def __init__(self):
# This determines source location on where to caputer picture
# QuickTime - MacOS has record feature for phone (best)
# WebCam - Use OpenCV to capture photo (untested)
self.use_quicktime = False
self.use_webcam = False
self.use_input = False
# The filename of picture (no extension means we're capturing image)
self.picture = 'source'
# Default location of where to work on self.picture
self.location = os.getcwd()
# Replace with your own auth file name
self.google_auth_json = 'blissend.json'
# Default the language for wikipedia searches
self.wiki = wikipediaapi.Wikipedia('en')
self.vb = Vocabulary()
# The OCR text
self.raw = ''
# The information we ultimately wanted to be analyzed
self.question = ''
self.answers = {}
self.definitions = {}
# For debugging
self.verbose = False
def __init__(self):
# QuickTime - MacOS has record feature for phone (best)
self.use_quicktime = False
self.use_input = False
# the filename of the image (no extension = capturing image)
self.picture = 'source'
# location of where to work on self.picture
self.location = os.getcwd()
# Replace with your own auth file name
self.google_auth_json = 'HQproject-a1a4e25e4b45.json'
# wikipedia setting (english)
self.wiki = wikipediaapi.Wikipedia('en')
self.vb = Vocabulary()
# The OCR text (directly converted from image)
self.raw = ''
# processed texts
self.question = ''
self.question_nouns = ''
self.answers = {}
self.lookup_info = {}
# For debugging
self.times = {}
self.verbose = False
from options_loader import *
if __name__ == '__main__':
#Vocab = loadFromPKL('../../vocab/gigaword.pkl')
log_vocab = mylog(logFile='log/log_vocab')
options = optionsLoader(log_vocab, True)
fileName = options['primary_dir']
inputCorpus = [fileName + options['trainSet'] + '.Ndocument']
outputCorpus = [fileName + options['trainSet'] + '.Nsummary']
Vocab = Vocabulary(options,
inputCorpus=inputCorpus,
outputCorpus=outputCorpus)
log_vocab.log(
str(Vocab.full_size) + ', ' + str(Vocab.n_in) + ', ' +
str(Vocab.n_out))
saveToPKL(fileName + sys.argv[1] + '.Vocab', Vocab)
f = codecs.open(fileName + sys.argv[1] + '.i2w', 'w', encoding='utf-8')
for item in Vocab.i2w:
if (item == '<unk>' or item == '<s>'):
print >> f, item, 'NAN'
else:
print >> f, item, Vocab.typeFreq[item]