Training: Kubeflow
Level
IntermediateDuration
16h / 2 daysDate
Individually arrangedPrice
Individually arrangedTraining: Kubeflow
The Kubeflow training is an intensive two-day course focusing on the practical application of this platform for managing the machine learning lifecycle on Kubernetes. The program is designed so that 80% of the time is dedicated to hands-on workshops and 20% to theory. Participants will learn how to leverage the full potential of Kubeflow for training, deploying, and monitoring ML models, working on real-world examples and use cases.
What will you learn?
- How to configure and manage Kubeflow on Kubernetes
- How to deploy ML models using Kubeflow Serving and monitor their performance
- How to perform exploratory data analysis (EDA) and train ML models with Kubeflow Pipelines
- How to integrate Kubeflow with other ML tools and cloud platforms, and automate ML processes using CI/CD tools
Prerequisites
- Basic knowledge of Python programming
- Basic skills in working with Kubernetes
- Basic understanding of machine learning
Who is this training for?
Developers and data engineers who want to enhance their skills in managing the ML lifecycle on Kubernetes-
IT specialists looking to use Kubeflow to automate data processing and prediction in their organizations
-
Data scientists and analysts aiming to train and deploy ML models in a scalable production environment
Training Program
-
Day 1: Introduction to Kubeflow and Platform Basics
-
Kubeflow fundamentals
- Introduction to Kubeflow and its architecture
- Installing Kubeflow on Kubernetes
- Data management and exploratory data analysis (EDA)
- Importing and processing data in Kubeflow
- Performing EDA with Kubeflow Pipelines
-
Training models in Kubeflow
- Introduction to training components in Kubeflow
- Automating model training with Kubeflow Pipelines
- Training the first model
- Hands-on exercises: training a model on a real dataset
- Analysis and evaluation of model results
-
Day 2: Advanced Techniques and Practical Applications
-
Advanced techniques for training models
- Using custom scripts for training models
- Leveraging GPUs and compute clusters to accelerate training
-
Deploying and monitoring models
- Deploying models with Kubeflow Serving
- Monitoring and managing deployed models
- Model deployment and optimization
- Hands-on exercises: deploying a Kubeflow model
- Model optimization and hyperparameter tuning
-
Integration with other tools and services (optional)
- Integrating Kubeflow with other ML tools and cloud platforms
- Using CI/CD tools to automate ML processes
