Getting error while scoring a ML model "ERROR: Met error <class 'Exception'>:No valid deployments to route to" on AKS via Azure DevOps Pipeline

Question

Hi,

I'm getting below error while scoring a ML model on AKS via Azure DevOps Pipeline.

Error:

ERROR: Met error <class 'Exception'>:No valid deployments to route to. Please check that the endpoint has at least one deployment with positive weight values or use a deployment specific header to route. [https://docs.microsoft.com/en-us/azure/machine-learning/how-to-safely-rollout-managed-endpoints#deploy-a-new-model-but-send-it-no-traffic-yet]

My ML Workspace, AKS its under Private network (Vnet & Private endpoint enabled ) with firewall.

Answer 1

@siddharth joshi Thanks for the question, Currently In Azure Machine Learning API/SDK/CLI v2 the AKS is not supported and will be supported in the near future.

Here are the samples for Batch endpoint using the Azure ML Compute cluster.