Sagemaker Canvas time series forecasting model build failed
Dear all, I am trying to build a timeseries forecasting model using Sagemaker Canvas but receive the following error :
Your model build failed because of an issue with the Canvas server. Failure reason: <The value of "[]" should be a list of strings.>.Contact your administrator and share the details below to resolve the issue.
The dataset only consist of time_stamp, item_id and demand. I notice that the data is a bit sparse not every SKU have rows for every month. Anybody know if this might be the cause of the error or there are other possible cause ?
- Topics
- Machine Learning & AI
- Language
- English
asked 2 years ago513 views
- Newest
- Most votes
- Most comments
Hi,
It seems to me that Canvas is complaining because some fields in your rows are not strings (i.e. not '123') but probably numbers (123).
Can you check that and sanitize the data if needed ?
Best,
Didier
Relevant content
asked 4 years ago
asked 2 years ago
asked 3 years ago
- AWS OFFICIALUpdated a year ago
- AWS OFFICIALUpdated 2 years ago
There are only three columns : time_stamps, item_id which is strings, and demand which is number. All the data has been check and there are no numbers in the item_id field.