Sep	OCT	Nov
	31
2000	2001	2002

success

fail

About this capture

TIMESTAMPS

Ronan Collobert, collober@idiap.ch

Currently, Torch is developed at IDIAP, in Switzerland.

Author: Ronan Collobert, collober@idiap.ch

Main Contributors:

• Samy Bengio, bengio@idiap.ch
• Johnny Mariethoz, marietho@idiap.ch

It's (again ?) a new machine-learning library, written in C++.
It's distributed under the GPL licence.

It is, and it will be in development forever.
It's for Unix and Linux systems: other systems aren't supported at all.
It has been successfully compiled on Linux, SunOS, FreeBSD and OSF1.

Currently, the main features are the following:

• A lot of things in gradient-machines, that is, machines which could be learned with gradient descent. This includes Multi-Layered Perceptrons, Radial Basis Functions and Mixtures of Experts. In fact there are a lot of small "modules" available (Linear module, Tanh module, SoftMax module...) that you can plug as you want to get what you want.
• Support Vector Machine, in classification and regression.
• A Distribution package which includes Kmeans, Gaussian Mixture Models and Hidden Markov Models.
• A few non-parametric models such as K-nearest-neighbors, Parzen Regression and Parzen Density Estimator.

As it's an open library, we encourage everybody to develop new packages which could then be included in future versions on the official website (as long as it follows the coding guidelines).

The aim of Torch is to provide the state-of-the-art of the best algorithms in machine learning. Therefore...

• If you know C++,

 
• If you're working in machine learning and want to develop your own algorithms,
• or if you want to use well-known machine-learning algorithms

... Torch is made for you!