sat_ade/task-6a02e23da6fe2e4ac16acf65-v1
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update rag_tools

Browse Source
This commit is contained in:
Аделина Саттарова
2026-05-28 13:25:58 +00:00
parent 1a1c746db5
commit cecde4145a
1 changed files with 4 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
rag_tools.py
+4 -1
View File
@@ -3,6 +3,7 @@ from qdrant_client import QdrantClient
from langchain.embeddings.ollama import OllamaEmbeddings from langchain.embeddings.ollama import OllamaEmbeddings
from langchain.vectorstores.qdrant import QdrantVectorStore from langchain.vectorstores.qdrant import QdrantVectorStore
from langchain.text_splitter import RecursiveCharacterTextSplitter from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.schema import Document
# Initialize embeddings and vector store # Initialize embeddings and vector store
embeddings = OllamaEmbeddings(model="nomic-embed-text") embeddings = OllamaEmbeddings(model="nomic-embed-text")
@@ -10,7 +11,9 @@ client = QdrantClient(host="localhost", port=6333)
collection_name = "knowledge_base" collection_name = "knowledge_base"
# Ensure collection exists # Ensure collection exists
if not client.has_collection(collection_name): if not client.has_collection(collection_name):
client.create_collection(name=collection_name, vectors_config={"size": embeddings.embed_query(["test"]).shape[1], "distance": "Cosine"}) # Determine embedding dimension by embedding a dummy text
dim = embeddings.embed_query(["test"])[0].shape[0]
client.create_collection(name=collection_name, vectors_config={"size": dim, "distance": "Cosine"})
vector_store = QdrantVectorStore(client=client, collection_name=collection_name, embedding=embeddings) vector_store = QdrantVectorStore(client=client, collection_name=collection_name, embedding=embeddings)
# Text splitter # Text splitter