DBND is an open-source framework for building and tracking data pipelines. It is used for processes ranging from data ingestion, preparation, machine learning model training, and production.
DBND includes a Python library, a set of APIs, and a CLI that enables you to collect metadata from your workflows, create a system of record for runs, and easily orchestrate complex processes.
DBND simplifies the process of building and running data pipelines:
from dbnd import task
@task
def say_hello(name: str = "databand.ai") -> str:
value = "Hello %s!" % name
return valueAdditionally, DBND makes it easy to track your critical pipeline metadata:
from dbnd import log_metric, log_dataframe
log_dataframe("my_dataset", my_dataset)
log_metric("r2", r2)This project is built using Python. To ensure compatibility, use the recommended version of Python supported by DBND. Make sure you have the following installed before proceeding:
- Virtualenv or Conda
To install DBND, run:
pip install dbndTo verify the successful installation, import the library and run a simple task:
from dbnd import task
@task
def verify_installation() -> str:
return "Installation successful!"See our documentation with examples and quickstart guides to get up and running with DBND.
If you have any questions, feedback, or contributions, feel free to get in touch with us at [email protected], and open pull request and issues in this repository.
Refer to additional documents for further guidance:
To stay updated with the latest features and improvements, ensure you update DBND regularly via pip install --upgrade dbnd.