This article should mention the job environment in the example does not support the SDK v2 API

[mstetka-fr]

This article demonstrates using the SDKv2 API to execute a python script for training a tensorflow model as a job.

It should have a disclaimer that the python script that the job executes utilizes SDKv1 APIs. The compute environment used in the example script does not contain SDKv2 packages.

For those newer to using Machine Learning Studio and are unfamiliar with the SDKv1 APIs (because the documentation strongly recommends that new work should use the v2 APIs), this can cause confusion and a need to migrate the scripts to SDKv1. There is documentation for migrating logging from SDKv1 to SDKv2, but not the reverse.

The root problem is that there are no tensorflow compute environments that support SDKv2 APIs
[Enter feedback here]

---

Document Details

⚠ Do not edit this section. It is required for learn.microsoft.com ➟ GitHub issue linking.

• ID: b5b90fbd-19a9-4fe7-032b-765457549f30
• Version Independent ID: a23a02a5-23fd-9935-4031-b943aa468030
• Content: Train and deploy a TensorFlow model (SDK v2) - Azure Machine Learning
• Content Source: articles/machine-learning/how-to-train-tensorflow.md
• Service: machine-learning
• Sub-service: training
• GitHub Login: @balapv
• Microsoft Alias: balapv

[TPavanBalaji]

@mstetka-fr
Thanks for your feedback! We will investigate and update as appropriate.

[AjayBathini-MSFT]

@mstetka-fr
Thanks for bringing this to our attention.
I'm going to assign this to the document author so they can take a look at it accordingly.