How can I compute the precision-recall cuvre with OCSVM?

[RamziFsm]

sklearn.metrics.precision_recall_curve(y_true, probas_pred, *, pos_label=None, sample_weight=None) According to Sklearn, we should pass a probability prediction as a parameter for precision_recall_curve. But, the OCSVM doesn't have a function that returns a probability as output. How can I compute the precision-recall curve with OCSVM?

[zeyiwen]

ocsvm doesn't support probabilistic output, but it is possible to do by revising some code in C++. I have labelled this issue as "enhancement", and also call for contribution.