Nowadays A.I. is the hype and everyone is talking on this awesome new neural network they trained. However, what is often overlooked is how to integrate such technology into a real system. In this talk I will show some results from our research lab on training a neural network to grasp objects. Instead of delving into the deep learning details however, I will focus on how we benefit from OSGi to integrate this onto a real robot system, and create some fancy API models using Promises and PushStreams.
Nowadays AI is reaching new heights on the hype cycle, especially due to recent advances in deep learning techniques. A lot of deep learning frameworks exist for creating and training deep neural networks, the most popular ones being PyTorch and TensorFlow. However, how to integrate, deploy and manage these neural networks in complex software systems is often overlooked. In this talk we show how OSGi can be used as a modular runtime for deep learning models. We embed those models inside OSGi bundles, and use the extender pattern to make these available as OSGi services.