원클릭으로 Manus에서 모든 스킬 실행

lemma-baseline

스타4

포크1

업데이트2026년 7월 15일 00:47

Establish a dumb baseline and an honest validation harness before any real model, so every later number means something.

설치

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

Manus에서 실행

출처

tkpratardan

tkpratardan/lemma

GitHub 저장소 열기 Creator 저장소 보기

다운로드

Manus에서 실행

Baseline

Deliver

Provide a reproducible validation design, a no-information or rule baseline, the simplest credible model, and the precise score later models must beat.

Check

Define the outcome, observation unit, prediction moment, intended use, and a fixed seed.
Match the split to deployment: time-ordered for temporal prediction, grouped for repeated entities, otherwise a fixed random split.
Keep preprocessing inside the training pipeline.
Choose a decision-relevant metric; do not rely on accuracy alone for imbalanced outcomes.
Compare subgroup behavior and variation across folds or seeds when material.

Do not tune complex models, repeatedly inspect a final test set, preprocess before splitting, or celebrate a score without the dumb baseline.

For feature iteration and metric guidance, read references/deep-guide.md.

이 저장소의 다른 Skills

같은 저장소

lemma-causal

tkpratardan/lemma

Rigor for causal questions and A/B tests (the effect of acting on X): confounding, post-treatment bias, valid control groups.

2026-07-154

lemma-describe

tkpratardan/lemma

Rigor for descriptive and diagnostic analytics (what happened and why): denominators, grain, and confounded slices, not model leakage.

2026-07-154

lemma-eda

tkpratardan/lemma

EDA kickoff for a fresh dataset: fixed opening scaffold (goal, imports, load, sanity), then chapters derived from the data; scan leakage, land a baseline.

2026-07-154

lemma-inference

tkpratardan/lemma

Rigor for statistical inference (is the difference real): hypothesis tests, power, multiple comparisons, effect size over p-value.

2026-07-154

lemma-leakage

tkpratardan/lemma

Audit a dataset or pipeline for the five leakages that inflate a metric: target, preprocessing, temporal, group, and sampling.

2026-07-154

lemma-model

tkpratardan/lemma

Final modeling once the baseline and feature set are locked: tune against validation, audit overfitting, touch the test set once, justify the complexity.

2026-07-154

name	lemma-baseline
description	Establish a dumb baseline and an honest validation harness before any real model, so every later number means something.
homepage	https://github.com/tkpratardan/lemma
license	MIT

Baseline

Deliver

Provide a reproducible validation design, a no-information or rule baseline, the simplest credible model, and the precise score later models must beat.

Check

Define the outcome, observation unit, prediction moment, intended use, and a fixed seed.
Match the split to deployment: time-ordered for temporal prediction, grouped for repeated entities, otherwise a fixed random split.
Keep preprocessing inside the training pipeline.
Choose a decision-relevant metric; do not rely on accuracy alone for imbalanced outcomes.
Compare subgroup behavior and variation across folds or seeds when material.

Do not tune complex models, repeatedly inspect a final test set, preprocess before splitting, or celebrate a score without the dumb baseline.

For feature iteration and metric guidance, read references/deep-guide.md.