メニュー
Rigor for clustering, dimensionality reduction, and anomaly detection: validity is stability under resampling, not a held-out score.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Establish a dumb baseline and an honest validation harness before any real model, so every later number means something.
Rigor for causal questions and A/B tests (the effect of acting on X): confounding, post-treatment bias, valid control groups.
Rigor for descriptive and diagnostic analytics (what happened and why): denominators, grain, and confounded slices, not model leakage.
EDA kickoff for a fresh dataset: fixed opening scaffold (goal, imports, load, sanity), then chapters derived from the data; scan leakage, land a baseline.
Rigor for statistical inference (is the difference real): hypothesis tests, power, multiple comparisons, effect size over p-value.
Audit a dataset or pipeline for the five leakages that inflate a metric: target, preprocessing, temporal, group, and sampling.
| name | lemma-unsupervised |
| description | Rigor for clustering, dimensionality reduction, and anomaly detection: validity is stability under resampling, not a held-out score. |
| homepage | https://github.com/tkpratardan/lemma |
| license | MIT |
Provide the discovered structure or anomalies, the preprocessing and selection choices that produced them, stability evidence, useful interpretation, and clear limits on what the pattern means.
Do not treat an attractive projection as validation, select cluster count from one index alone, or assign causal or essential meaning to algorithmic groups.
For stability and interpretation patterns, read references/deep-guide.md.
