search

dwofk / fast-depth

ICRA 2019 "FastDepth: Fast Monocular Depth Estimation on Embedded Systems"
MIT License
926 stars 189 forks source link

The Pretrained MobileNet has no "model " or "best_result" #44

Open Lightang6 opened 3 years ago

Lightang6 commented 3 years ago

I used the pretrained model directly , but there is a bug exist , the dict of Pretrained MobileNet has no "model " or "best_result"

niharsalunke commented 3 years ago

Hi @Lightang6, you are referring to the wrong file. Please download "mobilenet-nnconv5dw-skipadd-pruned.pth.tar" and then add the following command while loading the model: checkpoint = torch.load("mobilenet-nnconv5dw-skipadd-pruned.pth.tar")

I hope this helps.

PouyaNF commented 2 years ago

Hi @niharsalunke , when I want to load the model using torch.load("mobilenet-nnconv5dw-skipadd-pruned.pth.tar") I got the following error:

........ return super().find_class(mod_name, name) ModuleNotFoundError: No module named 'models'