[Python] Sklearn and cross_val_scores: can we use it for leave-one-out cross validation?

I am writing a loop where I have to execute the k-fold cross-validation on 1000 entries. The value of k changes at every iteration in the loop and it can have the value of 1000. However, when trying to do the leave-on-out cross-validation (i.e. k = #entries = 1000) I receive the following error:

ValueError: All the n_groups for individual classes are less than n_splits=1000.

test = [...,...,1000] for n in test:

knn = KNeighborsClassifier(n_neighbors=1)
scores = cross_val_score(knn, X_train, y_train, cv=n, scoring='accuracy')


Remember that X_train and y_train have 1000 rows and are numpy arrays. I know there are specific methods for the LOO but I am wondering whether there could be a way to use the same function in order to avoid inserting an if in the loop.

Continue reading...