Hands-On Neuroevolution with Python

Test environment for the modular retina problem

First, we need to create a test environment that can be used to evaluate the results of the neuroevolution process that aims to create a successful detector ANN. The test environment should create a dataset that consists of all possible patterns of pixels on the retina. Also, it should provide functions to evaluate the detector ANN against each pattern in the dataset. Thus, the test environment can be divided into two main parts:

• The data structure to hold visual patterns for the left-hand, the right-hand, or both sides of the retina
• The test environment storing the dataset and providing functions for detector ANN evaluation

In the following sections, we provide a detailed description of each ...