9
ML Deployments and Database Operations
The biggest sources of ongoing maintenance for RPA are due to handling changes to the business logic (Processes) and applications (Objects). With IA, another source of ongoing changes is added into the mix—updating BP to use new ML models as they’re changed. Imagine that models are only updated once per year and that we have five models. That’s still five times per year that we’ll need to modify how our IA solutions are configured due to ML model updates.
Even if the updated ML model keeps the exact same API endpoint definition as before, we still need to understand what ML deployment strategy is being used, as this affects how we plan for rollbacks in case the model is not performing as expected.
We ...
Get Intelligent Automation with Blue Prism now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.