def main():
#optional arguments:
argument_parser = argparse.ArgumentParser(description='Bipartite Clustering algorithm.')
argument_parser.add_argument('-wc','--wc',type=int,default=150,help='number of word clusters. default=100.')
argument_parser.add_argument('-dc','--dc',type=int,default=75,help='number of document clusters. default=100.')
argument_parser.add_argument('-test','--test',help="Work with test set.",action='store_true')
parsed_arg = argument_parser.parse_args()
test_docVector = "C:\Smuzi\CMU\ML Txt mining\HW\HW2\data\HW2_test.docVectors"
gold_std = "C:\Smuzi\CMU\ML Txt mining\HW\HW2\data\HW2_dev.gold_standards"
dev_docVector = "C:\Smuzi\CMU\ML Txt mining\HW\HW2\data\HW2_dev.docVectors"
word_df = "C:\Smuzi\CMU\ML Txt mining\HW\HW2\data\HW2_dev.df"
if parsed_arg.test:
filename = test_docVector
else:
filename = dev_docVector
doc_output = "doc_clusters.out"
word_output = "word_clusters.out"
begin_time = time.time()
criteria = 0.01
bpc.main(parsed_arg.wc,parsed_arg.dc,filename,doc_output,word_output,word_df,criteria)
output = {}
output['runtime'] = time.time()-begin_time
print output
sys.argv = ['eval.py',doc_output,gold_std]
execfile('eval.py')
print('')
print('')
def main():
#optional arguments:
#argument_parser = argparse.ArgumentParser(description='Bipartite Clustering algorithm.')
#argument_parser.add_argument('-wc','--wc',type=int,default=100,help='number of word clusters. default=100.')
#argument_parser.add_argument('-dc','--dc',type=int,default=50,help='number of document clusters. default=100.')
#argument_parser.add_argument('-test','--test',help="Work with test set.",action='store_true')
#parsed_arg = argument_parser.parse_args()
test_docVector = "HW2_test.docVectors"
gold_std = "HW2_dev.gold_standards"
dev_docVector = "HW2_dev.docVectors"
word_df = "HW2_dev.df"
filename = dev_docVector
doc_output = "doc_clusters.out"
word_output = "word_clusters.out"
begin_time = time.time()
for wc in (50,100,150,200):
for dc in (50,100,150,200):
begin_time = time.time()
bpc.main(wc,dc,filename,doc_output,word_output,word_df,0.1)
output = {}
output['wc']=wc
output['dc']=dc
output['runtime'] = time.time()-begin_time
print output
sys.argv = ['eval.py',doc_output,gold_std]
execfile('eval.py')
print('')
print('')
def start(event):
global ucontrol, listener
global image_tk
global testing
testing = False
if image_tk is not None:
del image_tk
clear_can()
if listener is None:
listener = MyListener()
if ucontrol is None:
ucontrol = bpc.uControl(listener)
try:
name = "test%d.uct" % start.tid
sys, dia = bpc.main(name, listener, ucontrol, abort=abort)
image_fn = 'images/%s.png' % name
figure(620)
title('%.0f/%.0f' % (sys, dia))
savefig(image_fn)
start.tid += 1
im = Image.open(image_fn)
display_image(im)
can.update_idletasks()
clf()
start_b.top()
finally:
ucontrol.deflate(10, fast=True)
def start(event):
global ucontrol, listener
global image_tk
global testing
testing = False
if image_tk is not None:
del image_tk
clear_can()
if listener is None:
listener = MyListener()
if ucontrol is None:
ucontrol = bpc.uControl(listener)
try:
name = "test%d.uct" % start.tid
sys, dia = bpc.main(name, listener, ucontrol, abort=abort)
image_fn = 'images/%s.png' % name
figure(620)
title('%.0f/%.0f' % (sys, dia))
savefig(image_fn)
start.tid += 1
im = Image.open(image_fn)
display_image(im)
can.update_idletasks()
clf()
start_b.top()
finally:
ucontrol.deflate(10, fast=True)
def main():
test_docVector = "C:\Smuzi\CMU\ML Txt mining\HW\HW2\data\HW2_test.docVectors"
gold_std = "C:\Smuzi\CMU\ML Txt mining\HW\HW2\data\HW2_dev.gold_standards"
dev_docVector = "C:\Smuzi\CMU\ML Txt mining\HW\HW2\data\HW2_dev.docVectors"
word_df = "C:\Smuzi\CMU\ML Txt mining\HW\HW2\data\HW2_dev.df"
filename = test_docVector
doc_output = "usmadja-test-clusters.txt"
word_output = "test-word_clusters.out"
begin_time = time.time()
criteria = 0.01
bpc.main(150,75,filename,doc_output,word_output,word_df,criteria)
output = {}
output['runtime'] = time.time()-begin_time
print output
def main():
test_docVector = "C:\Smuzi\CMU\ML Txt mining\HW\HW2\data\HW2_test.docVectors"
gold_std = "C:\Smuzi\CMU\ML Txt mining\HW\HW2\data\HW2_dev.gold_standards"
dev_docVector = "C:\Smuzi\CMU\ML Txt mining\HW\HW2\data\HW2_dev.docVectors"
word_df = "C:\Smuzi\CMU\ML Txt mining\HW\HW2\data\HW2_dev.df"
filename = dev_docVector
doc_output = "doc_clusters.out"
word_output = "word_clusters.out"
for criteria in (1,0.1,0.01,0.001):
begin_time = time.time()
bpc.main(150,75,filename,doc_output,word_output,word_df,criteria)
output={}
output['criteria']= criteria
output['runtime'] = time.time()-begin_time
print output
sys.argv = ['eval.py',doc_output,gold_std]
execfile('eval.py')
print('')
print('')
