Issue - Generating Explainability in Forecasting Tool


Hello AWS Community! I am currently working on a sales forecasting project for a company with over 150 branches across various countries. My predictive variable is 'amount,' and I am utilizing the forecasting tool provided by AWS. The tool seems straightforward and has the potential to significantly simplify our daily operations. In my analysis, I'm using AutoPredictor, I have incorporated several attributes such as exchange rates, the number of operations on previous days, etc. I have successfully generated predictions, selected 'holidays,' and loaded the corresponding CSV for 'related time series.' However, I am encountering an issue with generating explainability. Despite selecting the option to automatically generate explainability when creating the predictor, I am receiving a "CREATE_FAILED" message (always talking about Console). I would appreciate any insights or guidance on resolving this issue. Has anyone else experienced a similar problem or could provide assistance in troubleshooting this? Thank you in advance for your time and assistance.

asked 4 months ago570 views
1 Answer


I understand that you are seeing ‘CREATE_FAILED’ message upon trying to generate explainability when creating the predictor. Please refer to the best practices and restrictions from [1] when working with explainability. The DescribeExplainability[2][3] API can be used that describes the Explainability resource created for more details and complete error message. If Forecast successfully creates a Predictor but the Predictor Explainability job fails, you also can retry creating Predictor Explainability in the console or with the CreateExplainability[4][5] API using CLI/SDK.

Kindly try checking with the above resources. If you face any other issues or require further assistance, please reach out to AWS Support [6] along with the job ARNs and dataset details, and we would be happy to assist you further. Thank you!








answered 4 months ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions