Jane Man

  • S205 - Write your own database ML scoring function using Spark

    DB2 Sessions: Room C first floor

    Session Evaluation

    Are you interested in getting insight from your enterprise data? Is existing solution too costly and/or too general that cannot satisfy your needs? In this session, we will show you how to do database analytics and write your own ML scoring function customize for your own environment. Using an easy to adapt example, we will illustrate how you can achieve this step by step. We will also illustrate use cases such scoring function can be used (like embed in a SQL statement). Hints and tips are provided to help you to avoid pain points, and improve development efficiency.

  • S211 - Leverage Kubeflow for your Enterprise data in Db2 for z/OS

    DB2 Sessions: Room C first floor

    Session Evaluation

    A new era of applications focused around cloud and machine learning is becoming increasingly important in large enterprises. Kubeflow is an open source Machine Learning toolkits for Kubernetes. It also provides a platform for building, deploying, and managing multi-step workflows on Kubernetes (based on Docker containers). In this session, we will first walk through a sample application to train, serve, and test a ML model using Kubeflow. Then, we will illustrate how to leverage Kubeflow to create, invoke, and drop a database service.