pubtab.
root-5Clean and standardize messy datasets using pandas and numpy.
kollaboraiEnforce assumption-verified statistical workflows with effect sizes and confidence intervals.
