01-pytorch_workflow loss curve plotting fix with 'torch.inference_mode()'

[soumukhe]

Daniel, your YouTube course is 100% awesome!
Learning a lot and enjoying it.
Had a quick comment on 01-pytorch_workflow loss curve plotting. When you were showing the loss curve plotting error that occurred while plotting the graph, you converted the values from tensor to numpy arrays to fix the problem. As I was following you I stopped the video when the error occurred and tried to fix it before you showed how. I tried using 'with torch.inference_mode():' before plotting and that seemed to work.

error

[/usr/local/lib/python3.10/dist-packages/torch/_tensor.py](https://localhost:8080/#) in __array__(self, dtype)
   1028             return handle_torch_function(Tensor.__array__, (self,), self, dtype=dtype)
   1029         if dtype is None:
-> 1030             return self.numpy()
   1031         else:
   1032             return self.numpy().astype(dtype, copy=False)

RuntimeError: Can't call numpy() on Tensor that requires grad. Use tensor.detach().numpy() instead.

worked fine with use of torch.inference_mode()

# Plot the loss curves
with torch.inference_mode(): # turn off gradient decline function before plotting graph
  plt.plot(epoch_count, train_loss_values, label="Train loss")
  plt.plot(epoch_count, test_loss_values, label="Test loss")
  plt.title("training and test loss curves")
  plt.ylabel("Loss")
  plt.xlabel("epoch count")
  plt.legend();