Machine Learning
Neural Networks »
Mathematica Version 11 introduces a high-performance neural network framework with both CPU and GPU training support. A full complement of vision-oriented layers is included, as well as encoders and decoders to make trained networks interoperate seamlessly with the rest of the language. Constructing and training networks often requires only a few lines of code, putting deep learning in the hands of even nonexpert users.
Machine Learning »
Mathematica Version 11 extends and improves its machine learning capabilities. New functions allow users to extract features, reduce dimension, cluster data, optimize hyperparameters, and obtain interpretable models. The feature extraction functionality can be used to visualize datasets or to create semantic distances for search systems. Computer vision is substantially improved: ImageIdentify can recognize more than 10,000 objects, and performances of Classify on images are enhanced. Additionally, machine learning functions accept a wider range of data types.