tensorlearn 1.0.65

A Python Package for Advanced Tensor Learning Methods

tensorlearn

tensorlearn is a Python library distributed on Pypi for implementing tensor learning

This is a package under development. Yet, the available methods are final and functional. The backend is Numpy.

Installation

Use the package manager pip to install tensorlearn in Python.

pip install tensorlearn

methods

Decomposition Methods

Tensor-Train Decomposition with Auto Rank Determination

Tensor-Train Decomposition with Auto Rank Determination

Tensorlearn.decomposition.tensor_train.auto_rank_tt(tensor,epsilon)

This implementation of tensor-train decomposition determines rank automatically based on a given error bound written according to Oseledets (2011). Therefore the user does not need to specify a rank. Instead the user specifies an upper error bound (epsilon) which is the frobenius norm of the error divided by the frobenius norm of the given tensor to be decomposed.

Arguments

@tensor - dimension must be at least 3.

@epsilon - Error bound = frobenius norm of the error / frobenius norm of the given tensor.

Outputs

@factors - The list includes 2D numpy arrays of factors according to TT decomposition. Length of the list equals the dimension of the given tensor to be decomposed.