Skip to content
This repository was archived by the owner on Jul 2, 2023. It is now read-only.

Pretrained model 404 #160

Open
BlackDriver opened this issue May 30, 2021 · 4 comments
Open

Pretrained model 404 #160

BlackDriver opened this issue May 30, 2021 · 4 comments

Comments

@BlackDriver
Copy link

I tried to open http://drive.google.com/uc?id=0B9P1L--7Wd2vVGE3TkRMbWlNRms but get 404 for result

@oksidgy
Copy link

oksidgy commented Jun 11, 2021

I downloaded weights from the torch official hub.
Simply It needs to set parameter "pretrained" == True: torchvision.models.vgg16(pretrained=True)

In torchfcn/models/vgg.py

def VGG16(pretrained=False):
    model = torchvision.models.vgg16(pretrained=True)
    if not pretrained:
        return model


    #model_file = _get_vgg16_pretrained_model()
    #state_dict = torch.load(model_file)
    #model.load_state_dict(state_dict)
    return model

@xiangsanliu
Copy link

same problem

@sss07
Copy link

sss07 commented Jun 24, 2021

I try to download the pretrained FCN models but get 404 for result.

@ArthurRyan0803
Copy link

I downloaded weights from the torch official hub.
Simply It needs to set parameter "pretrained" == True: torchvision.models.vgg16(pretrained=True)

In torchfcn/models/vgg.py

def VGG16(pretrained=False):
    model = torchvision.models.vgg16(pretrained=True)
    if not pretrained:
        return model


    #model_file = _get_vgg16_pretrained_model()
    #state_dict = torch.load(model_file)
    #model.load_state_dict(state_dict)
    return model

What's the difference between the pytorch offical pretrained model and the model provided here?

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

5 participants