Iterative launches a machine learning management tool
Iterative, the MLOps company dedicated to streamlining the workflow of data scientists and machine learning (ML) engineers, has launched Machine Learning Engineering Management (MLEM) – a tool for deploying and open-source template registry that uses an organization’s existing Git infrastructure and workflows.
According to the company, MLEM is designed to bridge the gap between ML engineers and DevOps teams. DevOps teams can understand the underlying frameworks and libraries a model uses and automate deployment in a one-step process for production services and applications, iterative states.
Sriram Subramanian, Research Director of IDC AI/ML Lifecycle Management Softwrae, says, “Iterative allows customers to treat AI models like another type of software artifact. The ability to create ML model registries using the Git framework and DevOps principles allows models to go into production faster. »
MLEM is a building block for a Git-based ML model registry, as well as other iterative tools, such as GTO and DVC. A model registry stores and versions trained ML models. Model registries greatly simplify the task of tracking models throughout their ML lifecycle, from training to production deployments and ultimately retirement.
Dmitry Petrov, co-founder and CEO of Iterative, says, “Model registries make it easier to track models moving through the ML lifecycle by storing and managing versions of trained models, but organizations who create these ledgers end up with two different technology stacks for machine learning models and software development.
“MLEM as the building block for model registries uses Git and traditional CI/CD tools, aligning ML and software teams so they can get models into production faster.”
With the iterative tools, organizations can create a registry of ML models based on software development tools and best practices, the company says. This means that Git acts as a central source of truth for models, eliminating the need for machine learning-specific external tools.
All of the information around a model, including what’s in production, development, or deprecated, can all be viewed in Git. The modular nature of MLEM fits into any organization’s Git and CI/CD-based software development workflows, without engineers having to switch to a separate machine learning deployment and registry tool. This allows teams to use a similar process for ML models and deployment applications, eliminating process and code duplication.
Teams can then create a model register in hours rather than days. MLEM promotes a complete machine learning model lifecycle management workflow using a GitOps-based approach. Software development and MLOps teams can then be aligned, using the same tools to accelerate the time it takes for a model to move from development to production.
Iterative was founded in 2018. Since then, its tools have logged over 8 million sessions and are growing rapidly, with over 12,000 stars on GitHub between CML and DVC. DVC users grew by almost 95% in 2021 with over 3000 monthly users. Iterative now has more than 300 contributors to the various tools.