Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Duration 14 hours
Course Outline
Preparing Machine Learning Models for Production
- Encapsulating models using Docker
- Exporting models from TensorFlow and PyTorch
- Considerations for version control and storage
Serving Models on Kubernetes
- Introduction to inference servers
- Implementing TensorFlow Serving and TorchServe
- Configuring model endpoints
Optimizing Inference Performance
- Implementing batching methods
- Managing concurrent request processing
- Tuning for latency and throughput
Autoscaling ML Workloads
- Horizontal Pod Autoscaler (HPA)
- Vertical Pod Autoscaler (VPA)
- Kubernetes Event-Driven Autoscaling (KEDA)
GPU Allocation and Resource Control
- Setting up GPU-enabled nodes
- Overview of the NVIDIA device plugin
- Defining resource requests and limits for ML tasks
Model Release and Deployment Strategies
- Blue/green deployment patterns
- Canary release techniques
- Conducting A/B tests for model validation
Monitoring and Observability for Production ML
- Key metrics for inference operations
- Best practices for logging and tracing
- Creating dashboards and setting up alerts
Security and Reliability in Production
- Hardening model endpoints
- Applying network policies and access controls
- Maintaining high availability
Wrap-Up and Future Directions
Requirements
- A solid grasp of containerized application lifecycles
- Practical experience with Python-based machine learning models
- A working knowledge of core Kubernetes concepts
Target Audience
- ML engineers
- DevOps engineers
- Platform engineering teams
Testimonials (2)
As i said before , for a person like me (no exp. ) this was a gateway to understanding features and functions with these programs/tools & etc. .
Patrick V. Duylovski - UBB + DZI (KBC GROUP)
Course - Docker and Kubernetes
basic understanding of container/kubernetes and how they interact features of the openshift plattform
