Skip to main content

We understand one option is to push the result data from Celonis ML workbench back to the Celonis data pool and display in the UX. We are wondering is there a way for ML workbench embedded into Celonis UX in a dynamic manner: when a user changes filters and date range, the information can be passed to ML workbench, ML workbench computes the result and plot a graph and Celonis UX retrieves the graph and displays in Celonis UX.

 

Celonis Studio allows adding a image as a component, however my understanding is that the image is a static component as to be generated before hand and stored. We are hoping to add a image that's dynamically generated in ML workbench.

Xiao Huang,

Apologies for the late reply, I've also just sent this to you through a separate email. The support engineer sent the response to your question as an internal note by accident. I wanted to make sure that you received the proper information.

You were asking about a way for ML workbench to update the UX in a dynamic manner, ie have changes to filters update the Celonis UX through ML workbench.

The only way we see this working would be by adding the ML component to your view/analysis which allows you to trigger the ML notebook. Within the ML notebook script, you would need to reload the data model or specific table. Once the data is updated, then the components using the data should be updated with the correct data. But I do not have an example of any other customers doing this.

Please let me know if you have any additional questions.

Reply