def time_analysis(genomes, total_length, chunk_size='m'):
# analysis dictionary holds all data about the algorithms
analysis = {'substring_length':len(args.pattern), 'substring': args.pattern,
'text_length':total_length}
# Get time to run algorithm on all substrings
boyermoore_data = {'name': 'boyermoore'}
nlogn_data = {'name': 'nlogn'}
nlogm_data = {'name': 'nlogm'}
opencv_data = {'name': 'opencv'}
with Timer() as t:
bm_matches = bm.boyer_moore_mult_match_index(genomes, args.pattern)
boyermoore_data['time'] = t.msecs
boyermoore_data['accuracy'] = 1
with Timer() as t:
nlogn_matches = fft.fft_match_index_n_sq_log_n(genomes, args.pattern)
nlogn_data['time'] = t.msecs
accuracy = 0
for i in range(len(nlogn_matches)):
i_accuracy = len(nlogn_matches[i]) / len(bm_matches[i])
accuracy += i_accuracy
nlogn_data['accuracy'] = accuracy / len(bm_matches)
nlogm_matches = []
total_length = 0
with Timer() as t:
for g in genomes:
total_length += len(g)
nlogm_matches.append(fft.fft_match_index_n_log_m(g, args.pattern, chunk_size))
nlogm_data['time'] = t.msecs
accuracy = 0
for i in range(len(nlogm_matches)):
i_accuracy = len(nlogm_matches[i]) / len(bm_matches[i])
accuracy += i_accuracy
nlogm_data['accuracy'] = accuracy / len(bm_matches)
with Timer() as t:
cvmatch.cv_match_index(genomes, args.pattern)
opencv_data['time'] = t.msecs
accuracy = 0
for i in range(len(nlogm_matches)):
i_accuracy = len(nlogm_matches[i]) / len(bm_matches[i])
accuracy += i_accuracy
opencv_data['accuracy'] = accuracy / len(bm_matches)
algorithms = []
algorithms.append(boyermoore_data)
algorithms.append(nlogn_data)
algorithms.append(nlogm_data)
algorithms.append(opencv_data)
analysis['algorithms'] = algorithms
# make pretty json format
print json.dumps(analysis)
def k_analysis(genomes):
# analysis dictionary holds all data about the algorithms
analysis = {
'substring_length': len(args.pattern),
'substring': args.pattern
}
# Get time to run algorithm on all substrings
boyermoore_data = {'name': 'boyermoore'}
nlogn_data = {'name': 'nlogn'}
opencv_data = {'name': 'opencv'}
for i in range(0, len(genomes)):
analysis['k'] = i
if len(genomes[:i]) == 0:
k_genomes = [genomes[0]]
else:
k_genomes = genomes[:i]
with Timer() as t:
bm_matches = bm.boyer_moore_mult_match_index(
k_genomes, args.pattern)
boyermoore_data['time'] = t.msecs
boyermoore_data['accuracy'] = 1
with Timer() as t:
nlogn_matches = fft.fft_match_index_n_sq_log_n(
k_genomes, args.pattern)
nlogn_data['time'] = t.msecs
accuracy = 0
for i in range(len(nlogn_matches)):
i_accuracy = len(nlogn_matches[i]) / len(bm_matches[i])
accuracy += i_accuracy
nlogn_data['accuracy'] = accuracy / len(bm_matches)
with Timer() as t:
cvmatch.cv_match_index(k_genomes, args.pattern)
opencv_data['time'] = t.msecs
algorithms = []
algorithms.append(boyermoore_data)
algorithms.append(nlogn_data)
algorithms.append(opencv_data)
analysis['algorithms'] = algorithms
# make pretty json format
print json.dumps(analysis)
def k_analysis(genomes):
# analysis dictionary holds all data about the algorithms
analysis = {'substring_length':len(args.pattern), 'substring': args.pattern}
# Get time to run algorithm on all substrings
boyermoore_data = {'name': 'boyermoore'}
nlogn_data = {'name': 'nlogn'}
opencv_data = {'name': 'opencv'}
for i in range(0,len(genomes)):
analysis['k'] = i
if len(genomes[:i]) == 0:
k_genomes = [genomes[0]]
else:
k_genomes = genomes[:i]
with Timer() as t:
bm_matches = bm.boyer_moore_mult_match_index(k_genomes, args.pattern)
boyermoore_data['time'] = t.msecs
boyermoore_data['accuracy'] = 1
with Timer() as t:
nlogn_matches = fft.fft_match_index_n_sq_log_n(k_genomes, args.pattern)
nlogn_data['time'] = t.msecs
accuracy = 0
for i in range(len(nlogn_matches)):
i_accuracy = len(nlogn_matches[i]) / len(bm_matches[i])
accuracy += i_accuracy
nlogn_data['accuracy'] = accuracy / len(bm_matches)
with Timer() as t:
cvmatch.cv_match_index(k_genomes, args.pattern)
opencv_data['time'] = t.msecs
algorithms = []
algorithms.append(boyermoore_data)
algorithms.append(nlogn_data)
algorithms.append(opencv_data)
analysis['algorithms'] = algorithms
# make pretty json format
print json.dumps(analysis)
def time_analysis(genomes, total_length, chunk_size='m'):
# analysis dictionary holds all data about the algorithms
analysis = {
'substring_length': len(args.pattern),
'substring': args.pattern,
'text_length': total_length
}
# Get time to run algorithm on all substrings
boyermoore_data = {'name': 'boyermoore'}
nlogn_data = {'name': 'nlogn'}
nlogm_data = {'name': 'nlogm'}
opencv_data = {'name': 'opencv'}
with Timer() as t:
bm_matches = bm.boyer_moore_mult_match_index(genomes, args.pattern)
boyermoore_data['time'] = t.msecs
boyermoore_data['accuracy'] = 1
with Timer() as t:
nlogn_matches = fft.fft_match_index_n_sq_log_n(genomes, args.pattern)
nlogn_data['time'] = t.msecs
accuracy = 0
for i in range(len(nlogn_matches)):
i_accuracy = len(nlogn_matches[i]) / len(bm_matches[i])
accuracy += i_accuracy
nlogn_data['accuracy'] = accuracy / len(bm_matches)
nlogm_matches = []
total_length = 0
with Timer() as t:
for g in genomes:
total_length += len(g)
nlogm_matches.append(
fft.fft_match_index_n_log_m(g, args.pattern, chunk_size))
nlogm_data['time'] = t.msecs
accuracy = 0
for i in range(len(nlogm_matches)):
i_accuracy = len(nlogm_matches[i]) / len(bm_matches[i])
accuracy += i_accuracy
nlogm_data['accuracy'] = accuracy / len(bm_matches)
with Timer() as t:
cvmatch.cv_match_index(genomes, args.pattern)
opencv_data['time'] = t.msecs
accuracy = 0
for i in range(len(nlogm_matches)):
i_accuracy = len(nlogm_matches[i]) / len(bm_matches[i])
accuracy += i_accuracy
opencv_data['accuracy'] = accuracy / len(bm_matches)
algorithms = []
algorithms.append(boyermoore_data)
algorithms.append(nlogn_data)
algorithms.append(nlogm_data)
algorithms.append(opencv_data)
analysis['algorithms'] = algorithms
# make pretty json format
print json.dumps(analysis)
