Torch is a scientific computing framework with wide support for machine learning algorithms. It is easy to use and efficient, thanks to an easy and fast scripting language, LuaJIT, and an underlying C/CUDA implementation.

Source : http://torch.ch/

The main goal of torch is to implement Deep Neural Networks. The Torch community did a really great job to have a flexible and easy framework to do so.
The Lua language was picked because it is simple, flexible and fast thanks to LuaJit.

Yet, Lua code can quickly become crappy, tangled, inefficient … and bugged. This can become a big issues since Deep Learning is mainly based on experimental results. Results must be reproducible, code must be shared. However, there are little information regarding torch good practice and torch programming pitfalls.

This blog aims to provide some tutorials about how to use Torch with the right way. New Tutorials will be added upon time to help new Torch users.

If you are looking for bug reports, some basic helps, go to the forums (1) (2), many people (including me) will be happy to help!

If you are looking for a particular kind of tutorial, feel free to email me.