Optimizers

The optimizers module contains the low-level fitting functions. Most users will use the high-level df.fit() interface, but these functions are available for advanced use cases.

Optimization routines module for ezfit.

This module provides low-level optimization functions that wrap scipy.optimize and emcee. These functions are called internally by FitAccessor.fit() via a registry pattern and should not be called directly by users.

Features

• Multiple optimization algorithms (curve_fit, minimize, etc.)

• Global optimizers (differential_evolution, shgo, dual_annealing)

• MCMC sampling via emcee with automatic diagnostics

• Regularized regression via scikit-learn (Ridge, Lasso, ElasticNet)

• Automatic covariance matrix estimation for all methods

• Constraint handling for parameter relationships

Note

These functions are typically called internally by FitAccessor.fit(). Direct use is only recommended for advanced scenarios.