llama-index packs nebulagraph_query_engine integration
Project description
NebulaGraph Query Engine Pack
This LlamaPack creates a NebulaGraph query engine, and executes its query
function. This pack offers the option of creating multiple types of query engines, namely:
- Knowledge graph vector-based entity retrieval (default if no query engine type option is provided)
- Knowledge graph keyword-based entity retrieval
- Knowledge graph hybrid entity retrieval
- Raw vector index retrieval
- Custom combo query engine (vector similarity + KG entity retrieval)
- KnowledgeGraphQueryEngine
- KnowledgeGraphRAGRetriever
CLI Usage
You can download llamapacks directly using llamaindex-cli
, which comes installed with the llama-index
python package:
llamaindex-cli download-llamapack NebulaGraphQueryEnginePack --download-dir ./nebulagraph_pack
You can then inspect the files at ./nebulagraph_pack
and use them as a template for your own project!
Code Usage
You can download the pack to a ./nebulagraph_pack
directory:
from llama_index.core.llama_pack import download_llama_pack
# download and install dependencies
NebulaGraphQueryEnginePack = download_llama_pack(
"NebulaGraphQueryEnginePack", "./nebulagraph_pack"
)
From here, you can use the pack, or inspect and modify the pack in ./nebulagraph_pack
.
Then, you can set up the pack like so:
# Load the docs (example of Paleo diet from Wikipedia)
from llama_index import download_loader
WikipediaReader = download_loader("WikipediaReader")
loader = WikipediaReader()
docs = loader.load_data(pages=["Paleolithic diet"], auto_suggest=False)
print(f"Loaded {len(docs)} documents")
# get NebulaGraph credentials (assume it's stored in credentials.json)
with open("credentials.json") as f:
nebulagraph_connection_params = json.load(f)
username = nebulagraph_connection_params["username"]
password = nebulagraph_connection_params["password"]
ip_and_port = nebulagraph_connection_params["ip_and_port"]
space_name = "paleo_diet"
edge_types, rel_prop_names = ["relationship"], ["relationship"]
tags = ["entity"]
max_triplets_per_chunk = 10
# create the pack
nebulagraph_pack = NebulaGraphQueryEnginePack(
username=username,
password=password,
ip_and_port=ip_and_port,
space_name=space_name,
edge_types=edge_types,
rel_prop_names=rel_prop_names,
tags=tags,
max_triplets_per_chunk=max_triplets_per_chunk,
docs=docs,
)
Optionally, you can pass in the query_engine_type
from NebulaGraphQueryEngineType
to construct NebulaGraphQueryEnginePack
. If query_engine_type
is not defined, it defaults to Knowledge Graph vector based entity retrieval.
from llama_hub.llama_packs.nebulagraph_query_engine.base import (
NebulaGraphQueryEngineType,
)
# create the pack
nebulagraph_pack = NebulaGraphQueryEnginePack(
username=username,
password=password,
ip_and_port=ip_and_port,
space_name=space_name,
edge_types=edge_types,
rel_prop_names=rel_prop_names,
tags=tags,
max_triplets_per_chunk=max_triplets_per_chunk,
docs=docs,
query_engine_type=NebulaGraphQueryEngineType.KG_HYBRID,
)
NebulaGraphQueryEnginePack
is a enum defined as follows:
class NebulaGraphQueryEngineType(str, Enum):
"""NebulaGraph query engine type"""
KG_KEYWORD = "keyword"
KG_HYBRID = "hybrid"
RAW_VECTOR = "vector"
RAW_VECTOR_KG_COMBO = "vector_kg"
KG_QE = "KnowledgeGraphQueryEngine"
KG_RAG_RETRIEVER = "KnowledgeGraphRAGRetriever"
The run()
function is a light wrapper around query_engine.query()
, see a sample query below.
response = nebulagraph_pack.run("Tell me about the benefits of paleo diet.")
You can also use modules individually.
# call the query_engine.query()
query_engine = nebulagraph_pack.query_engine
response = query_engine.query("query_str")
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for llama_index_packs_nebulagraph_query_engine-0.0.1.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 9791d69b8124da9e1873f741abedcc33c596122b03d201b18e44a80bac7b9e25 |
|
MD5 | e4bfde6d39726e7431e84d038fcbd4a2 |
|
BLAKE2b-256 | 937050d1a1d6b86e5ea7f0638abadf297caafca360c37d45b3c2d8736c5428e8 |
Hashes for llama_index_packs_nebulagraph_query_engine-0.0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 86203915e349b7ed9dbc5077684e7412dc007f783c89cf4a80cc2d5377ff5b6a |
|
MD5 | becde55e81661cc41a1bb5a13a9934aa |
|
BLAKE2b-256 | 58bb3dda7263a8af17570a8e7c48cdd7a7d64260a37c3574aed821ff029e35e5 |