sehes2solve/Build-CNN-Models-for-Handwritten-Digit-Classification-using-Keras

using MNIST dataset of 60,000 training set images of handwritten single digits between 0 and 9, each image is a 28x28 pixel square. 4 different architectures of CNN variant in (convolutional layers, pooling layers, filter size, ....). Apply K-folding cross-validation during training to avoid overfitted model misinterpretation.