Skip to content
Products
Platform
Developer Tools
Portal

Setup the local environment for AI Model Experiments

Last updated on

To connect your local environment to the STACKIT service, you need to configure the following environment variables. These allow the MLflow™ client to authenticate and communicate with both the tracking server and the artifact storage.

  • MLFLOW_TRACKING_URI: The tracking URI of your AI Model Experiments instance. It can be retrieved directly from the instance details page in the STACKIT Portal as shown in Manage Instances.
  • MLFLOW_TRACKING_TOKEN: The authentication token generated for your instance. To create a tracking token please follow the Manage Tokens guide.
  • MLFLOW_S3_ENDPOINT_URL: The endpoint for STACKIT Object Storage. This has to be set to https://object.storage.eu01.onstackit.cloud.
  • AWS_ACCESS_KEY_ID: Your S3 access key ID for artifact storage.
  • AWS_SECRET_ACCESS_KEY: Your S3 secret access key for artifact storage.

S3 Access and Secret keys are generated within the Bucket Settings in the STACKIT Portal. For specific instructions, refer to the Manage Artifacts documentation.