Installation

Local / Notebook / Script (Python 3.10+)

From PyPI:

pip install cognite-data-quality

In a notebook:

%pip install cognite-data-quality

After install, restart the kernel or interpreter so imports pick up the package. SHACL validation code is inlined in the package — no external dependencies required.

Embedded function code

Function handler code is embedded in the package. There is no separate function/ directory or function_source_root parameter. When you deploy with deploy_validation_infrastructure(), the function zip is built from cognite_data_quality._function_code automatically. You do not need to clone or copy handler code elsewhere.

To confirm embedded files are present:

from cognite_data_quality.deploy import _get_embedded_function_files

embedded = _get_embedded_function_files()
print(f"Embedded function files: {len(embedded)}")

Cognite Function (deployed in CDF)

When the validation runs inside a Cognite Function, dependencies are declared in the function’s requirements.txt (folder deploy) or via [requirements] tags in the handler (function_handle deploy). The deployment process that uses deploy_validation_infrastructure() builds the function bundle from the package and includes the correct dependencies.

If you deploy a custom function that uses this package, list it in requirements:

cognite-data-quality

See Deploy for full deployment examples and Runtime options for usage contexts.