bomanimc
commented
3 years ago @NathanK20 thanks for your question! I would like to understand more about your use case. What are you trying to predict with your model? My understanding is that you would need to trIn a model first before it has the ability to make reasonable predictions on any data, so I'm not sure I can imagine the case where you'd want to do predictions with an untrained model.
I am trying to initialize a very simple network with random weights and biases that I can simply use to make a prediction on a given input. I do not want to train the network, I just want it to be able to make predictions. Everywhere I look, the documentation calls for adding data and training and I cant get the network to predict without doing the former.