A step-by-step guide to building an artificial neural network with the MNIST dataset using CNN architecture. The post covers accessing the data, data preprocessing, normalization, model architecture design, model compilation, training, evaluation, and detecting overfitting.
Table of contents
Importing the most important librariesCalling the datasetData processingNormalizationVisualizing the dataDecision MakingStep 1: Architecture Design (Model Building)compile.modelTraining Phasefit Phase (model.fit)All problems stem from overfitting!!!!Preventing overfitting!?!? How???The graphsFinal test98 Impressions