langchain-google-spanner 0.9.1

LangChain integrations for Google Cloud Spanner

• Client Library Documentation

• Product Documentation

Quick Start

In order to use this library, you first need to go through the following steps:

1. Select or create a Cloud Platform project.

2. Enable billing for your project.

3. Enable the Google Cloud Spanner API.

4. Setup Authentication.

Installation

Install this library in a virtualenv using pip. virtualenv is a tool to create isolated Python environments. The basic problem it addresses is one of dependencies and versions, and indirectly permissions.

With virtualenv, it’s possible to install this library without needing system install permissions, and without clashing with the installed system dependencies.

Supported Python Versions

Python >= 3.9

Mac/Linux

pip install virtualenv
virtualenv <your-env>
source <your-env>/bin/activate
<your-env>/bin/pip install langchain-google-spanner

Windows

pip install virtualenv
virtualenv <your-env>
<your-env>\Scripts\activate
<your-env>\Scripts\pip.exe install langchain-google-spanner

Vector Store Usage

Use a vector store to store embedded data and perform vector search.

from langchain_google_spanner import SpannerVectorstore
from langchain.embeddings import VertexAIEmbeddings

embeddings_service = VertexAIEmbeddings(model_name="textembedding-gecko@003")
vectorstore = SpannerVectorStore(
    instance_id="my-instance",
    database_id="my-database",
    table_name="my-table",
    embeddings=embedding_service
)

See the full Vector Store tutorial.

Document Loader Usage

Use a document loader to load data as LangChain Documents.

from langchain_google_spanner import SpannerLoader


 loader = SpannerLoader(
     instance_id="my-instance",
     database_id="my-database",
     query="SELECT * from my_table_name"
 )
 docs = loader.lazy_load()

See the full Document Loader tutorial.

Chat Message History Usage

Use ChatMessageHistory to store messages and provide conversation history to LLMs.

from langchain_google_spanner import SpannerChatMessageHistory


 history = SpannerChatMessageHistory(
     instance_id="my-instance",
     database_id="my-database",
     table_name="my_table_name",
     session_id="my-session_id"
 )

See the full Chat Message History tutorial.

Spanner Graph Store Usage

Use SpannerGraphStore to store nodes and edges extracted from documents.

from langchain_google_spanner import SpannerGraphStore


 graph = SpannerGraphStore(
     instance_id="my-instance",
     database_id="my-database",
     graph_name="my_graph",
 )

See the full Spanner Graph Store tutorial.

Spanner Graph QA Chain Usage

Use SpannerGraphQAChain for question answering over a graph stored in Spanner Graph.

from langchain_google_spanner import SpannerGraphStore, SpannerGraphQAChain
from langchain_google_vertexai import ChatVertexAI


graph = SpannerGraphStore(
    instance_id="my-instance",
    database_id="my-database",
    graph_name="my_graph",
)
llm = ChatVertexAI()
chain = SpannerGraphQAChain.from_llm(
    llm,
    graph=graph,
    allow_dangerous_requests=True
)
chain.invoke("query=Where does Sarah's sibling live?")

See the full Spanner Graph QA Chain tutorial.

Spanner Graph Retrievers Usage

Use SpannerGraphTextToGQLRetriever to translate natural language question to GQL and query SpannerGraphStore.

from langchain_google_spanner import SpannerGraphStore, SpannerGraphTextToGQLRetriever
from langchain_google_vertexai import ChatVertexAI


graph = SpannerGraphStore(
    instance_id="my-instance",
    database_id="my-database",
    graph_name="my_graph",
)
llm = ChatVertexAI()
retriever = SpannerGraphTextToGQLRetriever.from_params(
    graph_store=graph,
    llm=llm
)
retriever.invoke("Where does Elias Thorne's sibling live?")

Use SpannerGraphVectorContextRetriever to perform vector search on embeddings that are stored in the nodes in a SpannerGraphStore. If expand_by_hops is provided, the nodes and edges at a distance upto the expand_by_hops from the nodes found in the vector search will also be returned.

from langchain_google_spanner import SpannerGraphStore, SpannerGraphVectorContextRetriever
from langchain_google_vertexai import ChatVertexAI, VertexAIEmbeddings


graph = SpannerGraphStore(
    instance_id="my-instance",
    database_id="my-database",
    graph_name="my_graph",
)
embedding_service = VertexAIEmbeddings(model_name="text-embedding-004")
retriever = SpannerGraphVectorContextRetriever.from_params(
        graph_store=graph,
        embedding_service=embedding_service,
        label_expr="Person",
        embeddings_column="embeddings",
        top_k=1,
        expand_by_hops=1,
    )
retriever.invoke("Who lives in desert?")

Contributions

Contributions to this library are always welcome and highly encouraged.

See CONTRIBUTING for more information how to get started.

Please note that this project is released with a Contributor Code of Conduct. By participating in this project you agree to abide by its terms. See Code of Conduct for more information.

License

Apache 2.0 - See LICENSE for more information.

Disclaimer

This is not an officially supported Google product.

Limitations

• Approximate Nearest Neighbors (ANN) strategies are only supported for the GoogleSQL dialect

• ANN’s ALTER VECTOR INDEX is not yet supported by [Google Cloud Spanner](https://cloud.google.com/spanner/docs/find-approximate-nearest-neighbors#limitations)