Spectral Tensor Train Parameterization of Deep Learning Layers
-
Updated
Jul 1, 2021 - Python
Spectral Tensor Train Parameterization of Deep Learning Layers
A take on the Tensor Train decomposition
Python implementation of the Parametric Tensor Train Kernel (PTTK) method.
Image classification
The core idea is to parametrize the right-hand side of an ordinary differential equation (ODE) using a tensor train (TT) decomposition, such that the discretization of the ODE via standard numerical methods, such as the Explicit Euler scheme, implicitly induces a compositional TT structure.
Add a description, image, and links to the tensortrain topic page so that developers can more easily learn about it.
To associate your repository with the tensortrain topic, visit your repo's landing page and select "manage topics."