Document loaders

Prerequisites

• Document API Reference

Overview

Document loaders are designed to load document objects. LangChain has hundreds of integrations with various data sources to load data from: Slack, Notion, Google Drive, etc.

Available Integrations

You can find available integrations on the Document Loaders Integrations page.

Interface

Documents loaders implement the BaseLoader interface.

Each DocumentLoader has its own specific parameters, but they can all be invoked in the same way with the .load method or .lazy_load.

Here's a simple example:

from langchain_community.document_loaders.csv_loader import CSVLoader

loader = CSVLoader(
    ...  # <-- Integration specific parameters here
)
data = loader.load()

API Reference:CSVLoader

or if working with large datasets, you can use the .lazy_load method:

for document in loader.lazy_load():
    print(document)

Related Resources

Please see the following resources for more information:

• How-to guides for document loaders
• Document API Reference
• Document Loaders Integrations