Suppose I have data with two independent variable $X_1$, $X_2$ and one dependent variable say $y$, as follows:

$X_1$: $x_{1,1}$, $x_{1,2}$ , $x_{1,3}$

$X_2$: $x_{2,1}$, $x_{2,2}$, $x_{2,3}$

$y$: $y_1$, $y_2$, $y_3$

I built some Machine learning model which is good .

Now I want to generate predictions not just for test data but for all possible combinations of test data for example, if our test data looks like

$X_1$: $a$, $b$, $c$

$X_2$: $p$, $q$, $r$

then I want predictions for pairs $(a,p)$$(a,q)$,$(a,r)$,$(b,p)$,$(b,q)$....etc

I have tried np.ravel, Meshgrid kind of commands but find it difficult.

  • $\begingroup$ Please be clear in which way the $y$ interacts with the $X$s, so we can help you further $\endgroup$ – Juan Esteban de la Calle Apr 27 '19 at 15:08
  • $\begingroup$ I ran ANN Regression ..There is no explicit equation $\endgroup$ – Milan Amrut Joshi Apr 27 '19 at 15:16

It is possible.

It is called Multi-output regression. Allows you to predict more that one variable:

Multioutput regression

| improve this answer | |

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.