TSP Local Solver

Python implementation of different local heuristics for the TSP.

- Current Implementation has 2-opt Algorithm 
- Multiple Adj Matrices are being used as a single matrix with all almost 200k points causes a memory error
- Implemented 2-opt (n^2) on top 1000 points got from optimized set
- Current implementation is more focused on not using a distance matrix but only make sure entire dataset is used
- Will be Adding c++ code with better algorithms to this to make it more robust and optimized

Installation

• This is a standalone library, better included directly in large scale datasets. Works with Python 3 (3.7xx)
• The input to this needs to be a .csv files with the coordinates of the cities
• I am working on to make it more generic as a package

Optional

• Doctest for running tests.
• Pytest for test integration.
• Yapf for code formatting.

TODO

• Improve by adding LKH, and k-opt Algorithms
• Verify with more test cases

Travis-Build