I am using Automated ML timeseries for predicting sales from sample sales data of a retailer.
My historical data is available in Azure blob storage and I linked to the Azure Machine Learning Studio instance through datastore. I am creating dataset under new automated ML run
My historical sales data has details of transaction data at the store level and product level . it has the 2 time series identifier.
In the snap2 , we can observe that count is 10000 and time is 2019-02-03, even though original CSV file has data till 2020-08-31
I observer similar issue with other dataset. Also while generating the model it either fails with error "Unable to detect the frequency " or after generating model I am unable to test model for a valid product and store identifier
If I trim the data to less than10000 , I am able to create the model and run the model
It will be highly appreciated if there are work around to this problem
Hi Romungi
Thanks for the response
I am using Automated ML timeseries for predicting sales from sample sales data of a retailer.
My historical data is available in Azure blob storage and I linked to the Azure Machine Learning Studio instance through datastore. I am creating dataset under new automated ML run
My historical sales data has details of transaction data at the store level and product level . it has the 2 time series identifier.
In the snap2 , we can observe that count is 10000 and time is 2019-02-03, even though original CSV file has data till 2020-08-31
I observer similar issue with other dataset. Also while generating the model it either fails with error "Unable to detect the frequency " or after generating model I am unable to test model for a valid product and store identifier
If I trim the data to less than10000 , I am able to create the model and run the model
It will be highly appreciated if there are work around to this problem
Thanks

Ramabadran