This is a C implementation of the Sequence to Sequence model described in the paper:
Sequence to Sequence Learning with Neural Networks by Sutskever et.al, NIPS 2014.
This code is not very useful by itself. I originally planned to add CUDA and multi-GPU support. I didn't do so yet partly because my interests shifted, partly because of the release of TensorFlow which makes usage of multiple GPUs easier.
I wrote this code in C, not C++ because I am more familiar with it. However, the each of the layer modules are very "class-like" and if I were to write the code again I would write it in C++ to make of use STL vectors and to create a layer base-class.
For more information I refer to my Python Sequence to Sequence implementation which is arguably more compact.