Training cloud-based machine learning models for deployment in enterprise applications requires access to the domain-specific data that you want to use. Whether your data is on premises – the kind that lives on your local machine, within your data center, or behind a firewall – or already in the cloud, IBM Watson Studio allows you to easily and securely connect to your data source for initial model training and continuous learning, no matter where your data is located.

In the code patterns referenced below, we solve a problem for the City of Chicago using the Watson Studio Model Builder to model building code violations. We show how to do this using SQL data that is either on premises or on the cloud with Db2 offerings. We’ll predict which buildings are most likely to fail building inspections to help save time and resources for city inspectors.

Is your data in the cloud? Watson Studio can natively use data assets like Db2 Warehouse on Cloud located within IBM Cloud for model training and continuous learning. Continuous learning is a process where your machine learning models automatically improve over time, as your training data evolves and grows, thus closing the feedback loop between training data and deployed models. Configure your required triggers for retraining, and new models with competing algorithms are automatically trained, evaluated for performance, and conditionally deployed for immediate use by your own applications, all without having to update the applications themselves. Take a look at the Continuous learning with WML and Db2 Warehouse on Cloud code pattern to try it out.

Is your data on premises? On-premises data storage like in a Db2 database lets you have full control over your data, including its security and integrity. Often, maintaining a custom on-premises security posture presents a unique challenge when adopting hybrid cloud. To overcome this type of hurdle, Watson Studio uses the Secure Gateway Service to allow Watson Studio to securely access and train on your data sets on Db2. By colocating the lightweight Secure Gateway client along with your data, you can establish a secure, persistent connection between your environment and the cloud, with powerful options for implementing custom security policies on both ends of the connection. The Train a cloud-based machine learning model from Db2 on-premises data code pattern shows how it works.