pylibcugraph.all_pairs_cosine_coefficients#
- pylibcugraph.all_pairs_cosine_coefficients(
- ResourceHandle resource_handle,
- _GPUGraph graph,
- vertices,
- bool_t use_weight,
- topk,
- bool_t do_expensive_check,
Perform All-Pairs Cosine similarity computation.
Note that Cosine similarity must run on a symmetric graph.
- Parameters:
- resource_handleResourceHandle
Handle to the underlying device resources needed for referencing data and running algorithms.
- graphSGGraph or MGGraph
The input graph, for either Single or Multi-GPU operations.
- verticescudf.Series or None
- Vertex list to compute all-pairs. If None, then compute based
on all vertices in the graph.
- use_weightbool, optional
- If set to True, then compute weighted cosine_coefficients(
the input graph must be weighted in that case).
Otherwise, compute non-weighted cosine_coefficients
- topksize_t
Specify the number of answers to return otherwise will return all values.
- do_expensive_checkbool
If True, performs more extensive tests on the inputs to ensure validitity, at the expense of increased run time.
- Returns:
- A tuple of device arrays containing the vertex pairs with
- their corresponding Cosine coefficient scores.
Examples
# FIXME: No example yet