scaling-fit

Finite-size scaling analysis automates curve fitting on size-indexed quantum observable data to extract critical exponents. It supports power-law, log-L, polynomial, and data-collapse forms to generate research-grade fit parameters.

Bootstrap resampling estimates uncertainty by repeatedly drawing samples from the dataset to generate reliable error bars for fitted exponents. This automated statistical resampling ensures validity and diagnostic transparency without requiring manual calculations.

Data-collapse fitting is supported for identifying phase transitions in quantum simulations. The skill fits size-indexed observable data to data-collapse forms and automatically generates residual plots and data collapse curves to validate fit quality.

Automating weighted least-squares fitting for computational physics data is handled by applying regression to size-indexed observables. The skill combines this fitting method with bootstrap uncertainty estimation to ensure statistical rigor.

Extracting critical exponents requires size-indexed quantum observable data provided in a standard data table. You can point the skill to the results directory of your current run to automatically process the tabular data for fitting.