We're pleased to announce the release of Mathematica 11.3. The latest version expands Mathematica and the Wolfram Language's functionality in mathematical computation, audio and image processing, machine learning and more, as well as introduces several new front end features. Here are some highlights:
Mathematical Computation
Audio and Image Processing
Machine Learning and Neural Networks
- FacialFeatures for extracting facial characteristics such as age, emotion and landmarks
- FeatureSpacePlot3D for plotting dimension-reduced feature spaces in 3D
- Improved automation for Classify and Predict, new properties and measures for ClassifierMeasurements and on-the-fly model evaluation timing for ClassifierInformation
- Progress reporting, additional pre-trained models, new layer types and performance improvements for the neural network framework
- Several new functions for in-place modification of neural networks and layers
Front End Features
- Bring live interactivity to your technical presentations with Wolfram Presenter Tools
- Exchange text, code and other content in real time using the notebook-based Wolfram Chat interface (accessible via File > New > Chat)
Version 11.3 also includes a new integrated framework for browsing, importing and generating system models for simulation and data analysis, as well as functionality for interacting with blockchains. For more information about the latest additions, read through Stephen Wolfram's launch-day blog post and take a look at the summary of new features in 11.3.