I’m trying to automate our factory and as part of it I would like to develop some predictive models using DataIKU, but I want to deploy these models on edge instead of a an API node. Can you please advise if that is possible and if so, how do we do that?
Can you please share a how to document or a architecture that depicts the same.
My company is a heavy user of DataIKU for our enterprise needs, but this specific use is for a conceptual factory.
Thanks for your time
For model deployment at the edge, Dataiku has several options: API node, Java Model Export (POJO), PMML model export.
The best option would be to contact your CSM to discuss it further in the context of your use case.