What file formats does this skill support?

The skill is designed to work with directory structures containing bin_*.h5 files, which are the standard HDF5 output format for these DQMC simulations.

Can I use this for batch processing multiple runs?

Yes, the skill includes patterns for iterating through subdirectories to collect results from multiple simulations simultaneously, making it ideal for parameter sweeps.

Which physical observables can I extract?

Supported observables include fermion sign, density (den), double occupancy (docc), Green's functions (gr/gk), spin correlators (zzr/xxr), and various structure factors (nnq, zzq, swq0).

Does it support unequal-time measurements?

Yes, for measurements like zero-frequency susceptibilities (nnrw0, zzrw0) or d-wave pair susceptibility (dwq0t), provided the simulations were run with the period_uneqlt parameter set correctly.

How does it handle the fermion sign problem?

It extracts the mean and standard error of the sign; it also includes validation checks to warn the user if the error bar is too large relative to the mean, indicating the sign problem is too severe for reliable results.

DQMC Data Analysis

Name: DQMC Data Analysis
Author: edwnh

byedwnh

•

Data Science & ML

Analyzes Determinant Quantum Monte Carlo (DQMC) simulation data to extract physical observables with statistical error estimates.

The dqmc-analyze skill enables Claude to process and interpret raw results from Determinant Quantum Monte Carlo simulations. It automates the extraction of key physical quantities such as fermion sign, density, double occupancy, and spin correlations from HDF5 bin files. By providing standardized methods for error estimation and data aggregation across multiple simulation directories, this skill streamlines the post-processing workflow for researchers studying lattice models like the Hubbard model, ensuring rigorous statistical validation of simulation outputs and helping identify potential sign problems or convergence issues.

Key Features

01Batch processing for aggregating results from multiple simulation directories

02Validation logic to detect severe fermion sign problems or insufficient statistics

03Extraction of physical observables including density, sign, and spin correlations

04Automatic statistical error estimation providing mean and standard error tuples

0516 GitHub stars

06Support for advanced measurements like structure factors and zero-frequency susceptibilities

Use Cases

01Calculating magnetic moments and spin-z structure factors from Hubbard model simulations

02Monitoring the fermion sign problem across different temperature and interaction scales

03Post-processing large-scale simulation sweeps into consolidated datasets for plotting

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add edwnh/dqmc dqmc-analyze

For use in Claude.ai and ChatGPT

Download Skill