SOMTSP is short for Self-Organized Map Traveling Salesman Problem.
Traveling Salesman problem is a np-complete problem. We are trying to solve this problem, with SOM. However, main attempt in this project is improving SOM with Simulated Annealing.
Followings are version differences of SOMTSP,
0.8:
- Simmulated anneling with different cooling strategy added.
- Learning parameters improved.
0.5:
- Simmulated anneling fixed.
- Visualization completed.
0.2:
- Visualization with ACM function implemented.
- Distance finder function implimented.
0.1:
- Print functions implemented.
- Matrix excluded, phi function added.
- Variable names refactored.
0.0:
- SOM Node created.
- SOM Neuron created.
- SOM Kohonen main class implimented (%80).