checkpoint_dir not found

[meeng251]

Hi @zsdonghao @Laicheng0830 I know after per 10 epochs by code line:

• if (epoch != 0) and (epoch % 10 == 0):

They will create a checkpoint file or maybe folder but I can't see anything except saving fake he paths in code

• tl.vis.save_images(fake_hr_patchs.numpy(), [2, 4], os.path.join(save_dir, 'train_ginit{}.png'.format(epoch)))

I think block ##initialize learning (G) is working and ##adversarial learning (G, D) is not working while run training code Could you help me solve this problem, I use for academic. Thank you!

My device information:

• Tensorflow 2.0.0, Tensowlayer 2.2.3, Numpy 1.16.1
• I5-8300H + GTX1050Ti 4GB