Saved model for inference

#1

Hi all,

I am new to pyTorch and curious to know how can I use the saved model during training for inference (Testing), especially for the GAN saved model.

I am following this repository [https://github.com/eriklindernoren/PyTorch-GAN]. In the all GAN models which implemented here the coder saved the whole models and now I want to use the saved model (for example, Vanilla GAN model) for inference but don’t have Idea how to do that,

Any help will be appreciated,
Thanks in advance!

1 Like
#2

https://pytorch.org/tutorials/beginner/saving_loading_models.html

#3

I already know how to load the saved model with torch.load() and my problem is not loading the model. For example, in CycleGAN one domain is translating to other domain (apple2orange). I trained my model based on training data and saved my model parameters and everything. Now I want to use the saved model and give a single image (apple image which network never has seen) to saved model and get the output in other domain.

1 Like
#4

Hey did you figure out how to do it ?? im stuck on the same thing please help