Neural Network Demonstration

Tool to visualize the learning process of a neural network.

In this project there is a simple neural network that takes as input the distances between a car and the walls around it, and learns to drive around various tracks based on that information.

You can experiment with different values for each car's view radius, as well as different weights for the fitness evaluation.

You may also save a network you've been experimenting on through the tool's Save/Load system, which allows you to keep the progress of each network between sessions.

Included in the screenshots is a table detailing the results of a few experiments with different values within the network.

By writing "Case[number of case in the table]" inside the tool's Save/Load system, you can see for yourself the final state of each network by the time of adding its information to the table.