Copyright 2024 National Technology & Engineering Solutions of Sandia,
LLC (NTESS). Under the terms of Contract DE-NA0003525 with NTESS, the
U.S. Government retains certain rights in this software.

pyttb: Python Tensor Toolbox

Welcome to pyttb, a refactor of the Tensor Toolbox for MATLAB in Python.

This package contains data classes and methods for manipulating dense, sparse, and structured tensors, along with algorithms for computing low-rank tensor decompositions:

• Data Classes: tensor, sptensor, ktensor, ttensor, tenmat, sptenmat, sumtensor
• Algorithms: cp_als, cp_apr, gcp_opt, hosvd, tucker_als

Quick Start

Installation

python3 -m pip install pyttb

Example

>>> import pyttb as ttb
>>> X = ttb.tenrand((2,2,2))
>>> type(X)
<class 'pyttb.tensor.tensor'>
>>> M = ttb.cp_als(X, rank=1)
CP_ALS:
 Iter 0: f = 7.367245e-01 f-delta = 7.4e-01
 Iter 1: f = 7.503069e-01 f-delta = 1.4e-02
 Iter 2: f = 7.508240e-01 f-delta = 5.2e-04
 Iter 3: f = 7.508253e-01 f-delta = 1.3e-06
 Final f = 7.508253e-01

Memory layout

For historical reasons we use Fortran memory layouts, where numpy by default uses C. This is relevant for indexing. In the future we hope to extend support for both.

>>> import numpy as np
>>> c_order = np.arange(8).reshape((2,2,2))
>>> f_order = np.arange(8).reshape((2,2,2), order="F")
>>> print(c_order[0,1,1])
3
>>> print(f_order[0,1,1])
6

Getting Help

• Documentation
• Tutorials
• Info for users coming from MATLAB
• Learn about tensor decompositions: tensor paper, tensor book

Contributing

• Report a bug
• Guide for contributors
• List of contributors

Citing pyttb in your work

If you use pyttb in your work, please cite it using the citation info here.