Expand How Does it Work section

[nicholas-leonard]

Could you guys expand on the how does it work section in the docs? Specifying how the output, gradInput, weight, gradWeight and intermediate state buffers are handled would be a plus.

[fmassa]

Sure, I'll improve the README with more information about it, thanks for mentioning it Nicholas !
I'll find some time this week to do it.

[geert56]

I would love to see some more algorithm documentation: what is the approach you take, what do the main functions do, what are the assumptions in terms of existing Torch data structures. It would greatly help others in improving and extending the code. Thanks.

[fmassa]

Hi @geert56,
I have prepared some initial version of a more complete documentation in #16 .
I didn't add much more since then because I have been quite busy over the past few months, but let me know what else you'd like to see in the documentation and I'll try to add it when I have time (probably in October).