neo4j-graphrag-python
neo4j-graphrag-python copied to clipboard
neo4j
Neo4j GraphRAG for Python
# Description README re-write. ## Type of Change - [ ] New feature - [ ] Bug fix - [ ] Breaking change - [X] Documentation update - [ ]...
# Description > **Note** > > Please provide a description of the work completed in this PR below > > ## Type of Change - [ ] New feature -...
# Description Update VectorCypherRetriever to improve its usability. ## Type of Change - [ ] New feature - [ ] Bug fix - [ ] Breaking change - [x] Documentation...
# Description > **Note** > > Please provide a description of the work completed in this PR below > > ## Type of Change - [ ] New feature -...
# Description > **Note** > > Please provide a description of the work completed in this PR below > > ## Type of Change - [ ] New feature -...
# Description > **Note** > > Please provide a description of the work completed in this PR below > > ## Type of Change - [ ] New feature -...
# Description This PR adds a `LexicalGraphBuilder` component that enables users to create the lexical graph without entity/relation extraction (see added example). It also updates the `ERExtraction` component to use...
Regarding stuff like this in kg writer: https://github.com/neo4j/neo4j-graphrag-python/blob/main/src/neo4j_graphrag/experimental/components/kg_writer.py#L103 For unique node ids/keys used in loading, you want to use a Constraint: Either [Key](https://neo4j.com/docs/cypher-manual/current/constraints/#node-key) or [Uniqueness](https://neo4j.com/docs/cypher-manual/current/constraints/#unique-node-property) constraint. Key is like uniqueness...
Often we just want to define node labels, or just node labels and relationship types. We may not know a good `potential_schema`, especially while developing on new data. Currently, these...
Currently, `create_vector_index` will fail if the index already exists. I also don't have a method in this package, AFAIK, to check if an index already exists. I think it's important...