This IBM Redpaper® publication introduces and explains AI technologies on IBM Z and demonstrates the capabilities that are used in a use case that involves credit risk scoring on the platform.
It also describes and demonstrates the implementation and integration of an end-to-end solution, from developing and training a deep learning model, to deploying that model in an IBM z/OS® V2R5 environment on an IBM z15™ hardware, to integrating AI functions into an IBM z/OS CICS® application.
The benefits that are derived from this solution also are discussed, which includes how the open-source AI framework portability of the platform enables model development and training to be done anywhere, including Z, and the ability to easily integrate to deploy on IBM Z for optimal inferencing. You can then uncover insights at the transaction level, while taking advantage of the speed and depth of the platform.
This publication is intended for technical specialists, site reliability engineers, architects, system programmers, and system engineers. Technologies that are covered include, but are not limited to, TensorFlow Serving, IBM Watson Machine Learning for z/OS, IBM Cloud® Pak for Data, IBM z/OS CICS, Open Neural Network Exchange, and IBM Deep Learning Compiler.
Chapter 1. Foundations of Artificial Intelligence
Chapter 2. Methodology
Chapter 3. Real-time in-transaction scoring use case
Chapter 4. Other use case scenarios
Chapter 5. Key takeaways
Appendix A. Installation and configuration pointers
Appendix B. Additional material