Used in https://towardsdatascience.com/mixture-of-variational-autoencoders-a-fusion-between-moe-and-vae-22c0901a6675

VAE MoE - plot samples.py

def plot_samples(samples, num_epochs):
    IMAGE_WIDTH = 0.7
    epochs = np.linspace(0, len(samples) - 1, num_epochs).astype(int)
    plt.figure(figsize=(IMAGE_WIDTH * NUM_DIGITS,
                        len(epochs) * IMAGE_WIDTH))
    for epoch_index, epoch in enumerate(epochs):
        for digit, image in enumerate(samples[epoch]):
            plt.subplot(len(epochs),
                        NUM_DIGITS,
                        epoch_index * NUM_DIGITS + digit + 1)
            plt.imshow(image.reshape((28, 28)),
                       cmap='Greys_r')
            plt.gca().xaxis.set_visible(False)
            if digit == 0:
                plt.gca().yaxis.set_ticks([])
                plt.ylabel('epoch {}'.format(epoch + 1),
                           verticalalignment='center',
                           horizontalalignment='right',
                           rotation=0,
                           fontsize=14)
            else:
                plt.gca().yaxis.set_visible(False)
                
                
plot_samples(samples=samples, num_epochs=20)