Eliminate unnecessary system complexity in Data Ops and in Machine Learning Ops (MLOps) with this Django module for defining, running and monitoring flows.
Project description
django-flow-forge
This is a module built for Data Scientists and Data Engineers, here to help you forge your Data Ops or ML Ops processes in to simple, flexible and impactful work flows, steering clear of the unnecessary complexities of other packages and vendor lock-in that plague many projects.
Motivation
Read the Docs for more info on this
Features
- Easily define your flows / pipeline processes as a DAG
- Visualize your flows for stakeholders, both in planning and development phases (inspired by Kedro)
- Simple pipeline registration and execution through a script, Django website or via an API by plugging in Django-ninja (Fast API for Django))
- Task monitoring and visualization of outputs and failures.
- Visualise pipeline process outputs and failures for each task (inspired by Kedro)
- Compare data science experiment results easily
- Serve your models by plugging in Django Ninja
- A familiar environment for Django users, minimizing the learning curve
- Leverage Django's robust security
- Leverage Django's existing scalability features, including Django Celery and flexbility with Kubernetes engines, without the worry of vendor lock-in
- Encourages teams to move away from Notebooks, which cause a plethora of issues
Example stakeholder facing development visualisation
Quick Start
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
django_flow_forge-0.2.37.tar.gz
(116.2 kB
view hashes)
Built Distribution
Close
Hashes for django_flow_forge-0.2.37-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | abb39cb695decd860069cb046d3d3959a6429bed24037ebd3615bd02e39368a8 |
|
MD5 | 9eb59bdb55ee0da5634447562099c07b |
|
BLAKE2b-256 | 15233e4a663d3edd1297ebd75922498809058d024b28c302739e2bb1c6e8ed52 |