Float Global Operator

Class instance that manages the training process of a neural network

Structure and Functionalities

The structure of a global operator largely differs depending on the training algorithm in use, hence the abstract class is defined minimalistically.

The global operator should be able to start the training process and feedback loop using execute(). An execute_all() in the corresponding FeedbackQueryManager should be executed directly afterwards.

Code

The abstract class is defined in the unit_neuron.hpp file.