Study
LangChain · Vector stores
Embedding chunks with a local sentence-transformer, storing them in LangChain's InMemoryVectorStore, similarity search with scores, metadata filters, and turning the store into a retriever.
- A vector store keeps (embedding, document) pairs and returns the documents whose embeddings are nearest to the query's.
- as_retriever() turns the store into a Runnable that plugs straight into a chain.