The MNIST dataset contains 60,000 training images of handwritten digits from zero to nine and 10,000 images for testing. The handwritten digits images are represented as a 28ร28 matrix where each cell contains grayscale pixel value.
The model is trained using a Convolution Neural Network with loss function as 'sparse_categorical_crossentropy' and optimizer as 'adam'. The model gives an accuracy of 97.6 %. The confusion matrix of the predicted value and the actual value is given below: