Anith Gopal

This blog has been migrated from community.research.microsoft.com Original date of blog: November 8, 2010 Original blog author: John Winn Hello Infernauts!! Now that version 2.4 of Infer.NET is released, we’re planning a series of blog posts to describe the new features and capabilities that we’ve added over the last 12 months.    The main focus of this version is performance, so I’m going to start off the series by showing how we’ve made improvements in the speed of running inference for a range of models.  For those of you already using Infer.NET, the nice aspect of these changes is that your existing models will just run faster and use less memory without you having to do a thing! Significant speed ups I’ve picked two common models to compare on: Mixture of Gaussians – this is the multivariate mixture of Gaussians model from this tutorial , with the number of points increased to 30,000.

Here is the original post:
Performance improvements in Infer.NET 2.4