How to use the retentioneering.core.clustering function in retentioneering

targets = self.get_clusters(plot_type=None, **kwargs)
        elif plot_type == 'targets':
            targets = self._tsne_targets
        else:
            return self._tsne
        if proj_type == '3d':
            plot.tsne_3d(
                self._obj,
                clustering.aggregate_cl(targets, 7) if kwargs.get('method') == 'dbscan' else targets,
                old_targs,
                **kwargs
            )
        else:
            plot.cluster_tsne(
                self._obj,
                clustering.aggregate_cl(targets, 7) if kwargs.get('method') == 'dbscan' else targets,
                targets,
                **kwargs
            )
        return self._tsne

Parameters for ``sklearn.decomposition()`` and ``sklearn.manifold()`` methods. Keyword arguments for clusterers. For more information, please, see ``sklearn.cluster.KMeans()``, ``sklearn.cluster.DBSCAN()``, ``sklearn.mixture.GaussianMixture()`` docs.

        Returns
        -------
        Array of clusters

        Return type
        -------
        np.array
        """
        if hasattr(self, 'datatype') and self.datatype == 'features':
            features = self._obj.copy()
        else:
            features = self.extract_features(**kwargs)
        if not hasattr(self, 'clusters') or refit_cluster:
            clusterer = getattr(clustering, method)
            self.clusters, self._metrics = clusterer(features, **kwargs)
            self._create_cluster_mapping(features.index.values)

        if hasattr(self, 'datatype') and self.datatype == 'features':
            target = kwargs.pop('target')
        else:
            target = self.get_positive_users(**kwargs)
        target = features.index.isin(target)
        self._metrics['homogen'] = clustering.homogeneity_score(target, self.clusters)
        if hasattr(self, '_tsne'):
            features.retention._tsne = self._tsne
        if plot_type:
            func = getattr(plot, plot_type)
            res = func(
                features,
                clustering.aggregate_cl(self.clusters, 7) if method == 'dbscan' else self.clusters,