analyze-results

Analyze ML experiment results to compute statistics and generate comparison insights.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/KwongFuk/codex-skills --skill analyze-results-kwongfuk
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: analyze-results
Source: https://github.com/KwongFuk/codex-skills/tree/main/global/analyze-results
Command: npx skills add https://github.com/KwongFuk/codex-skills --skill analyze-results-kwongfuk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Analyze ML experiment results to compute statistics, generate comparison tables, and derive actionable insights to support data-driven decisions.

Core Features & Use Cases

  • Statistical summaries: compute mean, std, and confidence intervals across seeds and hyperparameter sweeps.
  • Comparison tables: organize results by model type, data config, and metrics to reveal trends.
  • Insights & reporting: generate concise interpretations and recommended next steps for experiments.

Quick Start

Run the analyze-results workflow on your latest experiment folder to produce a stats report and insights.

Frequently Asked Questions about analyze-results

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I analyze ML experiment results across multiple seeds?

You can analyze ML experiment results by computing statistical summaries like mean and std across seeds, generating comparison tables, and deriving actionable insights to support data-driven decisions.

What is the best way to compare model performance from parameter sweeps?

The best way to compare model performance from parameter sweeps is to organize results by model type and data config, calculate delta vs baseline, and produce structured tables with concise findings and recommended next steps.

Can I generate a stats report directly from my experiment folder?

Yes, you can run the analysis workflow on your experiment folder to automatically produce a stats report containing raw tables, mean +/- std, delta vs baseline, and recommended next steps.

Does this approach work for tracking metrics across different data configurations?

Yes, analyzing ML experiment results works for different data configurations by organizing results into comparison tables to reveal trends across models and data configurations.

What statistics are included when analyzing machine learning experiments?

Analyzing machine learning experiments includes computing mean, std, and confidence intervals across seeds and hyperparameter sweeps, along with delta vs baseline and concise findings.