search

Gauravv97 / flownet2-Colab

Google Colab notebook for running Nvidia flownet2-pytorch
29 stars 2 forks source link

Problems to run the code #6

Closed SaraLCO99 closed 2 years ago

SaraLCO99 commented 2 years ago

Hi! I am trying to run the code but I have problems in the part of "Generate .flo files using FlowNet2" because when it is trying to Build FlowNet2 model it says that can not Loading checkpoint './FlowNet2_checkpoint.pth.tar' and the operation is failed also said that:

Traceback (most recent call last): File "main.py", line 210, in checkpoint = torch.load(args.resume) File "/usr/local/lib/python3.7/dist-packages/torch/serialization.py", line 367, in load return _load(f, map_location, pickle_module) File "/usr/local/lib/python3.7/dist-packages/torch/serialization.py", line 528, in _load magic_number = pickle_module.load(f) _pickle.UnpicklingError: invalid load key, '<'.

when I go to the serialization.py there are errors on: from ._utils import _import_dotted_name from ._six import string_classes as _string_classes It said that ._utils and ._six can not be resolved

I am running on cuda_11.1 I do not know if can be a conflict with the versions. I would like to know if you can help me to know what can I do.

Thanks in advance,

Regards,

Gauravv97 commented 2 years ago

Hi @SaraLCO99, It seems that the model is not correctly being downloaded from google drive.

I suggest you download & upload it directly or create a new link from your personal Drive.

Regards,

SaraLCO99 commented 2 years ago

Hi, it works thank youuu!!

Regards,