0
$\begingroup$

If we have two cross validation from KNN and linear regression like this:

KNN:

[0.33762693  0.20510195  0.27689527  0.24034536  0.21975643  0.32372453
 0.25537508  0.22416264  0.29873961  0.27207764]
KNN Variance Accuracy:0.0018054681226660197
KNN Average Accuracy: 26.54% 

Linear Regression:

[0.53606717 0.33177835 0.51892388 0.41194928 0.42048795 0.55997986
 0.45753065 0.41717226 0.44168116 0.46072667]
Linear Regression Variance Accuracy:0.004155050225436886
Linear Regression Average Accuracy: 45.56% 

Can we say that fold 1 and 6 have the highest accuracies and it means that the data at that fold has more important for prediction of target value?

What can we say about the variance in these two cases?

$\endgroup$
  • $\begingroup$ Your question on variance is kind of broad. Also, it makes no sense to pick data from validation folds to make predictions. These folds only make sense aggregated, by picking a specific one you're bound to a lot of bias, which is exactly what you are trying to get rid of using CV. $\endgroup$ – Lucas Farias Mar 22 at 20:25

Your Answer

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

Browse other questions tagged or ask your own question.