Systems | Development | Analytics | API | Testing

Technology

Comprehending ClearML and MLOps - Enabling the New A-Z (ODSC East '21)

ClearML is an industry leading MLOps suite, fully open source and free in the best sense. Designed to ease the start, running and management of experiments and orchestration for every day practitioners, we will also see how it provides a clear path to deployment. Starting with a high level overview of the parts built into ClearML, we will then journey into what is and also, importantly, what is not part of ClearML's mandate. Along the way we will demonstrate how-to integrate into your PyTorch code, as well as the capabilities of reporting and possible workflows that could be made easier by pipeline usage.

[MLOps] The Clear SHOW - S02E10 - Everything You Wanted To Know About Model Stores*

Ariel (ft. G. Raffa) discusses the reasoning behind model stores, why you might want to build one, and reviews a model store library vs. ClearML to understand what needs to be built "on top" of our open-source MLOps Engine. + Operator AI ClearML is the only open-source tool to manage all your MLOps in a unified and robust platform providing collaborative experiment management, powerful orchestration, easy-to-build data stores, and one-click model deployment.

Building a Single Pipeline for Data Integration and ML with Azure Synapse Analytics and Iguazio

Across organizations large and small, ML teams are still faced with data silos that slow down or halt innovation. Read on to learn about how enterprises are tackling these challenges, by integrating with any data types to create a single end-to-end pipeline and rapidly run AI/ML with Azure Synapse Analytics with Iguazio.

ETL with Apache Airflow

Written in Python, Apache Airflow is an open-source workflow manager used to develop, schedule, and monitor workflows. Created by Airbnb, Apache Airflow is now being widely adopted by many large companies, including Google and Slack. Being a workflow management framework, Apache Airflow differs from other frameworks in that it does not require exact parent-child relationships. Instead, you only need to define parents between data flows, automatically organizing them into a DAG (directed acyclic graph).