MATH 490 Group 7: Philip Dohm, Roger Qiu, Jonathan Shobrook

• ./final.ipynb - Implements GNN to solve the Traveling Salesman Problem
• ./data/figures/train_stats.png - Visualization of training/generalization statistics
• ./data/data/pickles/ - Model training checkpoints and training/testing statisitics

To run: just run the final.ipynb file.

Description:

Tensorflow and graphs_nets implementation of solving the Traveling Salesman Problem using a Graph Neural Network.

Dependencies:

• random
• tensorflow
• networkx
• numpy
• pandas
• time
• pickle
• tsp_solver
• graphs_nets