import tensorwaves

A model optimization package for Partial Wave Analysis.

The tensorwaves package contains four main ingredients:

Function creation (tensorwaves.function)

Express arbitrary mathematical expressions as functions in different kinds of computational backends, such as JAX, TensorFlow, NumPy, and Numba.

Data generation (tensorwaves.data)

Generate phase space samples as well as hit-and-miss Monte Carlo samples for the input mathematical expression.

Optimization (tensorwaves.optimizer and estimator)

Optimize a ParametrizedFunction with respect to some data sample and an Estimator (loss function).

The interface module defines how the main classes interact.

Submodules and Subpackages