Are there any advantages or disadvantages of having multiple output nodes in a neural network compared to several?
For example, if a scenario can be represented by 10, 3, 2, or 1 output nodes, what is considered better? - obviously, it depends on how you want to present the results, but say that now it does not matter.
Or does the number of output nodes not affect the accuracy of the network, but just the computational time needed to train it?
Gjhix source
share