Chroma

Setup:

Install chromadb, langchain-chroma packages:

pip install -qU chromadb langchain-chroma

Key init args — indexing params: collection_name: Name of the collection. embedding_function: Embedding function to use.

Key init args — client params: client: Chroma client to use. client_settings: Chroma client settings. persist_directory: Directory to persist the collection. host: Hostname of a deployed Chroma server. port: Connection port for a deployed Chroma server. Default is 8000. ssl: Whether to establish an SSL connection with a deployed Chroma server. Default is False. headers: HTTP headers to send to a deployed Chroma server. chroma_cloud_api_key: Chroma Cloud API key. tenant: Tenant ID. Required for Chroma Cloud connections. Default is 'default_tenant' for local Chroma servers. database: Database name. Required for Chroma Cloud connections. Default is 'default_database'.

Instantiate:

from langchain_chroma import Chroma
from langchain_openai import OpenAIEmbeddings

vector_store = Chroma(
    collection_name="foo",
    embedding_function=OpenAIEmbeddings(),
    # other params...
)

Add Documents:

from langchain_core.documents import Document

document_1 = Document(page_content="foo", metadata={"baz": "bar"})
document_2 = Document(page_content="thud", metadata={"bar": "baz"})
document_3 = Document(page_content="i will be deleted :(")

documents = [document_1, document_2, document_3]
ids = ["1", "2", "3"]
vector_store.add_documents(documents=documents, ids=ids)

Update Documents:

updated_document = Document(
    page_content="qux",
    metadata={"bar": "baz"},
)

vector_store.update_documents(ids=["1"], documents=[updated_document])

Delete Documents:

vector_store.delete(ids=["3"])

Search:

results = vector_store.similarity_search(query="thud", k=1)
for doc in results:
    print(f"* {doc.page_content} [{doc.metadata}]")

*thud[{"baz": "bar"}]

Search with filter:

results = vector_store.similarity_search(
    query="thud", k=1, filter={"baz": "bar"}
)
for doc in results:
    print(f"* {doc.page_content} [{doc.metadata}]")

*foo[{"baz": "bar"}]

Search with score:

results = vector_store.similarity_search_with_score(query="qux", k=1)
for doc, score in results:
    print(f"* [SIM={score:3f}] {doc.page_content} [{doc.metadata}]")

* [SIM=0.000000] qux [{'bar': 'baz', 'baz': 'bar'}]

Async:

# add documents
# await vector_store.aadd_documents(documents=documents, ids=ids)

# delete documents
# await vector_store.adelete(ids=["3"])

# search
# results = vector_store.asimilarity_search(query="thud",k=1)

# search with score
results = await vector_store.asimilarity_search_with_score(query="qux", k=1)
for doc, score in results:
    print(f"* [SIM={score:3f}] {doc.page_content} [{doc.metadata}]")

* [SIM=0.335463] foo [{'baz': 'bar'}]

Use as Retriever:

retriever = vector_store.as_retriever(
    search_type="mmr",
    search_kwargs={"k": 1, "fetch_k": 2, "lambda_mult": 0.5},
)
retriever.invoke("thud")

[Document(metadata={"baz": "bar"}, page_content="thud")]