Closed scottrhay closed 2 months ago
I have the same issue. Have you figured out any resolution?
@scottrhay, @jackmmcgarry, let me work on this and find a solution, thanks for sharing.
Hi @sherzyang, I have the same issue. Have you found a solution?
Hi @scottrhay, @jackmmcgarry, @aparnakhadatkar, we have a solution. I will be updating the lab instructions later today. Here's a summary of the solution:
At the end we see that we're using v1 which is not compatable with testing. To "use" v2 we need to change the type of input dataset. A tabular dataset is a v1 type of dataset. We need a MLTable dataset to "force" the UI to use the v2 SDK in the backend.
Hi @scottrhay, @jackmmcgarry, @aparnakhadatkar, unfortunately, the Studio does not support creating the MLTable dataset at this time. I am going to close this bug and put a note in the lab for now -- the note will say that testing is not possible at this time, as use of the v2 in the Studio is not fully implementable. Long-term, I will find an alternate solution.
im facing the same issue what should i do then?
Okay, it sounds like we just need to skip the testing part of the lab.
Can you clarify what exactly v1 and v2 are referred to by this error message?
so should i just delete the workspace
Same issues I've been having here as well
Module: 01
Lab/Demo: Explore Automated Machine Learning in Azure Machine Learning
Task: Test the Deployed Service
Step: 2
Description of issue
Cannot test the model, Azure returns an error message that the model uses an outdated API.
Repro steps:
Follow the lab, when following the Test the deployed service section, I get the following error message when opening the Test tab in Azure: V1 deployment testing not supported This deployment is based on v1 API and doesn't support testing on the Studio. To get the key/token and invoke, please use CLI/SDK/REST v1 API. Consider migrating to v2 managed online endpoint.