8.2.10 Inferring Embeddings for a Supervised GraphWise Model
You can use a trained model to infer embeddings for unseen nodes and store in the database as described in the following code:
opg4j> var vertexVectors = model.inferEmbeddings(fullGraph, testVertices).flattenAll()
opg4j> vertexVectors.write().
db().
name("vertex vectors").
tablename("vertexVectors").
overwrite(true).
store()
PgxFrame vertexVectors = model.inferEmbeddings(fullGraph,testVertices).flattenAll();
vertexVectors.write()
.db()
.name("vertex vectors")
.tablename("vertexVectors")
.overwrite(true)
.store();
vertex_vectors = model.infer_embeddings(full_graph, test_vertices).flatten_all()
vertex_vectors.write().db().table_name("table_name").name("vertex_vectors").overwrite(True).store()
The schema for the
vertexVectors
will be as follows
without flattening (flattenAll
splits the vector column into
separate double-valued
columns):+---------------------------------------------------------------+
| vertexId | embedding |
+---------------------------------------------------------------+
Note:
All the preceding examples assume that you are inferring the embeddings for a model in the current logged in database. If you must infer embeddings for the model in a different database then refer to the examples in Inferring Embeddings for a Model in Another Database.