Convolutional Neural Networks: An Analysis of Varying Factors

In computer vision, Convolutional Neural Networks (CNN) are regarded well for their ability to process data with grid-like topology, such as images. However, with the wide variety of models available, as well as the number of external factors associated with the problem at hand, it can be difficult to determine what model works best for a specific problem. To address this, the paper aims to analyze and compare the performances of different Convolutional Neural Networks on a fashion image dataset. Additionally, this research highlights factors to consider when building or deciding on a model, as well as showing an example of what models are the most suitable for a specific problem size.