Evaluate the predictions of RST discourse parsers.
The initial release of the script and data is a companion to our short paper at EMNLP 2017:
Mathieu Morey, Philippe Muller and Nicholas Asher. 2017. How much progress have we made on RST discourse parsing? A replication study of recent results on the RST-DT. To appear at EMNLP 2017.
Shafiq Joty's implementation of the standard evaluation metrics for RST described in Marcu's 2000 book: http://alt.qcri.org/tools/discourse-eval/