langchain/templates/neo4j-vector-memory/ingest.py

24 lines
691 B
Python
Raw Normal View History

from pathlib import Path
from langchain.text_splitter import TokenTextSplitter
from langchain_community.document_loaders import TextLoader
from langchain_community.embeddings.openai import OpenAIEmbeddings
from langchain_community.vectorstores import Neo4jVector
txt_path = Path(__file__).parent / "dune.txt"
# Load the text file
loader = TextLoader(str(txt_path))
raw_documents = loader.load()
# Define chunking strategy
splitter = TokenTextSplitter(chunk_size=512, chunk_overlap=24)
documents = splitter.split_documents(raw_documents)
# Calculate embedding values and store them in the graph
Neo4jVector.from_documents(
documents,
OpenAIEmbeddings(),
index_name="dune",
)