I want to implement Multilayer perceptron for software effort estimation. Which function should I use feedforwardnet, fitnet or something else.

My input may be continuous or categorial. but target is always a continuous no.

Answer 1

[Maria Duarte Rosa]

For continuous outputs both fitnet and feedforwardnet are equivalent and the natural choice.

Perhaps an easier way to get strated is using the app:

https://uk.mathworks.com/help/deeplearning/gs/fit-data-with-a-neural-network.html

If further customization is needed then 'network' allows one to build more flexible networks:

https://uk.mathworks.com/help/deeplearning/ref/network.html

Note: for deep learning networks a good way to get sarted is by using the Deep Network Designer app:

https://uk.mathworks.com/help/deeplearning/gs/get-started-with-deep-network-designer.html

I hope this helps.

[sushma khatri]

in case of feedforwardnet target must be in 1/0 form.

so how can I use it for continuous target

[Maria Duarte Rosa]

Hi Sushma,

The targets for feedforwardnet are continuous. Please see here:

https://uk.mathworks.com/help/deeplearning/ref/feedforwardnet.html

[x,t] = simplefit_dataset;
net = feedforwardnet(10);
net = train(net,x,t);

t are continuous.

Petternnet is for categorical targets, see here:

https://uk.mathworks.com/help/deeplearning/ref/patternnet.html

[x,t] = iris_dataset;
net = patternnet(10);
net = train(net,x,t);

t in this case is the 1/0 form.

[sushma khatri]

Thank u so much.

what is the basic difference between feedforwardnet and fitnet.

Actually I want to predict software effort based on some features.

effort is always numeric but input may be continuous or categorial.

So I m confuse which tool should I use nftool , nprtool or ntstool

Answer 2

[sushma khatri]

Thank u so much.

what is the basic difference between feedforwardnet and fitnet.

Actually I want to predict software effort based on some features.

effort is always numeric but input may be continuous or categorial.

So I m confuse which tool should I use nftool , nprtool or ntstool

[Maria Duarte Rosa]

fitnet and feedforwardnet are equivalent. You can use one or the other. If 'effort' is always continuous then fitnet or feedforwardnet seems to me to be the most natural choice for your task.

nftool is the most appropriate UI for modelling continous outputs.

nprtool is for patternnet where the output is categorical.

ntstool is for modelling time-series, using narxnet and other similar networks.

[sushma khatri]

Awesome it worked

thank u so much