gds.degree.stream not respecting orientation?

[bSharpCyclist]

Describe the bug
I'm trying to reproduce a similar result that I would get in GDS.

To Reproduce
Cypher code:

call gds.graph.project('xdc-test-search', ['Event','Search'], { HAS_SEARCH:{orientation:'REVERSE'}})
call gds.degree.stream('xdc-test-search')
YIELD nodeId, score
return gds.util.asNode(nodeId).search_name_1 as SearchTerm, score As NumberOfSearches
Order by NumberOfSearches Descending, SearchTerm Limit 10;

The above returns the Search nodes.

Python code:

node_projection = ["Event","Search"]
relationship_projection = {"HAS_SEARCH": {"orientation": "REVERSE"}}
G, _ = gds.graph.project("xdc-test-search", node_projection, relationship_projection)
degree_stream = gds.degree.stream(G)

degree.stream above is returning the node ids of the Event nodes (other side of direction), even if I were to use G = gds.graph.get("xdc-test-search") and use the projection created with cypher.

I'm sure it's something I'm doing on my end :)

graphdatascience library version: 1.3
GDS plugin version: 2.1.7
Python version: 3.9.12
Neo4j version: 4.4.10
Operating system: macOS 12.6

The image below is what I would expect from the python approach.

[adamnsch]

Hi @bSharpCyclist,

The Python code gds.degree.stream(G) is really just calling call gds.degree.stream('xdc-test-search') and unpacking the result into a dataframe under the hood, so I'm surprised there's a difference.

What's the code for parsing the result of the Python call to gds.degree.stream? How do you know what you see are "Event" nodes?

Adam

[bSharpCyclist]

Hi @adamnsch

Thanks for the response. In the dataframe that is returned from gds.degree.stream there is a column of nodeIds, and a score column of all zeros, which what I would suspect for that direction of the relationship. Events are unique and will only point to one Search configuration node, but a Search node can have many edges coming into it. After using gds.util.asNode(nodeId) on one of the nodeIds and taking a look at that result, I can see that the node is an Event node.

[adamnsch]

Hi @adamnsch

Thanks for the response. In the dataframe that is returned from gds.degree.stream there is a column of nodeIds, and a score column of all zeros, which what I would suspect for that direction of the relationship. Events are unique and will only point to one Search configuration node, but a Search node can have many edges coming into it. After using gds.util.asNode(nodeId) on one of the nodeIds and taking a look at that result, I can see that the node is an Event node.

Ok, thanks for your reply.

When calling gds.degree.stream (either through Cypher or the Python client) you will get a list of all nodes, that is, both "Event" and "Search" nodes will show up in the results. It's therefore certainly possible that gds.util.asNode(nodeId) on one of the nodes will resolve to an "Event" node, using either method. You say the "score" column is all zeroes, but did you check every row?
In your Cypher query you order things so that "Event" nodes are likely at the bottom? "score" or "NumberOfSearches" are perhaps null for "Event" nodes? And since you limit how many results you see you thus don't see them at all?

Adam

[bSharpCyclist]

OMG facepalm, I just assumed they were all event nodes. That's exactly it 😊 Thanks!!