Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/project-codeflare/codeflare

Simplifying the definition and execution, scaling and deployment of pipelines on the cloud.
https://github.com/project-codeflare/codeflare

automl data-science hyperparameter-optimization machine-learning pipelines ray sklearn workflows

Last synced: 29 days ago
JSON representation

Simplifying the definition and execution, scaling and deployment of pipelines on the cloud.

Awesome Lists containing this project

README 

        










[![License](https://img.shields.io/badge/license-Apache--2.0-blue.svg)](http://www.apache.org/licenses/LICENSE-2.0)

[![Build

Status](https://travis-ci.com/project-codeflare/codeflare.svg?branch=main)](https://travis-ci.com/project-codeflare/codeflare.svg?branch=main) 

[![GitHub](https://img.shields.io/badge/issue_tracking-github-blue.svg)](https://github.com/project-codeflare/codeflare/issues)

[![GitHub](https://img.shields.io/badge/CodeFlare-Join%20Slack-blue)](https://invite.playplay.io/invite?team_id=T04KQQBTDN3)



# Simplified and efficient AI/ML on the hybrid cloud



CodeFlare provides a simple, user-friendly abstraction for developing, scaling, and managing resources for distributed AI/ML on the Hybrid Cloud platform with OpenShift Container Platform.



---



## 📦 Stack Components and Features



CodeFlare stack consists of the following main components. This project is organized as a metarepo, gathering pointers and artifacts to deploy and use the stack.



* **Simplified user experience**:

CodeFlare [SDK](https://github.com/project-codeflare/codeflare-sdk) and [CLI](https://github.com/project-codeflare/codeflare-cli) to define, develop, and control remote distributed compute jobs and infrastructure from either a python-based environment or command-line interface



* **Efficient resource management**:

Multi-Cluster Application Dispatcher [(MCAD)](https://github.com/project-codeflare/multi-cluster-app-dispatcher) for queueing, resource quotas, and management of batch jobs. And [Instascale](https://github.com/project-codeflare/instascale) for on-demand resource scaling of an OpenShift cluster



* **Automated and streamlined deployment**:

[CodeFlare Operator](https://github.com/project-codeflare/codeflare-operator) for automating deployment and configuration of the Project CodeFlare stack



With CodeFlare stack, users automate and simplify the execution and scaling of the steps in the life cycle of model development, from data pre-processing, distributed model training, model adaptation and validation.



Through transparent integration with [Ray](https://github.com/ray-project/ray) and [PyTorch](https://github.com/pytorch/pytorch) frameworks, and the rich library ecosystem that run on them, CodeFlare enables data scientists to **spend more time on model development and minimum time on resource deployment and scaling**. 



See below our stack and how to get started.



--- 

## ⚙️ Project CodeFlare Ecosystem



In addition to running standalone, Project CodeFlare is deployed as part of and integrated with the [Open Data Hub](https://github.com/opendatahub-io/distributed-workloads), leveraging [OpenShift Container Platform](https://www.openshift.com). 



With OpenShift, CodeFlare can be deployed anywhere, from on-prem to cloud, and integrate easily with other cloud-native ecosystems.










---



## 🛠️ Getting Started



### Learning



Watch [this video](https://www.youtube.com/watch?v=OAzFBFL5B0k) for an introduction to Project CodeFlare and what the

stack can do.



### Quick Start



To get started using the Project CodeFlare stack, try this [end-to-end example](https://github.com/opendatahub-io/distributed-workloads/blob/main/Quick-Start.md)!



For more basic walk-throughs and in-depth tutorials, see our [demo notebooks](https://github.com/project-codeflare/codeflare-sdk/tree/main/demo-notebooks/guided-demos)!



## Development



See more details in any of the component repos linked above, or get started by taking a look at the [project board](https://github.com/orgs/project-codeflare/projects/8) for open tasks/issues!



### Architecture



We attempt to document all architectural decisions in our [ADR documents](https://github.com/project-codeflare/adr). Start here to understand the architectural details of Project CodeFlare.



---



## 🎉 Getting Involved and Contributing



Join our [Slack community][slack] to get involved or ask questions.



## Blog



CodeFlare related blogs are published on our [Medium publication](https://medium.com/codeflare).



## License



CodeFlare is an open-source project with an [Apache 2.0 license](LICENSE).



[codeflare-sdk]: https://github.com/project-codeflare/codeflare-sdk

[codeflare-cli]: https://github.com/project-codeflare/codeflare-cli

[mcad]: https://github.com/project-codeflare/multi-cluster-app-dispatcher

[instascale]: https://github.com/project-codeflare/instascale

[codeflare-operator]: https://github.com/project-codeflare/codeflare-operator

[distributed-workloads]: https://github.com/opendatahub-io/distributed-workloads

[quickstart]: https://github.com/opendatahub-io/distributed-workloads/blob/main/Quick-Start.md

[slack]: https://invite.playplay.io/invite?team_id=T04KQQBTDN3

[adr]: https://github.com/project-codeflare/adr

[demos]: https://github.com/project-codeflare/codeflare-sdk/tree/main/demo-notebooks/guided-demos

[board]: https://github.com/orgs/project-codeflare/projects/8

[youtube-demo]: https://www.youtube.com/watch?v=OAzFBFL5B0k