Login Register

Auto-login on future visits

Forgot your password?

Newsletter December 2015

Visualizing differences between confusion marices to find the best error trade-off in multi-class ROC

New interactive tool to choose a good classifier trade-off

Feature highlight

In practice, we often need to tune our classifiers so that certain errors are avoided while others do not matter so much. Our new interactive tool makes this process much simpler and faster.
Watch this video (3min) to find out more.

training course on industrial machine learning

Spring Machine Learning course: 4-8 April 2016


Want to learn from industrial use-cases and gain practical skills?
Mark your calendar: The spring perClass course has been scheduled for 4-8 April 2016 in Delft, NL. It includes a new lecture on Deep Learning.
Join us and bring your own data to push ahead your project!

Learn more about the industrial pattern recognition course register now to the machine learning course for industry

Read more: How was our November course
perClass 4.7 running one-against-one multi-class SVM in Matlab 2015b

New perClass release


perClass 4.7 brings new features and important fixes for Matlab 2015b release that introduced a new computational engine:

  • one-against-one multi-class SVMs
  • interactive visualization of confusion matrix improvements in ROC
  • user-defined filter bank extractors

Read more in the release notes.

recent publications using perClass toolbox in pattern recognition

perClass in use: Latest research publications

From the blog

We are happy to share with you some of the recent publications using perClass:

  • Real-time conveyor belt tracking for x-ray sorting (DeBeers)
  • Hyper-spectral classification of metals (Uni.Liege)
  • Multi-class ROC optimization (Uni.Rouen)

Read more in the blog