embedstore

Python wrapper to access embeddings from embedstore.ai

Install

git clone https://github.com/embedstore/embedstore.git
cd embedstore
python setup.py install

APIS

Load an embedding

# load embeddings as a list
from embedstore import load_embedding

embeddings = load_embedding("langchain/docs-python")
print(len(embeddings))
print(embeddings[1])

# load embeddings for ingesting in Chroma db
result = load_embedding("langchain/docs-python", embed_for="chroma")
# result is {"embeddings": [], "documents": [] "ids": []}
print(result["embeddings"])
print(result["documents"])
print(result["ids"])

Get all embeddings info

from embedstore import get_embeddings_info

embeddings = get_embeddings_info()

# output
{"data": [{"name": "The Almanack Of Naval Ravikant", "id": "book/naval-almanack"}, {"name": "Langchain Python Docs", "id": "langchain/docs-python"}]}

Get all embedding info

from embedstore import search_embedding_info

embeddings = search_embedding_info("doc")

# output
{"data": [{"name": "Langchain Python Docs", "id": "langchain/docs-python"}]}