args = utils.config()
print(args)
experiment_name = utils.get_experiment_name()
utils.make_directory("../logs")
logging.basicConfig(filename=os.path.join(
'../logs/{}.log'.format(experiment_name)),
level=logging.INFO,
format='%(asctime)s %(name)s %(levelname)s %(message)s')
logger = logging.getLogger(__name__)
utils.print_log('START with Configuration : {}'.format(args))
data_path = utils.make_directory(args.data_path)
partitions, partitions_train, partitions_tune = utils.get_partitions()
if args.dataset == "Cars3D":
train_data = Cars3D(root=args.data_path,
mode="train",
image_size=args.image_size)
query_data = Cars3D(root=args.data_path,
mode="query",
image_size=args.image_size)
gallery_data = Cars3D(root=args.data_path,
mode="gallery",
image_size=args.image_size)
else:
raise ValueError("Provided dataset does not exist")
if args.sampling_method == "triplet":
from z_algo import zbox_search
from ed import levenshtein_distance_lt_1
from utils import get_partitions
if __name__ == '__main__':
text = "abdyabxdcyabcdz"
pattern = "abcd"
pattern_length = len(pattern)
first_half, second_half = get_partitions(pattern, 2)
result = []
i, j = first_half
ps = pattern[i:j]
ps2 = pattern[j:pattern_length]
hits = zbox_search(text, ps)
print("first half hits:", hits)
for h in hits:
print("hit:", h)
# check for substitution
print("match substitute")
print(ps2, text[h + j:h + pattern_length])
res, d = levenshtein_distance_lt_1(ps2, text[h + j:h + pattern_length])
if res:
result.append((h, d))
# check for deletion
print(ps2, text[h + j:h + pattern_length + 1])
res, d = levenshtein_distance_lt_1(ps2,
text[h + j:h + pattern_length + 1])
if res:
result.append((h, 1))
# check for insertion
np.random.seed(1)
filename = "%s%s.csv" % (data_dir, dataset)
print("Loading features...")
data, labels = data_load(filename)
posind = np.where(labels == 1)[0] # all known miRNAs
print("Done.")
startTime = time.time()
trainTime = []
testTime = []
for fold in range(nfolds):
print("Training deeSOM with %s (fold %d of %d)" %
(dataset, fold + 1, nfolds))
train_ind, test_ind = get_partitions(partitions_dir, dataset, len(labels),
fold)
startFoldTime = time.time()
# Train
labels[posind] = 1 # Starting with all known genome-wide positives...
print("original pos %d" % sum(labels))
# ...remove the positive labels that will be used for testing
labels[test_ind] = 0
print("train pos %d" % sum(labels))
toc = time.time()
# Train
deesom = DeeSOM(verbosity=1)
deesom.fit(data, labels)
from z_algo import zbox_search
from utils import get_partitions
if __name__ == '__main__':
text = "bbcaefadcabcpqr"
pattern = "abc"
pattern_length = len(pattern)
partitions = get_partitions(pattern, 2)
result = []
for i, j in partitions:
ps = pattern[i:j]
hits = zbox_search(text, ps)
for h in hits:
mismatch = 0
# verify string segment before partition
if i > 0:
for k in range(i):
if text[h - i + k] != pattern[k]:
mismatch += 1
# verify string segment after partition
if j < pattern_length:
for k in range(j, pattern_length):
if pattern[k] != text[h + k]:
mismatch += 1
if mismatch < 2:
result.append((h - i, mismatch))
result = list(set(result))
result.sort(key=lambda x: x[0])
for i, mis in result:
print(i + 1, mis)