Skip to content
/ kitops Public

An open source DevOps tool for packaging and versioning AI/ML models, datasets, code, and configuration into an OCI artifact.

License

Notifications You must be signed in to change notification settings

jozu-ai/kitops

Repository files navigation

KitOps

Description

The KitOps project provides tools for building, packaging, and deploying / running all kinds of ML models. The KitOps project enables data scientists and AI/ML developers to encapsulate their models along with all necessary dependencies, configurations, and environments into a standardized, portable format. This facilitates easy sharing and deployment across diverse computing environments, ensuring that models are readily usable without compatibility issues.

The project is composed of two main components:

  • A kitfile manifest that lists a model, its dataset(s), code, and metadata - everything required to build, deploy, or run that model
  • The kit command-line interface (CLI) tool designed to simplify the packaging, versioning, and distribution of AI/ML models as OCI (Open Container Initiative) artifacts

We refer to the final package that the kit CLI builds based on the kitfile as a Model Kit. Model Kits can be run in a Jupyter Notebook, or packaged as a Docker container that can be run anywhere. Kit supports a wide range of AI/ML frameworks, offers robust version control for model iterations, integrates security practices for safe distribution, and provides a seamless connection to model repositories for storing and retrieving models.

Kit's aim is to streamline the end-to-end lifecycle of AI/ML model management, making it as effortless as managing containerized applications.

Using Kit

Running Kit with Binaries

You can download the Kit CLI using one of our tagged versions. The latest tag is used for the latest tested and stable release. This is usually the best place to start. If you want to live on the cutting edge then next is the tag we use to for development builds.

Make sure you get the right binary for your platform:

  • MacOS: TODO
  • Linux: TODO
  • Windows: TODO

We suggest renaming the executable once it's downloaded to just kit then make sure it's in your path and executable.

Run Kit by opening a terminal and typing:

./kit

This will list all the commands you can use.

Building and Running Kit from Source Code

You can get the Kit CLI sources from our tagged versions. The latest tag is used for the latest tested and stable release. This is usually the best place to start. If you want to live on the cutting edge then next is the tag we use to for development builds.

go build -o kit

Then run the project:

./kit

Alternatively

go run kit

Filing Issues and Feature Requests

We want to see Kit become an open standard across the growing AI/ML industry so we deeply value the issues and feature requests we get from users in our community 💖. You can file an issue by selecting the Issues tab and hitting the New Issue green button. There are templates for Issues and Feature Requests, the more information you provide us the faster and better we can address your request. Thank you!

Contributing

We ❤️ our Kit community and contributors. To learn more about the many ways you can contribute (you don't need to be a coder) and how to get started see our Contributor's Guide. Please read our Governance and our Code of Conduct before contributing.

Support

If you need help there are several ways to reach our community and Maintainers outlined in our support doc

Code of Conduct

We run an inclusive, empathetic, and responsible community. Please read our Code of Conduct.