def train_and_trial(trn_file,
test_file,
clf,
posit_lex_file='positive-words.txt',
nega_lex_file='negative-words.txt',
pickled=False):
""" Train on the training file and test on the testing file,
given a classifier, for the aspect extraction task.
"""
if pickled:
f = open(trn_file, 'rb')
traind = cPickle.load(f)
f.close()
f = open(test_file, 'rb')
testd = cPickle.load(f)
f.close()
else:
traind = XMLParser.create_exs(trn_file)
testd = XMLParser.create_exs(test_file)
posi_words = semeval_util.get_liu_lexicon(posit_lex_file)
negi_words = semeval_util.get_liu_lexicon(nega_lex_file)
senti_dictionary = semeval_util.get_mpqa_lexicon()
#chunker = ConsecutiveChunker(traind['iob'], senti_dictionary)
chunker = clf.train(traind['iob'], senti_dictionary)
print "done training"
guessed_iobs = chunker.evaluate(testd['iob'])
XMLParser.create_xml(testd['orig'], guessed_iobs, testd['id'],
testd['idx'], 'trial_answers.xml')
compute_pr(testd['iob'], guessed_iobs)
parses_tests = ['laptops_test_phaseA-parse.txt','rest_test_phaseA-parse.txt','lap-trial-parse.txt']
results_files = ['lap_phaseA.xml','rest_phaseA.xml','lap-trial_phaseA.xml']
def get_data(dataset_name):
idx = names.index(dataset_name)
return pickle_trains[idx], pickle_tests[idx], parses_trains[idx], parses_tests[idx], results_files[idx]
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument("task_name", help="must be either lap or rest or dummy", type=str)
#later if time
parser.add_argument("-p", help="Specify that train_file is an already learned clf",type=bool, default=False)
parser.add_argument("-dep", help="If true, use dependency parse features", type=bool, default=False)
args = parser.parse_args()
train_file, test_file, parse_train_file, parse_test_file, out_xml_file = get_data(args.task_name)
results = semevalTask4.train_and_trial(train_file, test_file, parse_train_file, parse_test_file,
use_dep=args.dep, pickled=True)
#create results file
f = open(test_file, 'rb')
testd = cPickle.load(f)
f.close()
XMLParser.create_xml(testd['orig'], results, testd['id'], testd['idx'], out_xml_file)
for iob in traind['iob']:
polarities = semeval_util.create_sentiment_sequence(
iob, senti_dictionary, negate_wds)
translated = []
for p, n in polarities:
if p > n:
translated.append('positive')
elif n > p:
translated.append('negative')
else:
translated.append('neutral')
results.append(translated)
semeval_util.compute_sent_acc(traind['polarity'], results)
XMLParser.create_xml(traind['orig'],
traind['iob'],
traind['id'],
traind['idx'],
sentiments=results,
outfile='baseline.xml')
sys.exit()
else:
results = task4_stask2.train_and_trial(train_file, test_file)
#create results file
f = open(test_file, 'rb')
testd = cPickle.load(f)
f.close()
XMLParser.create_xml(testd['orig'],
testd['iob'],
testd['id'],
testd['idx'],
sentiments=results,
senti_dictionary = semeval_util.get_mpqa_lexicon()
negate_wds = semeval_util.negateWords
results = []
for iob in traind['iob']:
polarities = semeval_util.create_sentiment_sequence(iob, senti_dictionary, negate_wds)
translated = []
for p, n in polarities:
if p > n:
translated.append('positive')
elif n > p:
translated.append('negative')
else:
translated.append('neutral')
results.append(translated)
semeval_util.compute_sent_acc(traind['polarity'], results)
XMLParser.create_xml(traind['orig'], traind['iob'], traind['id'], traind['idx'], sentiments=results,
outfile='baseline.xml')
sys.exit()
else:
results = task4_stask2.train_and_trial(train_file, test_file)
#create results file
f = open(test_file, 'rb')
testd = cPickle.load(f)
f.close()
XMLParser.create_xml(testd['orig'], testd['iob'], testd['id'], testd['idx'], sentiments=results, outfile=out_xml_file)
