How to use the leidenalg.SignificanceVertexPartition function in leidenalg

def setUp(self):
    super(SignificanceVertexPartitionTest, self).setUp();
    self.partition_type = leidenalg.SignificanceVertexPartition;

def test_aggregate_partition(self, graph):
      if 'weight' in graph.es.attributes() and self.partition_type != leidenalg.SignificanceVertexPartition:
        partition = self.partition_type(graph, weights='weight');
      else:
        partition = self.partition_type(graph);
      self.optimiser.move_nodes(partition);
      aggregate_partition = partition.aggregate_partition();
      self.assertAlmostEqual(
          partition.quality(),
          aggregate_partition.quality(),
          places=5,
          msg='Quality not equal for aggregate partition.');
      self.optimiser.move_nodes(aggregate_partition);
      partition.from_coarse_partition(aggregate_partition);
      self.assertAlmostEqual(
          partition.quality(),
          aggregate_partition.quality(),
          places=5,

if graph is None:
            knn_conn_mat = self._sdm.s_knn_connectivity_matrix(
                k=k, use_pca=use_pca, use_hnsw=use_hnsw,
                index_params=index_params, query_params=query_params,
                verbose=verbose)
            graph = SampleDistanceMatrix.knn_conn_mat_to_aff_graph(
                knn_conn_mat, aff_scale=aff_scale)

        if partition_method == "RBConfigurationVertexPartition":
            la_part_cls = la.RBConfigurationVertexPartition
        elif partition_method == "RBERVertexPartition":
            la_part_cls = la.RBERVertexPartition
        elif partition_method == "CPMVertexPartition":
            la_part_cls = la.CPMVertexPartition
        elif partition_method == "SignificanceVertexPartition":
            la_part_cls = la.SignificanceVertexPartition
        elif partition_method == "SurpriseVertexPartition":
            la_part_cls = la.SurpriseVertexPartition
        else:
            raise ValueError(
                "Unknown partition method: {}".format(partition_method))

        la_res = la.find_partition(graph, la.RBConfigurationVertexPartition,
                                   seed=random_state, weights='weight',
                                   resolution_parameter=resolution)
        # keep track of results and parameters
        self._graph = graph
        self._la_res = la_res
        self._labs = la_res.membership
        self._k = k
        self._use_pca = use_pca
        self._use_hnsw = use_hnsw