mirror of
https://github.com/hwchase17/langchain
synced 2024-11-20 03:25:56 +00:00
fa5d49f2c1
ran ```bash g grep -l "langchain.vectorstores" | xargs -L 1 sed -i '' "s/langchain\.vectorstores/langchain_community.vectorstores/g" g grep -l "langchain.document_loaders" | xargs -L 1 sed -i '' "s/langchain\.document_loaders/langchain_community.document_loaders/g" g grep -l "langchain.chat_loaders" | xargs -L 1 sed -i '' "s/langchain\.chat_loaders/langchain_community.chat_loaders/g" g grep -l "langchain.document_transformers" | xargs -L 1 sed -i '' "s/langchain\.document_transformers/langchain_community.document_transformers/g" g grep -l "langchain\.graphs" | xargs -L 1 sed -i '' "s/langchain\.graphs/langchain_community.graphs/g" g grep -l "langchain\.memory\.chat_message_histories" | xargs -L 1 sed -i '' "s/langchain\.memory\.chat_message_histories/langchain_community.chat_message_histories/g" gco master libs/langchain/tests/unit_tests/*/test_imports.py gco master libs/langchain/tests/unit_tests/**/test_public_api.py ```
48 lines
1.4 KiB
Python
48 lines
1.4 KiB
Python
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.graphs import Neo4jGraph
|
|
from langchain_community.vectorstores import Neo4jVector
|
|
|
|
txt_path = Path(__file__).parent / "dune.txt"
|
|
|
|
graph = Neo4jGraph()
|
|
|
|
# Load the text file
|
|
loader = TextLoader(str(txt_path))
|
|
documents = loader.load()
|
|
|
|
# Define chunking strategy
|
|
parent_splitter = TokenTextSplitter(chunk_size=512, chunk_overlap=24)
|
|
child_splitter = TokenTextSplitter(chunk_size=100, chunk_overlap=24)
|
|
|
|
# Store parent-child patterns into graph
|
|
parent_documents = parent_splitter.split_documents(documents)
|
|
for parent in parent_documents:
|
|
child_documents = child_splitter.split_documents([parent])
|
|
params = {
|
|
"parent": parent.page_content,
|
|
"children": [c.page_content for c in child_documents],
|
|
}
|
|
graph.query(
|
|
"""
|
|
CREATE (p:Parent {text: $parent})
|
|
WITH p
|
|
UNWIND $children AS child
|
|
CREATE (c:Child {text: child})
|
|
CREATE (c)-[:HAS_PARENT]->(p)
|
|
""",
|
|
params,
|
|
)
|
|
|
|
# Calculate embedding values on the child nodes
|
|
Neo4jVector.from_existing_graph(
|
|
OpenAIEmbeddings(),
|
|
index_name="retrieval",
|
|
node_label="Child",
|
|
text_node_properties=["text"],
|
|
embedding_node_property="embedding",
|
|
)
|