gpd-parameter-sweep

Executes systematic parameter sweeps with wave-based parallel execution and result aggregation.

Updated Mar 18, 2026

Search Tags:#parallel execution #parameter sweep #computational physics #result aggregation #phase diagram #grid search #adaptive refinement

Ready to run

npx skills add https://github.com/MichaelsEngineering/sentinel-multiscale-field-dynamics --skill gpd-parameter-sweep-michaelsengineering

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: gpd-parameter-sweep
Source: https://github.com/MichaelsEngineering/sentinel-multiscale-field-dynamics/tree/main/.agents/skills/gpd-parameter-sweep
Command: npx skills add https://github.com/MichaelsEngineering/sentinel-multiscale-field-dynamics --skill gpd-parameter-sweep-michaelsengineering

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires numpy.

What problem does it solve? Ad hoc parameter sweeps waste compute on uniform grids, miss phase transitions between grid points, and scatter results across disorganized files. This Skill structures computational physics sweeps from design through parallel execution to aggregated analysis. ## Core Features & Use Cases - Structured Sweep Design: Define 1D or 2D parameter grids with linear, logarithmic, or adaptive spacing, then generate self-contained execution plans for each point. - Wave-Based Parallel Execution: Batch independent sweep points into waves of 5-8 plans executed concurrently by executor subagents, with per-point failure isolation. - Adaptive Refinement: Detect regions with large derivatives, non-monotonic segments, or gaps, then automatically add refined sampling points near transitions and extrema. - Use Case: Map a phase diagram by sweeping temperature from 0.1 to 10.0 across 20 points, computing ground state energy at each point, then refining near a detected crossover and producing a summary table with identified features. ## Quick Start Run a parameter sweep over temperature from 0.1 to 10.0 in 20 logarithmic steps with adaptive refinement enabled for the current phase.

Frequently Asked Questions about gpd-parameter-sweep

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

FAQPage Schema

How do I run a parameter sweep in computational physics?▼

Define the parameter name, range as start:end:steps, and observable, then the sweep generates one plan per grid point. Points execute in parallel waves, and results aggregate into a sweep-results.json file plus a markdown summary table.

What grid spacing should I use for a parameter sweep?▼

Use linear spacing for smooth observables and logarithmic spacing for parameters spanning orders of magnitude like couplings or masses. Phase diagrams need 20-50 points per dimension, while scaling law fits need 10-20 log-spaced points.

How does adaptive refinement work in a parameter sweep?▼

Adaptive refinement computes numerical first and second derivatives of the results and flags regions where change exceeds three times the median, where monotonicity breaks, or where data gaps exist. New points are added in those regions, executed, and merged into the dataset.

Can I run a 2D parameter sweep over two variables?▼

Yes, specify two parameters with their own ranges to generate a Cartesian product grid, such as a 10x10 temperature-coupling grid. Each pair is an independent computation executed in parallel waves.

What happens when a single sweep point fails?▼

A failed point is recorded in its result JSON with status agent_failed and a null observable, and the sweep continues with remaining points. Failed points are documented in the SWEEP-SUMMARY.md under a Failed Points section.

Why do parameter sweeps return NaN or identical values?▼

NaN or Inf results typically come from division by zero, overflow, or underflow at special parameter values and are excluded from feature analysis. Identical values across all points mean either the parameter does not affect the observable or the parameter is not being set correctly.