mprod-package 0.0.2a1

Software implementation for tensor-tensor m-product framework

mprod_package

Software implementation for tensor-tensor m-product framework [1]. The library currently contains tubal QR and tSVDM decompositions, and the TCAM method for dimensionality reduction.

Installation

using pip

The package is available at pypi and can be installed via the command

pip install mprod-package 

from source

Make sure that all dependencies listed below are installed in a newly created conda environment, preferably - using the conda-forge channel.

We stated the exact versions used to locally test the code, more recent versions of these packages should work as well.

Dependencies:

• python 3.6.8
• scipy 1.5.3
• scikit-learn 0.24.1
• numpy 1.19.2
• dataclasses 0.7 (Only for python version < 3.7)
• pip 21.0.1

Clone the repository, then from the package directory, run

pip install -e .

References

[1] Misha E. Kilmer, Lior Horesh, Haim Avron, and Elizabeth Newman. Tensor-tensor algebra for optimal representation and compression of multiway data. Proceedings of the National Academy of Sciences, 118(28):e2015851118, jul 2021.