DEVELOPING A BASIC NEURAL NETWORK TO CLASSIFY IMAGES FROM THE MNIST DATASET

Abstract

This research discusses one of the important topics, which is entitled " Developing a basic neural Network to classify images from the MNIST dataset". The research aims mainly to talk about the concept of MNIST dataset. It also aims to reveal the importance of developing a basic neural network in classifying images from the MNIST dataset and discuss how to create a neural network to recognize handwritten digits from the famous MNIST dataset. The process of collecting data is based on the descriptive study by collecting and measuring the required information, which can be obtained using many variables. So, and this process is very important for a number of reasons, which are: the most important that It helps in obtaining answers and solutions of many problems and questions, it helps in facilitating the decision-making and increases the quality of decisions that were taken, and also it helps in improving the quality of different outputs,the findings of this paper are that the MNIST dataset consists of a total of 70,000 images, with 60,000 images designated for training and 10,000 images for testing purposes. İt was found out that the MNIST dataset consists of 10 classes, enabling us to classify numbers ranging from 0 to 9. İt was concluded that the MNIST dataset is widely recognized as a popular benchmark for learning the usage of neural networks.