Node Sampling¶
-
class
RandomNodeSampler
(number_of_nodes: int = 100, seed: int = 42)[source]¶ An implementation of random node sampling. Nodes are sampled with uniform probability. “For details about the algorithm see this paper.”
- Parameters
-
sample
(graph: Union[networkx.classes.graph.Graph, networkit.graph.Graph]) → Union[networkx.classes.graph.Graph, networkit.graph.Graph][source]¶ Sampling nodes randomly.
- Arg types:
graph (NetworkX or NetworKit graph) - The graph to be sampled from.
- Return types:
new_graph (NetworkX or NetworKit graph) - The graph of sampled nodes.
-
class
DegreeBasedSampler
(number_of_nodes: int = 100, seed: int = 42)[source]¶ An implementation of degree based sampling. Nodes are sampled proportional to the degree centrality of nodes. “For details about the algorithm see this paper.”
- Parameters
-
sample
(graph: Union[networkx.classes.graph.Graph, networkit.graph.Graph]) → Union[networkx.classes.graph.Graph, networkit.graph.Graph][source]¶ Sampling nodes proportional to the degree.
- Arg types:
graph (NetworkX or NetworKit graph) - The graph to be sampled from.
- Return types:
new_graph (NetworkX or NetworKit graph) - The graph of sampled nodes.
-
class
PageRankBasedSampler
(number_of_nodes: int = 100, seed: int = 42, alpha: float = 0.85)[source]¶ An implementation of PageRank based sampling. Nodes are sampled proportional to the PageRank score of nodes. “For details about the algorithm see this paper.”
- Parameters
-
sample
(graph: Union[networkx.classes.graph.Graph, networkit.graph.Graph]) → Union[networkx.classes.graph.Graph, networkit.graph.Graph][source]¶ Sampling nodes randomly proportional to the normalized pagerank score.
- Arg types:
graph (NetworkX or NetworKit graph) - The graph to be sampled from.
- Return types:
new_graph (NetworkX or NetworKit graph) - The graph of sampled nodes.