llama-index packs chroma_autoretrieval integration
Project description
Chroma AutoRetrieval Pack
This LlamaPack inserts your data into chroma and insantiates an auto-retriever, which will use the LLM at runtime to set metadata filtering, top-k, and query string.
CLI Usage
You can download llamapacks directly using llamaindex-cli
, which comes installed with the llama-index
python package:
llamaindex-cli download-llamapack ChromaAutoretrievalPack --download-dir ./chroma_pack
You can then inspect the files at ./chroma_pack
and use them as a template for your own project!
Code Usage
You can download the pack to a the ./chroma_pack
directory:
from llama_index.core.llama_pack import download_llama_pack
# download and install dependencies
ChromaAutoretrievalPack = download_llama_pack(
"ChromaAutoretrievalPack", "./chroma_pack"
)
From here, you can use the pack, or inspect and modify the pack in ./chroma_pack
.
Then, you can set up the pack like so:
# setup pack arguments
from llama_index.core.vector_stores.types import MetadataInfo, VectorStoreInfo
vector_store_info = VectorStoreInfo(
content_info="brief biography of celebrities",
metadata_info=[
MetadataInfo(
name="category",
type="str",
description=(
"Category of the celebrity, one of [Sports Entertainment, Business, Music]"
),
),
],
)
import chromadb
client = chromadb.EphemeralClient()
nodes = [...]
# create the pack
chroma_pack = ChromaAutoretrievalPack(
collection_name="test",
vector_store_info=vector_store_index,
nodes=nodes,
client=client,
)
The run()
function is a light wrapper around query_engine.query()
.
response = chroma_pack.run("Tell me a bout a Music celebritiy.")
You can also use modules individually.
# use the retreiver
retriever = chroma_pack.retriever
nodes = retriever.retrieve("query_str")
# use the query engine
query_engine = chroma_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_chroma_autoretrieval-0.0.1.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | d24f7a1e8302d72f1aef1f61c90d3c15a994a49c0bedc75c9f24e31adb1fd93f |
|
MD5 | 332cf4c4481cefa02c38cb140da171da |
|
BLAKE2b-256 | c6acf1a4683dc51a204f84823528f318092379f5bdec454bf82a6e3fa31f1612 |
Hashes for llama_index_packs_chroma_autoretrieval-0.0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | e860e0e396d259d9aa4824e7828e399bf8439986d27a4f13ce91d573cb0a6184 |
|
MD5 | c0a6e70dd24b2595921b5ee4de192756 |
|
BLAKE2b-256 | 4730cf0360ddb973bde2d52be8a480aadfb442cca4d46ca191a361bcbcbf0c1f |