Run Train_cnn.py that will train on MNISt data and save the trianed weights. Then run test_cnn.py that reads the trained weights and then gives the prediction. I made a test_sr_v1.py for superesolution (can't recall if it runs but it's a modification of test_cnn.py) Run the test_sr_v1.py for testing.
kanishkaisreal/Numpy-CNN
A numpy-only implementation of a Convolutional Neural Network, from the ground up.
Python