REGRESSION AND CLASSIFICATION USING HIGH ORDER POLYNOMIALS AND TENSOR DECOMPOSITIONS

This folder contains the separate codes for the tensor decomposition algorithms CPD, tensor train (TT) and tensor ring (TR) as well as the results of these. Moreover, it includes a FFNN as comparison model for our classification problem. Regarding the regression model, the Alternating Least Squares TT was used .

Additionally, it contains scripts doing classification and regression using the three models.

Data sets

These data sets are used for creating the results.

• Classification results using Make Blobs data.
• Regression results using California Housing data.

Reference

Part of the code is inspired by exercice 3.4-EXE-FFN-MNIST from the DTU course 02456 Deep learning.