How to predict output of one image in Pytorch

So me and y friends wanted to make a webapp to be hosted in heroku but heroku is not supporting latest version of pytorch in its build. So we are not able to use " model.load_state_dict(torch.load(‘’)) " in our code for predicting the output of one image. Is there any way we can do this for torch 1.3.1?? Thanks in advace

Are you seeing an error while trying to load the state_dict because it’s incompatible with the old PyTorch version? If so, could you post the error message?