I've trained a neural network network that given a minmax normalized input, provides a minmax normalized output.

this might be late, but is it possible from a minmax normalized output to create the actual output, given you know the actual min and max value?

so an unormalized output?


Yes, just rearrange the formula for finding the normalised value. Where $x_i$ is the original attribute, and $z_i$ is the normalised value:

$$ \begin{align} z_i &= \frac{x_i - min(x)}{max(x) - min(x)} \\ x_i &= z_i(max(x) - min(x)) + min(x) \end{align} $$


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.