A simple tool for computing objects of interest in general relativity using Automatic differentiation.
See lib_test.ipynb for a simple example use case. To create a metric module one needs a function of the form:
def metric_function(coordinate_vector: torch.Tensor) -> torch.Tensor:
return torch.diag([-1,1,1,1])See GRPytorch_metrics.py for example metrics. For now only schwarchild was added.
The example in schwar_test.ipynb uses Einsteinpy to compare the symbolic output to the numerical estimates obtained via automatic differentiaiton.
Due to round-off errors, the Ricci and Einstein tensor in vacuum may not necessarily be zero. I have not figured out a good method to fix the round-off issue.