ML for forecasting in Dataflow

I have a dataflow with cases, where the number of cases would be a countrows of this table. 

I'm wondering if I could use an ML model on this dataflow to forecast the number of cases (rows) per day on a given period. 

When I try to set up the model, I select table on the dataflow, but in "Outcome column", the model is obviously expecting a column, but the result that I want to forecast is a measure (a countrows of the table). 

Is is possible to use the auto ML for such a scenario? Would it work if I create a numeric column with a 1 on each row?

Thank you in advance