For those who aren't familiar with the dataset, it consists of bunch of grayscale images. And if the image were to be represented as a matrix, each of the elements is a floating point of range 0 - 1.0. What format is that kind of image?
I ask because I want to predict my own image using a classifier trained through MNIST dataset, but I couldn't figure out how should I pre-process my image.
Thanks a lot in advance!