Forum Discussion
Problem Time Series Forecasting from Azure ML to Power Bi Desktop
Hello, i have created a model from azure machine learning with "Automated ML", its a time series forecasting model. I configured that model to predict from now to 12 months. Then i created a endpoint to connect to my power bi desktop. I tested the endpoint and it worked fine with more than 100 rows (you are gonna understand why i say 100 rows later). (The endpoint parameters are the date and customer_id)
But when i do try to do it in power query with a table from a sql query i got this error.
The weird thing is that it only happens when i bring more than 100 rows from the sql query because when i bring less than 100 rows or exactly 100 rows it works perfect.
Please, help me! :c
3 Replies
Well, maybe there's a limit on how many rows you can bring? You should read the documentation. If you're using something that you don't pay for, then it usually comes with some restrictions.
But if that's true then why i can use more than 100 rows on the endpoint test. Besides, i am paying for it (i am using my organization account)
I forget to say that i can bring more than 100 rows when i bring only the next month of each client. I mean something like this:
