A key to your deep learning research.
'cle' is not a building in which it costs a lot to live,
but a playground which is open to everyone for free.

We try to manage minimum lines of implementation.
Over-engineering is not beautiful.

If you happen to use this library, or use some of the implementations in this library,
please don't forget to cite and give credits to the right libraries (do not remove references in the docstrings).

If you happen to use GF-RNN implementation, please do cite this paper:
Junyoung Chung, Caglar Gulcehre, Kyunghyun Cho and Yoshua Bengio, "Gated Feedback Recurrent Neural Networks", in ICML, 2015.