0
$\begingroup$

đź‘‹

Anyone know whether it is possible to use Azure Databricks Feature Store and combine it with training in Azure ML?

Databricks: “Integration with MLflow ensures that the features are stored alongside the ML models, eliminating drift between training and serving time” If possible how can I utilize this in my Azure ML hosted model? I need some connection between those two.

Improve this question
$\endgroup$

1 Answer 1

Reset to default
-1
$\begingroup$

Yes, it is possible.

Next question? (Hopefully one that is well thought out, and accompanied by technical details such as source code.)

Improve this answer
$\endgroup$
2
  • $\begingroup$ I want to use the feature store in Databricks to log the training data set using MLFlow while also logging the runs in Azure ML using MLFlow. This would e.g make it possible to see which features (in Databricks feature store) is used by which models (hosted in Azure ML). That needs a connection between the two. How exactly do you achieve that? It is very limited documentation, and not so straightforward I think. $\endgroup$
    – Kjetil Haukås
    Commented Jan 25 at 23:10
  • $\begingroup$ “Integration with MLflow ensures that the features are stored alongside the ML models, eliminating drift between training and serving time” How can I utilize this in my Azure ML hosted model? $\endgroup$
    – Kjetil Haukås
    Commented Jan 25 at 23:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.