How to use Pytorch pretrained model (.pth)

Viewed 31

https://github.com/ZJULearning/resa I have used above GitHub repo, for lane detection, it contains already pretrained model for lane Detection on CULane dataset. I wanted to use this pretrained model and test it on other data, but I am not sure, how to approach.

On loading data model = torch.load('culane_resnet50.pth', map_location=torch.device('cpu')) it is loaded succesfully.

but on using model.eval() it is showing an error " 'dict' object has no attribute 'eval' "

On printing model print(model) the output is something like this

output

I want to know what the model represents, Also is it possible to use this pretrained model to make prediction on new data sets and if yes, how?

0 Answers
Related