README for icy-parses (formerly, icToolDist)

This package provides support for extracting integration cost from the output of the Stanford Dependency Parser.

Usage

Usage for plain SENTS files:

./integration_cost.py SENTS_FILE OUTPUT_FILE --stories

where a SENTS_FILE is a standard one-sentence-per-line corpus file. It may or may not be tokenized already. This script assumes they are not tokenized.

For plain SENTS files and sentence-level summary output:

./integration_cost.py SENTS_FILE OUTPUT_FILE --sentfeats

Inventory

• Stanford-Parser/
  contains the Stanford Parser instance, downloaded from the official page

• Tests-StanParser/
  contains some test files, its purpose is to check the output of the stanford parser for special sentences

• integration_cost.py
  main script for running icy-parses

• InvokeStanfordParser.java interface with the stanford-parser, to be later used with popen provided by Asad Sayeed

• InvokeStanfordParser.class
  class file of the previous file
  this is called by the Python scripts

• README.md
  this file

Metadata

Created by Asad Sayeed and/or Jesus Calvillo prior to 2015.
Last modified by David M. Howcroft, April 2017.