nan output

[Unknown]

Hi,
I use fuzzylite to implement a fuzzy system. I have three inputs and 1 output. For some inputs’ values, I obtain nan as an output ( I verified that all the input values are all in the predefined ranges). And I define all the possible combination of rules ( 81 rules).
I use mandani for the fuzzy inference system and I set centroid as a deffuzifier.
What are the possible reasons to obtain nan as output?

[Juan Rada-Vilela (admin)]

Hi,

Thanks for your post and for using fuzzylite.

You get nan because it is the default value of the output variable (which you can change) and no rules are activated for the given input.

You would see this easily using QtFuzzyLite.

Cheers.

[Author]

Posts