rafal1
6 years agoNew Member
Although it is possible to output Azure ML into a table/blob/file and then use that as a data source in Power BI, it would be clearer and easier to integrate the two technologies more directly. I would like to see a way to select an Azure ML experiment or web service in Power BI, especially for forecasting scenarios, as well as recommendations and more run-of-the-mill classifications (like highlight high-value customers, decide which segment the customer belongs too etc). As a bonus, having Power BI selectable as an output in Azure ML would complete the integration.
Recent ideas
Adjust column width in SQL Analytics Endpoint user interface
I can't see the column names and cell values in some of the columns in my tables in SQL Analytics Endpoint because the column width is too narrow. Also, the tooltip also doesn't show the entire...frithjof_v1 day agoCommunity ChampionCompleted4.2KViews75likes8Comments