Hello I am trying to load yolo7-w6-pose.pt model in a separate project, in order to do this I start with the following code:
path = 'yolov7-w6-pose.pt'
model = torch.hub.load("WongKinYiu/yolov7","custom",f"{path}",trust_repo=True)
This code returns the following error:
219 self.no_det=(nc + 5) # number of outputs per anchor for box and class
220 self.no_kpt = 3*self.nkpt ## number of outputs per anchor for keypoints
--> 221 self.no = self.no_det+self.no_kpt
222 self.nl = len(anchors) # number of detection layers
223 self.na = len(anchors[0]) // 2 # number of anchors
TypeError: unsupported operand type(s) for +: 'int' and 'str'
After this I tried with restart the kernel and this code:
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
path = 'yolov7-w6-pose.pt'
#model = torch.hub.load("WongKinYiu/yolov7","custom",f"{path}",trust_repo=True)
weigths = torch.load(f"{path}")
model = weigths['model']
_ = model.float().eval()
if torch.cuda.is_available():
model.half().to(device)
but this return this error:
ModuleNotFoundError: No module named 'models'
because I don't have the models folder from the origin repository and I don't want to copy that folder into my current project.
Anyway if I restart the kernel again and I run the first piece of the code, and after that, I execute the second piece of the code, the model works, but I am not sure how, and I don't like put an extra try catch only to allow that the first code can fail in order to load a model.
Can I solve this and load the model clearly? why is failing? is there any way to fix this?
Thanks.
