This is a tutorial to implement a feedforward neural network from scratch in Python in under 225 lines. The training technique is using a genetic algorithm with crossover and mutation operators along with elite and tournament selection.
Refer to genetic-neural-network-tutorial.ipynb for the tutorial. If you want to jump straight to the code, code/genetic_network_uc.py is the uncommented code and code/genetic_network.py is the commented code.