[FEA]: Support creating a subgraph as a Property graph from an existing Property Graph

[acostadon]

Is this a new feature, an improvement, or a change to existing functionality?

New Feature

How would you describe the priority of this feature request

High

Please provide a clear description of problem this feature solves

a way to extract a subgraph or sample of a PropertyGraph object without converting it to a cugraph object.

Describe your ideal solution

the PropertySelection objects returned from e.g. select_vertices can be used to extract a new PropertyGraph

Describe any alternatives you have considered

No response

Additional context

No response

Code of Conduct

• I agree to follow cuGraph's Code of Conduct
• I have searched the open feature requests and have found no duplicates for this feature request

[acostadon]

Please notify Christian Conroy in the Rapids-goAi Slack when this is complete.

[rlratzel]

We may want to do this by allowing extract_subgraph to take a PropertyGraph for the create_using arg.

Here's a workaround I posted:

>>> from cugraph import Graph
>>> from cugraph.experimental import PropertyGraph
>>> import cudf

>>> df = cudf.DataFrame(columns=["src", "dst", "some_property"],
...                     data=[(99, 22, "a"),
...                           (98, 34, "b"),
...                           (97, 56, "c"),
...                           (96, 88, "d"),
...                     ])
>>> pG = PropertyGraph()
>>> pG.add_edge_data(df, vertex_col_names=("src", "dst"))
>>> pG.get_edge_data()
   _EDGE_ID_ some_property  _SRC_  _DST_ _TYPE_
0          0             a     99     22
1          1             b     98     34
2          2             c     97     56
3          3             d     96     88

>>> # Select edges 'a' and 'b'
>>> selection = pG.select_edges("(some_property=='a')|(some_property=='b')")

>>> # Extract a subgraph as a cugraph directed Graph, include the edge property
>>> # meta-data for use in future calls to annotate_dataframe()
>>> G2 = pG.extract_subgraph(selection=selection,
...                          create_using=Graph(directed=True),
...                          add_edge_data=True)

>>> # Create a DataFrame containing the property data for the extracted edges
>>> # referencing the original PropertyGraph from which it was extracted
>>> subgraph_edge_data = pG.annotate_dataframe(G2.edges(), G2, edge_vertex_col_names=["src", "dst"])

>>> # Create a new PropertyGraph using the extracted data
>>> subPG = PropertyGraph()
>>> subPG.add_edge_data(subgraph_edge_data,
...                     vertex_col_names=["src", "dst"],
...                     property_columns=["some_property"])
>>> subPG.get_edge_data()
   _EDGE_ID_ some_property _TYPE_  _SRC_  _DST_
0          0             b            98     34
1          1             a            99     22