Activations of every layer

[neural-everything]

Hi @lululxvi ,
Is there a way to get the activation values at every layer of the network? I am trying to investigate the effect of scaling the PDE on training behaviour. Thanks.

[lululxvi]

Use Model.print_model() or Model.state_dict()

[ruchitkini]

Hi,
I tried using model.print_model() command but I am getting the following error: NotImplementedError: state_dict hasn't been implemented for this backend. I am using pytorch as the backend. I am using this command after I call the stored trained model.

[lululxvi]

Use Model.state_dict()