For anyone interested in Markov chains (aimed at language) in Python, and their relationship to the larger world of language modeling (including "modern" RNNLMs

For anyone interested in Markov chains (aimed at language) in Python, and their relationship to the larger world of language modeling (including "modern" RNNLMs and so on) this post by Yoav Goldberg is an excellent introduction, with clear and simple code [1]. Extending the Markov chain even a few steps can give really impressive generated results. Clone/download [2] if you want to run the code yourself. [1] http://nbviewer.jupyter.org/urls/gist.githubusercontent.com/yoavg/d76121dfde2618422139/raw/564dfb8ad396e987c1f1c45ce3c6b942757cdb9e/lm_example http://nbviewer.jupyter.org/urls/gist.githubusercontent.com/... [2] https://gist.github.com/yoavg/d76121dfde2618422139 https://gist.github.com/yoavg/d76121dfde2618422139