The original MNIST dataset contains a lot of handwritten digits. Members of the AI/ML/Data Science community love this dataset and use it as a benchmark to validate their algorithms. In fact, MNIST is often the first dataset researchers try. “If it doesn’t work on MNIST, it won’t work at all”, they said. “Well, if it does work on MNIST, it may still fail on others.”
To Serious Machine Learning Researchers
MNIST is too easy.
MNIST is overused.
MNIST can not represent modern CV tasks.
MNIST Dataset contains 60000 images of 10 different classes.
I was able to get an accuracy of 91.2 % on this Dataset.
It is not interactive