name	data-analysis
description	Reproducible data analysis. Uses pandas/numpy/scipy/statsmodels for cleaning, statistics, and modeling; scripts and random seeds all land in the run, and every reported number is recomputable. Use to analyze tabular/matrix data, run statistical tests, or model.

data-analysis

Put data in runs/<run>/inputs/ (SHA256-registered).
Write the analysis script to runs/<run>/outputs/analyze.py, fix the random seed, and write products (tables, intermediates) to outputs/.
Run the script with Python — don't hand-compute in a REPL; hand-computed numbers aren't reproducible.
Export key statistics to outputs/results.json for figure-maker and manuscript to cite.

Every number that enters text/figures must be recomputable from outputs/ scripts + inputs/ data.
Check and record statistical assumptions (normality, variance homogeneity, sample size, multiple-comparison correction).
Write non-ASCII results to files; do not print them to a GBK stdout.

name	data-analysis
description	Reproducible data analysis. Uses pandas/numpy/scipy/statsmodels for cleaning, statistics, and modeling; scripts and random seeds all land in the run, and every reported number is recomputable. Use to analyze tabular/matrix data, run statistical tests, or model.

Put data in runs/<run>/inputs/ (SHA256-registered).
Write the analysis script to runs/<run>/outputs/analyze.py, fix the random seed, and write products (tables, intermediates) to outputs/.
Run the script with Python — don't hand-compute in a REPL; hand-computed numbers aren't reproducible.
Export key statistics to outputs/results.json for figure-maker and manuscript to cite.

Every number that enters text/figures must be recomputable from outputs/ scripts + inputs/ data.
Check and record statistical assumptions (normality, variance homogeneity, sample size, multiple-comparison correction).
Write non-ASCII results to files; do not print them to a GBK stdout.

このリポジトリの他の Skills