# Modelling of output neuron for mixed features?

Dataset in artificial intelligence, in general, consists of some features (say $$n$$). Assume that $$m$$ among them are output features. I want to model this function using neural network. So, input to my neural network is $$n-m$$ features and output is $$m$$ features. My question is about the output features.

If an output feature is a continuous random variable, then its corresponding output neuron can be trained to give continuous output. Similarly, if an output feature is a discrete random variable, then its corresponding output neuron can be trained to give discrete output.

But, I never came across the features that are mixed random variables. What is the nature of output of the neuron that is intended to give the output value for mixed random variable, which is neither discrete nor continuous in nature?

• Can you please provide an example of a feature that is neither continuous nor discrete? Because, right now, I don't understand what you mean by that. Moreover, it may be a good idea to explain why you're interested in this question. Which problem are you having that involves such "neither continuous nor discrete features" (whatever that means)?
– nbro
Sep 6 at 11:17
• Okay @nbro, will update. Thanks. Sep 6 at 11:28