ワンクリックでManusで任意のスキルを実行

missingness-profile

スター0

フォーク0

更新日2026年5月20日 14:32

Report null percentages and recommended imputations for water quality data. Use when profiling missing data, deciding imputation strategy, or assessing data completeness.

インストール

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。

Manusで実行

ソース

DevExpGbb

DevExpGbb/data-science

GitHub リポジトリを開く Creator のリポジトリを見る

ダウンロード

Manusで実行

Missingness Profile

Profile null/missing values and recommend imputation strategies.

When to Use

Before model training or KPI computation
When missing data percentage is unknown
When deciding between imputation strategies

Procedure

Compute per-column null counts and percentages
Classify missingness severity:
- Low (< 5%): safe to impute with median/mode
- Medium (5–20%): impute with care; flag in output
- High (> 20%): warn; consider dropping column or using forward-fill for time series
Recommend imputation method per column based on dtype and missingness pattern
Check for systematic missingness patterns (e.g., always null for a specific site_id)

Output

Column	Null %	Severity	Recommended Action
...	...	...	...

Plus a summary of any systematic patterns found.

このリポジトリの他の Skills

同じリポジトリ

contamination-threshold-check

DevExpGbb/data-science

Flag water quality records that exceed contamination thresholds. Use when checking purity_score < 75, contamination_flag = 1, or identifying at-risk sites and records.

2026-05-200

kpi-definition

DevExpGbb/data-science

Standardize water purity, alert-rate, and site-risk metrics. Use when defining KPI calculations, ensuring metric consistency across reports, or setting up a new KPI pipeline.

2026-05-200

narrative-synthesis

DevExpGbb/data-science

Convert water quality metrics into concise operations guidance. Use when writing executive summaries, translating KPIs into plain-language recommendations, or producing briefing narratives.

2026-05-200

schema-check

DevExpGbb/data-science

Validate expected columns and coercible dtypes in water quality datasets. Use when checking schema conformance, column presence, or type compatibility before analysis.

2026-05-200

name	missingness-profile
description	Report null percentages and recommended imputations for water quality data. Use when profiling missing data, deciding imputation strategy, or assessing data completeness.
user-invocable	true
argument-hint	Path to data file or DataFrame variable name

Missingness Profile

Profile null/missing values and recommend imputation strategies.

When to Use

Before model training or KPI computation
When missing data percentage is unknown
When deciding between imputation strategies

Procedure

Compute per-column null counts and percentages
Classify missingness severity:
- Low (< 5%): safe to impute with median/mode
- Medium (5–20%): impute with care; flag in output
- High (> 20%): warn; consider dropping column or using forward-fill for time series
Recommend imputation method per column based on dtype and missingness pattern
Check for systematic missingness patterns (e.g., always null for a specific site_id)

Output

Column	Null %	Severity	Recommended Action
...	...	...	...

Plus a summary of any systematic patterns found.