cugraph.dask.all_pairs_sorensen#

cugraph.dask.all_pairs_sorensen( input_graph, vertices: Series = None, use_weight: bool = False, topk: int = None, )[source]#

Compute the All Pairs Sorensen similarity between all pairs of vertices specified. All pairs Sorensen coefficient is defined between two sets as the ratio of twice the volume of their intersection over the volume of each set. In the context of graphs, the neighborhood of a vertex is seen as a set. The Sorensen similarity weight of each edge represents the strength of connection between vertices based on the relative similarity of their neighbors.

cugraph.all_pairs_sorensen, in the absence of specified vertices, will compute the two_hop_neighbors of the entire graph to construct a vertex pair list and will return the sorensen coefficient for all the vertex pairs in the graph. This is not advisable as the vertex_pairs can grow exponentially with respect to the size of the datasets.

If the topk parameter is specified then the result will only contain the top k highest scoring results.

Parameters:

input_graphcugraph.Graph: cuGraph Graph instance, should contain the connectivity information as an edge list (edge weights are not supported yet for this algorithm). The graph should be undirected where an undirected edge is represented by a directed edge in both direction. The adjacency list will be computed if not already present.

This implementation only supports undirected, non-multi Graphs.
verticesint or list or cudf.Series, dask_cudf.Series, optional (default=None): A GPU Series containing the input vertex list. If the vertex list is not provided then the current implementation computes the sorensen coefficient for all vertices that are two hops apart in the graph.
use_weightbool, optional (default=False): Flag to indicate whether to compute weighted sorensen (if use_weight==True) or un-weighted sorensen (if use_weight==False). ‘input_graph’ must be weighted if ‘use_weight=True’.
topkint, optional (default=None): Specify the number of answers to return otherwise returns the entire solution

Returns:

resultdask_cudf.DataFrame

GPU distributed data frame containing 3 dask_cudf.Series

ddf[‘first’]: dask_cudf.Series: The first vertex ID of each pair (will be identical to first if specified).
ddf[‘second’]: dask_cudf.Series: The second vertex ID of each pair (will be identical to second if specified).
ddf[‘sorensen_coeff’]: dask_cudf.Series: The computed sorensen coefficient between the first and the second vertex ID.