Missing context from the README: This library has powered most of CMU's NLP papers for the last year or so. During that period, the group has been enormously successful. For example, here are the papers on which Chris Dyer is a co-author: http://www.cs.cmu.edu/~cdyer/pubs.html

This network has gone under the confusing name CNN as it is a neural network framework implemented in C++

It recently changed name, which I believe prompts this submission.

It has powered a number of NLP research papers and projects.

It has some really interesting features in terms of how data can flow through the graph, as you can dynamically restructure the graph for each datapoint. I assume that is why they've gone with dynet.

I am wondering what do they mean by "works well with networks that have dynamic structures that change for every training instance".

Can somebody explain why it could be useful and point towards some use case?

I checked the only currently given example https://github.com/clab/dynet/blob/master/doc/tutorial.md and it seems to be just a very simple neural network.

Their papers indicate that their main interest is in textual models that is a bit different from the main focus of my interest.

It's also cute because it's from Chris Dyer's group.

Be aware, there is existing DyNet protocol in Dynalite systems: https://en.wikipedia.org/wiki/Dynalite