How to implement customised loss function in patternnet (pattern recognition neural network)?

I am trying to implement my own loss function in the second hidden layer for multiclass classification problem. can anyone tell me how to start with.

I am aware of how to increase the hidden layers but not about implementing lossfunction assisted hidden layer.

Thanks in advance

Raja

Answer 1

[Mandar]

I understand that you want to define custom loss function.

Refer to the following documentation link to know more about implementation of the custom loss function.

Link: https://in.mathworks.com/help/deeplearning/ug/define-custom-training-loops-loss-functions-and-networks.html

[RAJA SEKHAR BATTU]

@Mandar Thank you for the answer

I am aware of this link. I am trying to implement a custom loss function

when I change the loss function from cross entropy to my custom loss function

Net.perFuncn='customloss'

Its not working but giving error not enough input arguements

Can you provide any example of custom loss function for classification using patternnet or feedforwardnet

many thanks

Raja