Which function to use for my neural network?

Not sure which one of these functions best applies to my neural network model:

feedforwardnet, fitnet, cascadeforwardnet, newff

Not sure when each of the above functions should be applied

the network has 11 input parameters and attempts to predict one output. the 11 inputs each represent different chemical properties for different MOF's.

Answer 1

[Jinal]

Hello Emer,

As per my understanding, you need to know which of, “feedforwardnet”, “fitnet”, "cascadeforwardnet", and “newff’ functions, can be used to create a neural network with eleven inputs and one output.

The choice of the appropriate neural network function in MATLAB depends on the complexity of the problem.

You can use “feedforwardnet” or “fitnet” to create your neural network.

Both functions are used for creating and training neural networks. However, "fitnet” is used for specialized versions of the feedforward network including fitting.

Please refer the following links to know more about “feedforwardnet”, “fitnet”, and difference between the two functions.

• feedforwardnet: https://www.mathworks.com/help/deeplearning/ref/feedforwardnet.html
• finet: https://www.mathworks.com/help/deeplearning/ref/fitnet.html
• Difference between “feedforwardnet” and “fitnet”: https://www.mathworks.com/matlabcentral/answers/35772-what-is-the-difference-between-feedforwardnet-with-fitnet

I hope this helps answer your query.

[Emer Gorman]

Hi Jinal,

Thank you for you help, I have looked at Matlab help files already. However, I can't seem to find detailed information on when each of the above functions should be applied.