jwf-zz

#!/bin/bash

# Requires vw (https://github.com/JohnLangford/vowpal_wabbit/wiki/),
# the IMDB dataset (http://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz),
# and the perf utility from http://osmot.cs.cornell.edu/kddcup/software.html.

cat aclImdb/train/labeledBow.feat | \
  sed -n 's/^\([7-9]\|10\)\s/&/p' | \
  sed -e "s/^\([7-9]\|10\)\s//" | \
  awk '{ print "1 '"'"'pos_" (NR-1) " |features " $0}' > train.vw