tucker-riemopt 0.0.2

Tucker toolbox for Riemannian optimization.

Tucker Riemopt

Python implementation of the Tucker toolbox. Package allows users to manipulate tensors in Tucker and SF-Tucker [1] formats. It also provides tools for implementing first-order optimization methods of the Riemannian optimization on the manifolds of tensors of fixed Tucker rank or fixed SF-Tucker rank. For instance, package implements a method for efficiently computing the Riemannian gradient of any smooth function via automatic differentiation.

The library is compatible with several computation frameworks, such as PyTorch and JAX, and can be easily integrated with other frameworks.

Installation

NumPy, SciPy, PyTorch and opt-einsum are required for installation. Additionally, you need to install your special computation framework (e.g. JAX).

Package may be installed using

pip install tucker_riemopt

We recommend use poetry for installation, as preferred computational framework may be specified by

poetry install -E torch

or

poetry install -E jax

for PyTorch/JAX correspondingly.

Use cases

See this repository for examples of package usage.

Documentation

Detailed information may be found here.

License

MIT License