Overview: Watson Machine Learning

Using IBM Watson Machine Learning, you can build analytical models and neural networks, trained with your own data, that you can deploy for use in applications.

Watson Machine Learning provides a full range of tools and services so you can build, train, and deploy Machine Learning models. Choose from tools that fully automate the training process for rapid prototyping to tools that give you complete control to create a model that matches your needs.

This graphic illustrates a typical process for a machine learning model.

Note that some tools, such as AutoAI, do some data refining as part of the model building process, but generally you need to prepare data prior to using it in a model. For details, see Preparing data.

IBM Watson Studio tools

Watson Studio is a collaborative environment with graphical tools for designing, training, deploying, and managing models with your Watson Machine Learning services:

• AutoAI experiments automatically preprocesses your data, selects the best estimator for the data, and then generates model candidate pipelines for you to review and compare. Deploy the best performing pipeline as a machine learning model.
• SPSS modeler presents a graphical view of your model while you build it by combining nodes representing objects or actions
• Notebooks provide an interactive programming environment for working with data, testing models, and rapid prototyping
• Experiment builder automates running hundreds of training runs while tracking and storing results
• Decision Optimization model builder guides you through building and solving prescriptive models

IBM Watson Machine Learning architecture and services

Watson Machine Learning is a service on IBM Cloud with features for training and deploying machine learning models and neural networks. Built on a scalable, open-source platform based on Kubernetes and Docker components, Watson Machine Learning enables you to build, deploy, and manage machine learning and deep learning models using:

• Programming Interfaces:
  • Python client library
  • Command line interface
  • REST API
• Training infrastructure:
  • Distributed training to run training jobs across multiple servers
  • Popular frameworks are supported so you can work with your preferred libraries
  • GPUs for faster training
  • Hyperparameter optimization for training complex neural networks
• Deployment infrastructure:
  • Deploy trained models - as a web service, for batch processing, or with streaming data
  • Deploy Python functions to simplify AI solutions