Continuous Machine Learning: CML for CI/CD Need to introduce C | Big Data Science

Continuous Machine Learning: CML for CI/CD
Need to introduce CI / CD in the development of ML systems? Try CML, an open source CLI tool from Iterative.ai for implementing CI/CD within MLOps. It is suitable for automating ML model development workflows, including provisioning, training and evaluation, comparison of experiments in the history of the project, and monitoring of changing datasets. CML is based on the following principles:
• GitLab or GitHub for managing ML experiments, monitoring model training and data changes using DVC;
• Automated reports for machine learning experiments with metrics and graphs on every Git pull to make informed decisions based on data.
• no additional services - only GitLab, Bitbucket or GitHub, Docker and DVC. Optionally, you can add cloud storage, as well as self-hosted or cloud workers such as AWS EC2 or MS Azure.
CML introduces CI/CD-style automation into the workflow: most of the configurations are defined in the cml.yaml file stored in the repository. This file specifies what actions should be taken when a new feature branch is ready to be merged into the main branch. When a pull request is created, GitHub Actions uses this workflow and performs the actions specified in the configuration file.
Source code: https://github.com/iterative/cml
Documentation: https://cml.dev/doc
Use case example: https://towardsdatascience.com/continuous-machine-learning-e1ffb847b8da