skills.rest
lix965996-art
lix965996-artCOMMUNITY
Explore all skills by this author →
GitHub

analyze-results

Analyze machine learning experiment JSON/CSV results to compute statistics and compare runs.

1|Updated May 14, 2026
Search Tags:#statistics#analyze#hyperparameter sweep#comparison table#ml experiments#baseline delta#research insights
One-click install
npx skills add https://github.com/lix965996-art/MMM --skill analyze-results-lix965996-art
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: analyze-results
Source: https://github.com/lix965996-art/MMM/tree/main/resources/app/skills/analyze-results
Command: npx skills add https://github.com/lix965996-art/MMM --skill analyze-results-lix965996-art

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you make sense of machine learning experiment outputs by converting raw result files into readable comparisons and actionable findings, so you can quickly understand what changed and why it mattered.

Core Features & Use Cases

  • Results discovery & parsing: Finds JSON/CSV outputs across common project folders and structures them for analysis.
  • Comparison tables & baselines: Groups results by independent variables and computes deltas against a baseline for faster model selection.
  • Statistics & insight generation: Summarizes mean and variability across seeds, identifies trends across sweeps, flags outliers, and drafts observation/interpretation/implication/next-step statements.

Quick Start

Use the analyze-results skill with the path or description of your experiment results directory to produce a raw data table, key findings with numbers, and suggested next experiments.

Frequently Asked Questions about analyze-results

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

FAQPage Schema
How do I compare machine learning experiment results from multiple runs?▼

Comparing machine learning experiment results involves parsing JSON or CSV output files, grouping metrics by independent variables like hyperparameters, and computing deltas against a baseline to identify the best performing model.

How do I analyze hyperparameter sweep results to find the best configuration?▼

Analyzing hyperparameter sweep results requires structuring independent and dependent variables from experiment logs, computing statistics like mean and standard deviation across seeds, and identifying trends to recommend next-step configurations.

What is the best way to summarize multi-seed evaluation metrics for research insights?▼

Summarizing multi-seed evaluation metrics involves calculating mean and variability to capture model stability, flagging outliers, and drafting observation and interpretation statements to generate research-ready insights.

Can I parse JSON and CSV metrics files from common project folders automatically?▼

Parsing JSON and CSV metrics files automatically requires discovering result artifacts across common project directories and structuring the raw data into comparison tables for analysis.

Does this approach work for generating baseline comparison tables from raw experiment logs?▼

Generating baseline comparison tables from raw experiment logs works by grouping results by independent variables, computing deltas against a defined baseline, and producing numbered findings for model selection.

Related Skills

Explore More Data & Analytics Skills→
zhuyingqinzhuyingqin
1

analyze-results

Analyze machine learning experiment result files to generate statistics, comparisons, and findings.

Community
Intermediate
lightrain-alightrain-a

analyze-results

Parse JSON/CSV ML experiment outputs into comparable statistics and numbered findings.

Community
Intermediate
wildwasserwildwasser
35

ml-experiment

Track machine learning experiments with structured logging and reproducibility.

Community
Advanced
KwongFukKwongFuk

analyze-results

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

Community
Advanced