A Python package for the approximation of functions and tensors.
tensap features low-rank tensors (including canonical, tensor-train and tree-based tensor formats or tree tensor networks), sparse tensors, polynomials, and allows the plug-in of other approximation tools. It provides different approximation methods based on interpolation, least-squares projection or statistical learning.
See the tutorials folder and its subfolders for examples of use.
Install from PyPi:
pip install tensap
Install from conda:
conda config --add channels conda-forge
conda config --set channel_priority strict
conda install tensap