def main(crowdflower_csv, pos_data_dir, output_file, debug):
""" Transform the crowdflower results into training data
:param file crowdflower_csv: The CSV containing crowdflower data
:param str pos_data_dir: The directory containing POS tagging for each sentence
:param file output_file: The file in which to write the training data
:return: 0
:rtype: int
"""
results = read_full_results(crowdflower_csv)
if debug:
print 'Results from crowdflower'
print json.dumps(results, indent=2)
set_majority_vote_answer(results)
if debug:
print 'Computed majority vote'
print json.dumps(results, indent=2)
tag_entities(results)
if debug:
print 'Entities tagged'
print json.dumps(results, indent=2)
output = produce_training_data(results, pos_data_dir, debug)
for l in output:
output_file.write(l.encode('utf-8') + '\n')
return 0
def main(crowdflower_csv, pos_data_dir, output_file, debug):
""" Transform the crowdflower results into training data
:param file crowdflower_csv: The CSV containing crowdflower data
:param str pos_data_dir: The directory containing POS tagging for each sentence
:param file output_file: The file in which to write the training data
:return: 0
:rtype: int
"""
results = read_full_results(crowdflower_csv)
if debug:
print 'Results from crowdflower'
print json.dumps(results, indent=2)
set_majority_vote_answer(results)
if debug:
print 'Computed majority vote'
print json.dumps(results, indent=2)
tag_entities(results)
if debug:
print 'Entities tagged'
print json.dumps(results, indent=2)
output = produce_training_data(results, pos_data_dir, debug)
for l in output:
output_file.write(l.encode('utf-8') + '\n')
return 0
def main(crowdflower_output, num_judgments):
"""
this script computes the agreement of judgments given in the crowdflower
job using a metric called Fleiss kappa
"""
cf_results = read_full_results(crowdflower_output)
mat = compute_matrix(cf_results, num_judgments)
print computeFleissKappa(mat)
def main(crowdflower_output, num_judgments):
"""
this script computes the agreement of judgments given in the crowdflower
job using a metric called Fleiss kappa
"""
cf_results = read_full_results(crowdflower_output)
mat = compute_matrix(cf_results, num_judgments)
print computeFleissKappa(mat)
def main(crowdflower_output, num_judgments):
"""
this script computes the agreement of judgments given in the crowdflower
job using a metric called Fleiss kappa
:param file crowdflower_output: CSV file containing the results from crowdflower
:param int num_judgments: Consider only this number of results, skip if not enough
"""
cf_results = read_full_results(crowdflower_output)
mat = compute_matrix(cf_results, num_judgments)
print computeFleissKappa(mat)
def main(crowdflower_output, num_judgments):
"""
this script computes the agreement of judgments given in the crowdflower
job using a metric called Fleiss kappa
:param file crowdflower_output: CSV file containing the results from crowdflower
:param int num_judgments: Consider only this number of results, skip if not enough
"""
cf_results = read_full_results(crowdflower_output)
mat = compute_matrix(cf_results, num_judgments)
print computeFleissKappa(mat)
def main(crowdflower_csv, pos_data_dir, output_file, debug):
results = read_full_results(crowdflower_csv)
if debug:
print 'Results from crowdflower'
print json.dumps(results, indent=2)
set_majority_vote_answer(results)
if debug:
print 'Computed majority vote'
print json.dumps(results, indent=2)
tag_entities(results)
if debug:
print 'Entities tagged'
print json.dumps(results, indent=2)
output = produce_training_data(results, pos_data_dir, debug)
for l in output:
output_file.write(l.encode('utf-8') + '\n')
return 0
def main(crowdflower_csv, pos_data_dir, output_file, debug):
results = read_full_results(crowdflower_csv)
if debug:
print 'Results from crowdflower'
print json.dumps(results, indent=2)
set_majority_vote_answer(results)
if debug:
print 'Computed majority vote'
print json.dumps(results, indent=2)
tag_entities(results)
if debug:
print 'Entities tagged'
print json.dumps(results, indent=2)
output = produce_training_data(results, pos_data_dir, debug)
for l in output:
output_file.write(l.encode('utf-8') + '\n')
return 0
def main(crowdflower_output, num_judgments):
cf_results = read_full_results(crowdflower_output)
mat = compute_matrix(cf_results, num_judgments)
print computeFleissKappa(mat)