Comparison of Different Convolutional Neural Network Structures Based on Keras

Abstract

Deep learning is subfield of machine learning, which aims to learn hierarchy of features from input data. Deep learning technologies are becoming major approaches for natural signal and information processing like image classification and speech recognition. Deep learning is a technology inspired by functioning of human brain. Convolutional neural network (CNN) becomes very popular for image classification in deep learning. In this paper, it is discussed about the deep learning’s convolutional structures based on Keras. Four different structures of CNN are compared on CPU system with four different combinations of classifiers and activated functions.