Overparametrization of the MNIST images
Tal-Golan opened this issue · 2 comments
Tal-Golan commented
It seems that the MNIST images are tiled so they have three (identical) color channels, matching the CIFAR-10 format (data_loaders/get_mnist_cifar, lines 43-44:)
x_train = np.tile(np.reshape(x_train, (-1, 32, 32, 1)), (1, 1, 1, 3))
x_test = np.tile(np.reshape(x_test, (-1, 32, 32, 1)), (1, 1, 1, 3))
Since this triples the size of the latent representation, I wonder whether this is the best way to model MNIST images with Glow. Any reason why I shouldn't disable this tiling by removing the np.tile call?
sophiejwalton commented
Was the reason you used this method is so you can load a model with pretrained weights? Or was it for something else.
Tal-Golan commented
No pre-trained weights, training an MNIST model from scratch.