Error with loading model_____01_PyTorch_workflow______TypeError: Expected state_dict to be dict-like, got <class '__main__.LinearRegressionModel'>

[Arun-Kumar-Rathod]

When i am trying to load the state_dict of my previous model into a new instance of that class(as the code shown below)

# To load in a saved state dict we have to instantiate a new instance of our model class
loaded_model_0 = LinearRegressionModel()

# Load the saved state dict of model 0 (this will update the new instance with updated parameters)
loaded_model_0.load_state_dict(torch.load(f=MODEL_SAVE_PATH))

I am getting this error

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
[<ipython-input-44-d71b900f1161>](https://localhost:8080/#) in <cell line: 5>()
      3 
      4 # Load the saved state dict of model 0 (this will update the new instance with updated parameters)
----> 5 loaded_model_0.load_state_dict(torch.load(f=MODEL_SAVE_PATH))

[/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py](https://localhost:8080/#) in load_state_dict(self, state_dict, strict, assign)
   2101         """
   2102         if not isinstance(state_dict, Mapping):
-> 2103             raise TypeError(f"Expected state_dict to be dict-like, got {type(state_dict)}.")
   2104 
   2105         missing_keys: List[str] = []

TypeError: Expected state_dict to be dict-like, got <class '__main__.LinearRegressionModel'>.

I was trying to do this step of the video where I was trying to learn how to load a model
video link at exact timeframe

Please help me navigate this problem. Thanks.

[supattavir]

Hi

Make sure that you added the state_dict() when you create the torch.save.
it should be

torch.save(obj=model_0.state_dict(), 
           f=MODEL_SAVE_PATH)

If you did not add state_dict(), you will run to that error.

Hope this helps.