Skip to main content
Creates a CyborgVectorStore instance from a list of LangChain Document objects. 
@classmethod
from_documents(
    documents: List[Document],
    embedding: Union[str, Embeddings, SentenceTransformer],
    **kwargs
) -> CyborgVectorStore

​
Parameters

ParameterTypeDescription
documentsList[Document]List of LangChain Document objects
embeddingUnion[str, Embeddings, SentenceTransformer]Embedding model or model name
**kwargsAnyAdditional arguments passed to from_texts

​
Returns

CyborgVectorStore: Initialized vector store with documents added

​
Example Usage

from langchain_core.documents import Document
from cyborgdb_core import DBConfig

# Create documents
documents = [
    Document(
        page_content="Introduction to natural language processing",
        metadata={"source": "nlp_guide.pdf", "page": 1}
    ),
    Document(
        page_content="Tokenization and text preprocessing",
        metadata={"source": "nlp_guide.pdf", "page": 15}
    ),
    Document(
        page_content="Word embeddings and semantic similarity",
        metadata={"source": "nlp_guide.pdf", "page": 42}
    )
]

# Create store from documents
store = CyborgVectorStore.from_documents(
    documents=documents,
    embedding="sentence-transformers/all-mpnet-base-v2",
    index_name="nlp_documents",
    index_key=CyborgVectorStore.generate_key(),
    api_key="your-api-key",
    index_location=DBConfig("memory"),
    config_location=DBConfig("memory")
)