Teaching Neural Network Programming

In this book we briefly address the theoretical foundation of neural networks, from the basic principles of how a neuron works and its similarity with the biological part, in which we explain its axons (inputs), the weights of the inputs, the bias, the neuron body, the activation function and the axon output function. Subsequently, the architecture of a multilayer neural network, the learning process of the neural network through the functions of backpropagation and feedforward propagation are described. Finally, the book proceeds to train a neural network from scratch, with the aim of teaching the reader in a simple way how to design a neural network and the processes involved in learning it. Subsequently, a problem is defined for the detection of late blight in potato crops. The theoretical foundation of a convolutional network and definitions related to potato blight are presented. The reader is then presented with a step-by-step guide on how to create a project in Jupyter-lab for the detection of potato leaf blight.