Handwritten digits classification is the task of classifying a handwritten digit image into one of the ten digits, 0 through 9. This is a challenging task because handwritten digits can vary greatly in appearance, depending on the handwriting of the person who wrote them
The MNIST dataset is a popular dataset used for handwritten digits classification. The MNIST dataset contains 60,000 training images and 10,000 testing images of handwritten digits. The images in the MNIST dataset are 28x28 pixels in size.