Document loaders
Prerequisites
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.
Integrationsβ
You can find available integrations on the Document loaders integrations page.
Interfaceβ
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
When working with large datasets, you can use the .lazy_load
method:
for document in loader.lazy_load():
print(document)
Related resourcesβ
Please see the following resources for more information: