Software
New deep learning toolbox
Mathworks presents the 2018b release of Matlab and Simulink. It contains enhancements for deep learning as well as new functions and improvements in all product families.
The latest Deep Learning Toolbox, which replaces the Neural Network Toolbox, provides engineers and scientists with a framework for designing and implementing deep neural networks. With this release, image processing, computer vision, signal processing and systems engineers can use Matlab to more easily design complex network architectures and increase the performance of their deep learning models. The 2018b release also features the Deep Network Designer app, which allows users to create complex network architectures or modify pre-trained networks for transfer learning. Support for cloud providers provides greater computing power when training networks. In addition, subject-specific workflows are supported with apps for ground truth labeling of audio and video data as well as application-specific datastores, which make working with large amounts of data easier and faster.
The GPU Coder increases interference performance by supporting Nvidia libraries and adding optimizations such as auto-tuning, layer fusion and buffer minimization.










