ABSTRACT
This chapter focuses on a convolutional layer and two other types of layers that are generally included in a convolutional neural network (CNN). CNNs have proved extremely successful with image recognition. They are also used for numerous other applications. The chapter presents the CNN to show that coding skills have been brought to a remarkable place. It discusses two types of layers in neural networks: fully connected layers and activation layers. It finds that the modular approach affords a great deal of flexibility in the design of neural networks. The chapter provides with the analysis of the MNIST data and classifies handwritten digits with suprising accuracy. The MNIST database contains 60,000 training images and 10,000 testing images. The database is also widely used for training and testing in the field of machine learning.
