Working with AI in the Data Lake

This section describes how to work with AI in the Data Lake.

Use the DEFAIR documentation to learn how to prepare AI-ready Earth observation data.

Contents:

• DEFAIR
  • Getting Started
  • User Guide
  • Tutorials
  • Plugin Development
  • Reference
  • What to do next
• MLOps Studio
  • Access the MLOps Studio workspace
  • Create and manage a MLOps Studio namespace
  • Download and use kubeconfig for a MLOps Studio namespace
  • Use the MLOps Studio dashboard
  • Notebook Servers
  • Access a MLOps Studio notebook environment with Visual Studio Code Remote Tunnels
  • Access a MLOps Studio notebook environment with kubectl port-forward
  • Choose how to run Python workloads in MLOps Studio
  • Run Python code in a MLOps Studio notebook environment
  • Run Python code as a MLOps Studio PyTorchJob
  • Access MLOps Studio Pipelines API from outside the cluster
  • MLOps Studio Pipelines: Accessing Model Registry via Istio
  • Use DEDL standalone S3 object storage with KServe
  • Choosing Gloo or NCCL for Distributed PyTorch Training in PyTorchJob
  • Known limitations in MLOps Studio
  • MLOps Studio Scenarios