Course Outline
Introduction
Overview of Kubeflow Features and Components
- Containers, manifests, etc.
Overview of a Machine Learning Pipeline
- Training, testing, tuning, deploying, etc.
Deploying Kubeflow to a Kubernetes Cluster
- Preparing the execution environment (training cluster, production cluster, etc.) for government use.
- Downloading, installing, and customizing Kubeflow for government operations.
Running a Machine Learning Pipeline on Kubernetes
- Building a TensorFlow pipeline for government applications.
- Building a PyTorch pipeline for government projects.
Visualizing the Results
- Exporting and visualizing pipeline metrics to support decision-making for government initiatives.
Customizing the Execution Environment
- Customizing the stack for diverse infrastructures in government settings.
- Upgrading a Kubeflow deployment to meet evolving government requirements.
Running Kubeflow on Public Clouds
- AWS, Microsoft Azure, Google Cloud Platform, and other platforms suitable for government use.
Managing Production Workflows
- Running with GitOps methodology to ensure robust governance for government projects.
- Scheduling jobs to optimize resource utilization for government operations.
- Spawning Jupyter notebooks to facilitate collaborative research and development for government initiatives.
Troubleshooting
Summary and Conclusion
Requirements
- Familiarity with Python syntax for government applications
- Experience with TensorFlow, PyTorch, or other machine learning frameworks
- An account with a public cloud provider (optional)
Audience
- Developers
- Data Scientists
Testimonials (1)
I enjoyed participating in the Kubeflow training, which was held remotely. This training allowed me to consolidate my knowledge for AWS services, K8s, all the devOps tools around Kubeflow which are the necessary bases to properly tackle the subject. I wanted to thank Malawski Marcin for his patience and professionalism for training and advice on best practices. Malawski approaches the subject from different angles, different deployment tools Ansible, EKS kubectl, Terraform. Now I am definitely convinced that I am going into the right field of application.
