How can we plot ROC Curve for leave one out (LOO) cross validation using scikit-learn?

Asked Jan 29 '19 at 10:36

Active Jan 29 '19 at 10:36

Viewed 674 times

On the scikit-learn website there is code for ROC curve for stratified kfold but no code is present for leave one out (LOO) cross validation. I tried by changing the code of kfold by loo but the result is nan. What can be the problem?

I checked scikit-learn website but no code is there for loo cross validation.

The code for stratified kfold is given in the link given below.

https://scikit-learn.org/stable/auto_examples/model_selection/plot_roc_crossval.html

asked Jan 29 '19 at 10:36

Kumar

Please show us the code and unexpected result explicitly. – rvf Jan 29 '19 at 12:58
Have you tried to set n_splits to the number of data points in your dataset? – Calimo Jan 29 '19 at 15:28
@Calimo Yes. I have tried this. Still the result is nun. No curve is displayed. – Kumar Jan 30 '19 at 16:24
@Kumar please show us the code you used and unexpected result explicitly, as rvf asked already. – Calimo Jan 30 '19 at 17:03
did you manage to find an answer for this? – seralouk Sep 02 '19 at 11:36
https://stackoverflow.com/questions/57756804/roc-curve-with-leave-one-out-cross-validation-in-sklearn – seralouk Sep 02 '19 at 12:20

How can we plot ROC Curve for leave one out (LOO) cross validation using scikit-learn?

0 Answers0

Linked