xitorch: differentiable scientific computing library¶
xitorch (pronounced “sigh-torch”) is a library based on PyTorch that provides differentiable operations and functionals for scientific computing and deep learning. xitorch provides analytic first and higher order derivatives automatically using PyTorch’s autograd engine. It is inspired by SciPy, a popular Python library for scientific computing.
Example operations available in xitorch:
xitorch.linalg.symeig: symetric eigendecomposition for large sparse matrix or implicit linear operator,
xitorch.optimize.rootfinder: multivariate root finder, and
xitorch.integrate.solve_ivp: initial value problem solver or commonly known as ordinary differential equations (ODE) solver.
Why use xitorch:
contains differentiable functionals;
provides 1st, 2nd, and higher order gradients of functionals;
enables the use of functionals in the object-oriented way.
Source code: https://github.com/xitorch/xitorch/
Getting started