Skip to content

A MATLAB Toolbox for High-order Tensor Data Decompositions and Analysis

License

Notifications You must be signed in to change notification settings

andrewssobral/TDALAB

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

TDALAB

Laboratory for Tensor Decomposition and Analysis

by Guoxu Zhou, Andrzej Cichocki
2012 Cichocki Laboratory for Advanced Brain Signal Processing

Download TDALAB manual
Current version 1.1, released May 1st, 2013.
For any support, request and bug reports contact guoxu.zhou(at)riken.jp.


Datafiles for testing are available in TDALAB benchmark [~30MB].

Tensor Toolbox (TDALAB) provides fundamental data structures and functions for tensor data processing. TDALAB attempts to provide an easy to use, user-friendly toolbox for experimentation and application of tensor decomposition and analysis.

TDALAB highlights and features

  • friendly graphical user interface (GUI) for tensor decompositions, enabling easy selection of decomposition model, algorithm, and parameters
  • platform for comparison and evaluation of a large number of state-of-the-art tensor decomposition algorithms, and provides friendly GUI to access the widely used functions included in N-way Toolbox and Tensor Toolbox, and some of the latest developments in tensor decompositions.
  • implementation of constrained tensor decomposition by incorporating standard 2D Penalized Matrix Factorization (PMF) methods in order to impose diversity/constraints on components (columns of factor matrices), such as orthogonality, statistical independence, sparsity, nonnegativity, etc (as in Multilinear Blind Source Separation (MBSS)).
  • implementation of 2D constrained matrix factorization (also referred to as 2D Blind Source Separation (BSS))
  • multiple visualization approaches for tensor objects are provided, users can explore the components and their connections

Link to the Laboratory homepage