This is the computational appendix for the following paper:
Ding Liu, Shi-Ju Ran, Peter Wittek, Cheng Peng, Raul Blázquez García, Gang Su, Maciej Lewenstein. Machine Learning by Two-Dimensional Hierarchical Tensor Networks: A Quantum Information Theoretic Perspective on Deep Architectures. arXiv:1710.04833, 2017.
The code uses tncontract for tensor contractions. Other dependencies are SciPy, Matplotlib, and Scikit-learn.
The data files can be downloaded from here.
-
tree_tensor_network_mnist.py: The implementation of the tree tensor network for the MNIST dataset. -
tsne_mnist.py: Plotting the t-SNE embedding. -
utilities_mnist.py: Helper functions. -
TTN_mnist.py: The main file to train and test the tree tensor network on MNIST. -
TTN_tsne.py: The script to generate the model for t-SNE embedding.