TypeError: only integer tensors of a single element can be converted to an index

[lakiet1609]

Why after i load this code:

    model_path = r'models\model_2.pth'
    loaded_model_2 = cnn_model.custom_model(input_param=channels, output_param=num_classes) 
    # Load in the saved state_dict()
    loaded_model_2.load_state_dict(torch.load(f=model_path))
    # Send model to GPU
    loaded_model_2 = loaded_model_2.to(device)

Then i use pred_and_plot_image.

It has the type error like the titile and it happens the error at this line of pred_and_plot_image
function:

plt.title(f"Pred: {class_names[target_image_pred_label]} | Prob: {target_image_pred_probs.max():.3f}")

[mrdbourke]

Hi @lakiet1609 ,

Did you check the value of target_image_pred_label?

It may be because it's not a valid index.

I'd step through your code to make sure the inputs and outputs are in the right data format.

E.g. it looks like target_image_pred_label should be an integer rather than a torch.Tensor.