Load a pretrained .pth generative ProGAN model

I have a .pth model as title says, how can I input a latent vector in the the model and show a correspond images?

Assuming the *.pth file contains the state_dict of the model, you would need to create an instance of the model and load the state_dict afterwards. Once this is done, you can pass an input tensor to the model to get the output:

model = ProGAN()
state_dict = torch.load('PATH_TO_STATE_DICT/file.pth')