Python bindings for KeOps¶
We now fully document the public interface of the
pykeops module, which is a NumPy or PyTorch front-end for the KeOps C++/Cuda library. This package contains four sets of instructions:
The LazyTensor wrapper: our high-level interface, which provides a pythonic support for the most useful features of KeOps.
The Genred module: our low-level syntax, compatible with NumPy and PyTorch.
The generic_reduction functions: a math-friendly set of helper routines for the Genred module.
kernel_product: a specific syntax with convenient aliases for operations in kernel spaces. This module is only compatible with PyTorch.
- Python install
- Generic reductions
- Generic solver
- Math-friendly syntax
- The kernel_product helper
- Block-sparse reductions