diffjeom

Automated differential geometry, powered by jax. Given a metric function, diffjeom can compute:

• Christoffel symbols (get_christoffel2());

• Riemann curvature tensor (get_riemann());

• Ricci tensor (get_ricci_tensor());

• Ricci scalar/scalar curvature (get_ricci_scalar()).

These work with the usual jax transformations like jit and vmap.

The functions check_christoffel2_sym(), check_riemann_sym(), check_diff_bianchi() and check_ricci_tensor_sym() are provided to check identities for these objects.

Take a look at the full API documentation for more details.

Installation

Install diffjeom and its few dependencies (jax, jaxlib, numpy) from PyPI with

pip install diffjeom

Requires python 3.6 or newer.

Citing

Please acknowledge diffjeom if you use it in your work.