The tutorial Serverless Web Application and API describes how to develop serverless APIs and web applications via IBM Functions, the IBM Functions API Gateway and a No-SQL database. The static web resources are hosted via GitHub Pages. I've open sourced an extension to the tutorial that allows hosting web resources directly via Apache OpenWhisk. (more…)

This week my colleagues Stephan Reimann and Thomas Südbroecker and I attended data2day in Heidelberg. data2day is a technical conference for data scientists and developers focussed on topics related to big data, artificial intelligence and machine learning. Below are some pictures, most of them taken by Thomas. (more…)

In order to train the object detection model for my sample TensorFlow Object Detection for Anki Overdrive Cars, I used a custom Docker container. I've extended the description of my repo to also document how Fabric for Deep Learning (FfDL) can be used to run this container. (more…)

I finally found some time to start playing with Knative. It took me a few hours to get a simple sample running. Below I describe some pitfalls when installing Knative and tips how to avoid them. With the official instructions and these tips it shouldn't take longer than half an hour to run your first container on Knative. (more…)

TensorFlow Object Detection is a powerful technology to recognize different objects in images including their positions. The trained Object Detection models can be run on mobile and edge devices to execute predictions really fast. I've used this technology to build a demo where Anki Overdrive cars and obstacles are detected via an iOS app. When obstacles are detected, the cars are stopped automatically. (more…)

For software development there are many methodologies, patterns and techniques to build, deploy and run applications. DevOps is the state of the art methodology which describes a software engineering culture with a holistic view of software development and operation. For data science there is a lot of information how machine and deep learning models can be built. The operational aspects seem to still be evolving. I'm currently trying to understand better how to deploy models in the cloud and ...

Early this year IBM announced Deep Learning as a Service within Watson Studio. The core of this service is available as open source and can be run on Kubernetes clusters. This allows developers and data scientists to train models with confidential data on-premises, for example on the Kubernetes-based IBM Cloud Private. (more…)

Building machine and deep learning models from scratch is often not trivial, not for developers and sometimes not even for data scientists. Fortunately over the last years several models have been developed and shared that can be reused and sometimes extended. This allows developers adding AI to applications without having to be data scientists. (more…)

Many developers, including myself, want to use AI in their applications. Building machine learning models however, often requires a lot of expertise and time. This article describes a technique called AutoML which can be used by developers to build models without having to be data scientists. While developers only have to provide the data and define the goals, AutoML figures out the best model automatically. (more…)

While it has been possible to deploy TensorFlow models to mobile and embedded devices via TensorFlow for Mobile for some time, Google released an experimental version of TensorFlow Lite as an evolution of TensorFlow for Mobile at the end of last year. This new functionality allows building exciting AI scenarios on edge devices and the performance of the models is amazing. (more…)