KeyError: Unable to find "net" key in the trained model from detect.py

[kkarnatak]

Hi Guys,

I am using this project on conda env with gpu configured. I was trying to just run the inference files first to try it out, but I get the following error:

Traceback (most recent call last):
File "c:\CULane\lanedet\tools\detect.py", line 86, in
process(args)
File "c:\CULane\lanedet\tools\detect.py", line 72, in process
detect = Detect(cfg)
File "c:\CULane\lanedet\tools\detect.py", line 24, in init
load_network(self.net, self.cfg.load_from)
File "c:\culane\lanedet\lanedet\utils\net_utils.py", line 48, in load_network
net.load_state_dict(pretrained_model['net'], strict=True)
KeyError: 'net'

I have used following command:
$ python detect.py' 'lanedet/configs/resa/resa34_culane.py' '--img' 'image\' '--load_from' 'C:\Users\blackbug\.cache\torch\hub\checkpoints\resnet34-333f7ec4.pth' '--savedir' './vis'

I tried to look at the model loaded from the downloaded resnet model file; it looks valid with all the trained layers, just "net" isnt part of the dictionary. Any help is appreciated! Thank you!