How to infer the online feature store table via an mlflow registered model, which is deployed to a sagemaker endpoint?

[Maverick1]

Can an mlflow registered model automatically infer the online feature store table, if that model is trained and logged via a databricks feature store table and the table is pushed to an online feature store (like AWS RDS)?

[Kaniz]

Hi @Saurabh Verma​ ,

You may find something similar to what you're looking for here?

Please let me know if this helps.

[Maverick1]

@Kaniz Fatma​ : Not working for Sagemaker and AWS RDS or Dynamo DB component.

[Kaniz]

Hi @Saurabh Verma​ , Would you like to raise a feature request or submit an idea ?

[Maverick1]

@Kaniz Fatma​ : I had been told recently that this feature is actually going to be released with the new version of DB runtime. Could you please check which DBR will come with this feature ? and what is the timeline.?

[Kaniz]

Hi @Saurabh Verma​ , I'll get back to you with your requested information. Thank you.

[Kaniz]

Hi @Saurabh Verma​ , We're looking for an answer for you. Please bear with me for the delay. Apologies.

[Christian]

Thanks for sharing [this](http://www.google.com) article

[Atanu]

@Saurabh Verma​  let us know if you need further help on this! Thanks.

[Maverick1]

@Atanu Sarkar​ : Your colleague @Kaniz Fatma​  is looking for an answer. Maybe you can also help her.

[Kaniz · Accepted solution]

Hi @Saurabh Verma​ , Feature Store <> SageMaker integration is not fully rolled out yet. We are looking to roll that out in Private Preview mode soon. It will need DynamoDB online store type which will be available soon.