CI/CD for Machine Learning

In this chapter, you will explore the essential principles of Continuous Integration/Continuous Delivery (CI/CD) and YAML. You'll grasp the software development life cycle and key terms like build, test, and deploy. Discover the differences between Continuous Integration, Continuous Delivery, and Continuous Deployment. Moreover, you'll investigate the significance of CI/CD in machine learning and experimentation.

Get ready to explore GitHub Actions (GHA), an influential platform for executing CI/CD workflows. Uncover the diverse components of GHA, encompassing events, actions, jobs, steps, runners, and context. Gain insights into crafting workflows that activate upon events like push and pull requests, and tailor runner machines. Dive into hands-on learning as you establish fundamental CI pipelines and grasp the intricacies of the GHA log.

In this chapter, you'll explore the integration of machine learning model training into a GitHub Action pipeline using Continuous Machine Learning GitHub Action. You'll generate a comprehensive markdown report including model metrics and plots. You will also delve into data versioning in Machine Learning by adopting Data Version Control (DVC) to track data changes. The chapter also covers setting DVC remotes and dataset transfers. Finally, you'll explore DVC pipelines, configuring a DVC YAML file to orchestrate reproducible model training.

In this chapter, you will direct your attention towards the analysis of model performance and the fine-tuning of hyperparameters. You will acquire practical expertise in comparing metrics and visualizations across different branches to assess changes in model performance. You will conduct hyperparameter tuning using scikit-learn's GridSearchCV. Furthermore, you will delve into the automation of pull requests using the optimal model configuration.