Charmed MLFlow is part of Canonical's growing MLOps portfolio. This newly released platform is ideal for model registry and experiment tracking and can be integrated with other AI and big data tools, such as Apache Spark and Kubeflow.
Charmed MLFlow can be deployed within minutes (even on off-the-shelf hardware such as laptops and desktops) to help facilitate fast experimentation. Although Charmed MLFlow has been fully tested on Ubuntu, it can be used on other operating systems, via Canonical's Multipass or with Windows Subsystem for Linux (WSL).
According to Cédric Gégout, VP of product management at Canonical, "MLFlow has become the leading AI framework for streamlining all ML stages. Its popularity arises from its flexibility in facilitating modest local desktop experimentation and extensive cloud deployment, catering to both individual and enterprise needs." Gégout added, "This made Charmed MLFlow a fitting addition to our Canonical MLOps suite, offering cost-effective solutions that enable developers to start small and scale up as their business grows, without the typical ML infrastructure hassle and with a simple Ubuntu Pro subscription."
Charmed MLFlow can also run on just about any environment, including public, private, and hybrid clouds and CNCF-conformant Kubernetes distributions, such as Microk8s. In addition, data scientists can migrate their models from laptops to whatever infrastructure they choose using the same tooling (which allows for seamless migration between clouds).
With Charmed MLFlow you can work with automated lifecycle management and integrations, generative AI, and much more. This new technology also benefits from security patching via Canonical's Ubuntu Pro subscription, which means it will receive timely patches for CVEs (Common Vulnerabilities and Exposures), as well as hardening and compliance with standards such as FedRAMP, HIPAA, and PCI-DSS.