798882c255701943d530d7024d0da2db22e33cc0,pyod/models/ocsvm.py,OCSVM,fit,#OCSVM#Any#Any#Any#,46

Before Change

        self.detector_.fit(X=X_train, y=y, sample_weight=sample_weight,
                           **params)
        // invert scores. Outliers comes with higher scores
        self.decision_scores = self.detector_.decision_function(X_train) * -1
        self.threshold_ = scoreatpercentile(self.decision_scores,
                                            100 * (1 - self.contamination))
        self.y_pred = (self.decision_scores > self.threshold_).astype("int").ravel()

        return self

    def decision_function(self, X):

After Change

                                     max_iter=self.max_iter,
                                     random_state=self.random_state)

    def fit(self, X_train, y=None, sample_weight=None, **params):
        self._isfitted = True
        self.detector_.fit(X=X_train, y=y, sample_weight=sample_weight,
                           **params)
        // invert decision_scores. Outliers comes with higher decision_scores
        self.decision_scores = self.detector_.decision_function(X_train) * -1
        self._process_decision_scores()
        return self

    def decision_function(self, X):
        if not self._isfitted:

Instances