I am interested in integrating a SageMaker model into QuickSight. My plan is to train the model on SageMaker and use it to make predictions. I have previously used the built-in ML Insights feature from QuickSight for forecasting, which worked well. However, I need to perform more in-depth analysis based on predictions, and the current feature does not provide an option to store the data.
I also want to apply “what-if” analysis, similar to the functionality available within ML Insights, to my deployed model. For example, I would like to have two predicted columns: the first one with predictions based on the data, and the second one with predictions based on a “what-if” scenario, such as setting a target label to achieve a specific value after a certain number of days.
Is there a way to integrate this functionality into a SageMaker model within QuickSight?
Hello @duncan Thanks for your help. I can import the sagemaker model to quicksight it is easy. But my problem is to set conditions like e.g I want the predicted value to be 50000 after 20 days (The model should make predictions based on target value and target days). This can be done in ml insights by setting what if analysis scenario but we cant store the data using quicksight(I want the data to be stored because I will use these predicted values to calculate other variables like how many editors will be required per day etc). This is the reason I moved to sagemaker but in sagemaker I dont see any what if analysis scenario. Do you know any thing about it?
Hi @zainab,
It’s been awhile since we last heard from you. Did you have any additional questions regarding your topic or were you able to find a work around?
If we do not hear back within the next 3 business days, I’ll go ahead and close out this topic.
In SageMaker Canvas Studio once model is trained & available we can do Single Prediction option which allows users to change the input values to predict the target value . You may try this option and see if it meets the requirement. Otherwise would recommend to reach out your AWS Solution Architect to do deep dive discussion with AI & ML Solution Architects.
Thanks, I dont have issue with integration. I have problem with what if analysis where it is asking me to provide dataset that will contain future values which is not possible. I need to make predictions based on what if analysis
It has been some time since we have heard from you so I am going to archive this topic. If you still need help please feel free to repost your question at the top of the community.
I would also recommend creating a support ticket if you believe you are encountering an error with SageMaker Canvas. Here are the steps to open a support case. If your company has someone who manages your AWS account, you might not have direct access to AWS Support and will need to raise an internal ticket to your IT team or whomever manages your AWS account. They should be able to open an AWS Support case on your behalf.
