Autoencoders Project

Purpose

To estimate the uniqueness of test roads by using Tensorflow Autoencoders

Approach

I have applied the approach from the paper of Langner et al. to build out the neural networks for encoding and decoding the test routes. If a new tested route has a high error, this indicates that the route was not seen by the model before.

How to run

• git clone https://github.com/SteveBanin/SemesterProject.git
• If using Jupyter: open the final.ipynb file and run cells one after another
• If using Python: execute python final.py within the cloned repository