def test_accuracy():
for size in size_list:
for file_size in filesize_list:
kwargs = {
'true_posi':'right',
'average':'binary',
'truth_column':'label',
'pred_column':'label',
}
result_file = 'report/test_accuracy_{}_{}.txt'.format(size, file_size)
if os.path.exists(result_file):
print('skip {}'.format(result_file))
continue
dfmaker = LabelDataMaker(size, file_size, 100, 2)
multip = len(dfmaker)
if multip == 0:
continue
timer = Timer()
print('running {}'.format(result_file))
new_data = next(dfmaker)
timer(accuracy, new_data, new_data, **kwargs)
cost = timer.avg_cost * multip
with open(result_file, 'w') as f:
f.write('total,each,number\n')
f.write(str(cost)+","+str(timer.avg_cost)+","+str(multip))
def test_kmeans():
for size in size_list:
for file_size in filesize_list:
kwargs = {
'n_cluster':'2',
'max_iter':'300',
'predict_labels':'label',
'store_origin':'False',
'n_jobs':'10',
}
result_file = 'report/test_kmeans_{}_{}.txt'.format(size, file_size)
if os.path.exists(result_file):
print('skip {}'.format(result_file))
continue
dfmaker = LabelDataMaker(size, file_size, 100, 2)
multip = len(dfmaker)
if multip == 0:
continue
timer = Timer()
print('running {}'.format(result_file))
timer(kmeans, next(dfmaker), **kwargs)
cost = timer.avg_cost * multip
with open(result_file, 'w') as f:
f.write('total,each,number\n')
f.write(str(cost)+","+str(timer.avg_cost)+","+str(multip))
def test_iforest():
for size in size_list:
for file_size in filesize_list:
kwargs = {
'contamination': '0.1',
'n_jobs': '10',
}
result_file = 'report/test_iforest_{}_{}.txt'.format(
size, file_size)
if os.path.exists(result_file):
print('skip {}'.format(result_file))
continue
dfmaker = LabelDataMaker(size, file_size, 100, 2)
multip = len(dfmaker)
if multip == 0:
continue
timer = Timer()
print('running {}'.format(result_file))
new_data = next(dfmaker)
timer(outlier_iforest, new_data, **kwargs)
cost = timer.avg_cost * multip
with open(result_file, 'w') as f:
f.write('total,each,number\n')
f.write(
str(cost) + "," + str(timer.avg_cost) + "," + str(multip))
def test_random():
for size in size_list:
for file_size in filesize_list:
kwargs = {}
result_file = 'report/test_random_{}_{}.txt'.format(
size, file_size)
if os.path.exists(result_file):
print('skip {}'.format(result_file))
continue
dfmaker = LabelDataMaker(size, file_size, 100, 2)
multip = len(dfmaker)
if multip == 0:
continue
timer = Timer()
print('running {}'.format(result_file))
timer(random, next(dfmaker), **kwargs)
cost = timer.avg_cost * multip
with open(result_file, 'w') as f:
f.write('total,each,number\n')
f.write(
str(cost) + "," + str(timer.avg_cost) + "," + str(multip))
def test_sort():
for size in size_list:
for file_size in filesize_list:
kwargs = {
'columns': '1',
'ascending': 'True',
'na_position': 'last',
}
result_file = 'report/test_sort_{}_{}.txt'.format(size, file_size)
if os.path.exists(result_file):
print('skip {}'.format(result_file))
continue
dfmaker = LabelDataMaker(size, file_size, 100, 2)
multip = len(dfmaker)
if multip == 0:
continue
timer = Timer()
print('running {}'.format(result_file))
timer(sort, next(dfmaker), **kwargs)
cost = timer.avg_cost * multip
with open(result_file, 'w') as f:
f.write('total,each,number\n')
f.write(
str(cost) + "," + str(timer.avg_cost) + "," + str(multip))