-1

I have trained a yolov5l model for object detection and classification. I want to use the exported weights to identify images in a program I am creating. I am having trouble finding much of anything on how to use .pt weights in a python program.

I believe I use the "torch.load" method from the pytorch library, but when I try: torch.load(path_to_weights) I get a ModuleNotFoundError for not having a module named 'models'.

Any help is much appreciated. Thank you very much.

Nawx
  • 1
  • 2

1 Answers1

1

You should use torch.load_state_dict() method to load your trained parameters to your model in addition to torch.load().

There are some issues with your torch.load() method. You should provide your path parameter as a either string or os.PathLike object. (These are written in the docs).

I am going to provide a simple code block to show you the way.

#Initializing model

model = Model() # Assuming your model's name is Model

model.load_state_dict(torch.load(path_to_weights))

But don't forget that your path_to_weights parameter must be either string or os.PathLike object.

hus
  • 56
  • 5