def nlogm_chunk_analysis(genomes, chunk_max, total_length):
# 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'}
nlogm_data = {'name': 'nlogm'}
with Timer() as t:
bm_matches = bm.boyer_moore_mult_match_index(genomes, args.pattern)
boyermoore_data['time'] = t.msecs
boyermoore_data['accuracy'] = 1
for i in range(3,chunk_max,3):
nlogm_matches = []
with Timer() as t:
for g in genomes:
nlogm_matches.append(fft.fft_match_index_n_log_m(g, args.pattern, chunk_size=i))
nlogm_data['time'] = t.msecs
nlogm_data['chunk_size'] = i
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)
algorithms = []
algorithms.append(boyermoore_data)
algorithms.append(nlogm_data)
analysis['algorithms'] = algorithms
# make pretty json format
print json.dumps(analysis)
def test_chunk_sizes(self):
text = "AAACCCAAA"
chunk_size = 'm'
pattern = "CC"
self.assertTrue((
fftmatch.fft_match_index_n_log_m(text, pattern, chunk_size) == \
np.array(boyermoore.boyer_moore_match_index(text,pattern))).all())
def test_chunk_sizes(self):
text = "AAACCCAAA"
chunk_size = 'm'
pattern = "CC"
self.assertTrue((
fftmatch.fft_match_index_n_log_m(text, pattern, chunk_size) == \
np.array(boyermoore.boyer_moore_match_index(text,pattern))).all())
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 nlogm_chunk_analysis(genomes, chunk_max, total_length):
# 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'}
nlogm_data = {'name': 'nlogm'}
with Timer() as t:
bm_matches = bm.boyer_moore_mult_match_index(genomes, args.pattern)
boyermoore_data['time'] = t.msecs
boyermoore_data['accuracy'] = 1
for i in range(3, chunk_max, 3):
nlogm_matches = []
with Timer() as t:
for g in genomes:
nlogm_matches.append(
fft.fft_match_index_n_log_m(g, args.pattern, chunk_size=i))
nlogm_data['time'] = t.msecs
nlogm_data['chunk_size'] = i
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)
algorithms = []
algorithms.append(boyermoore_data)
algorithms.append(nlogm_data)
analysis['algorithms'] = algorithms
# make pretty json format
print json.dumps(analysis)
title = title + str(count[title])
genomes[title] = genome
sorted_genomes = collections.OrderedDict(sorted(genomes.items(),
key=lambda t: t[0]))
genome_strings = sorted_genomes.values()
genome_titles = sorted_genomes.keys()
# Parse args
if args.algorithm == 'nlogn':
for gn in genomes:
matches = fft.fft_match_index_n_log_n(genomes[gn], args.pattern)
print gn, ': Found matches at indices', matches.tolist()
elif args.algorithm == 'nlogm':
if len(genomes) > 1:
matches = fft.fft_match_index_n_sq_log_m(genomes.values(),\
args.pattern[0], args.b)
print 'found matches at', matches.tolist()
else:
for gn in genomes:
matches = fft.fft_match_index_n_log_m(genomes[gn], args.pattern[0],args.b)
print gn, ': Found matches at indices', matches.tolist()
elif args.algorithm == 'boyermoore':
for gn in genomes:
matches = bm.boyer_moore_match_index(genomes[gn], args.pattern)
print gn, ': Found matches at indices', matches.tolist()
elif args.algorithm == 'opencv':
matches = cvmatch.cv_match_index_chunk(genomes.values(), args.pattern[0], args.b)
print genomes[genomes.keys()[0]]
print genomes.keys(), ': Found matches at indices', matches.tolist()
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)
sorted_genomes = collections.OrderedDict(
sorted(genomes.items(), key=lambda t: t[0]))
genome_strings = sorted_genomes.values()
genome_titles = sorted_genomes.keys()
# Parse args
if args.algorithm == 'nlogn':
for gn in genomes:
matches = fft.fft_match_index_n_log_n(genomes[gn], args.pattern)
print gn, ': Found matches at indices', matches.tolist()
elif args.algorithm == 'nlogm':
if len(genomes) > 1:
matches = fft.fft_match_index_n_sq_log_m(genomes.values(),\
args.pattern[0], args.b)
print 'found matches at', matches.tolist()
else:
for gn in genomes:
matches = fft.fft_match_index_n_log_m(genomes[gn], args.pattern[0],
args.b)
print gn, ': Found matches at indices', matches.tolist()
elif args.algorithm == 'boyermoore':
for gn in genomes:
matches = bm.boyer_moore_match_index(genomes[gn], args.pattern)
print gn, ': Found matches at indices', matches.tolist()
elif args.algorithm == 'opencv':
matches = cvmatch.cv_match_index_chunk(genomes.values(), args.pattern[0],
args.b)
print genomes[genomes.keys()[0]]
print genomes.keys(), ': Found matches at indices', matches.tolist()
