def checkRealtion7FinishedBy(epsilon, maxGap):
rh = RelationHandler(RelationHandler.RELATION_ALLEN_7)
logger.debug('Testing FINISH_BY in 7 relations, epsilon: ' + str(epsilon) +
', max gap: ' + str(maxGap))
sti1 = SymbolicTimeInterval(1, 5, 'A', 345)
sti2 = SymbolicTimeInterval(3, 5, 'B', 567)
checkRelation(rh, sti1, sti2, epsilon, maxGap,
AllenSevenRelationEngine.FINISHBY)
def checkRealtion7Overlaps(epsilon, maxGap):
rh = RelationHandler(RelationHandler.RELATION_ALLEN_7)
logger.debug('Testing OVERLAPS in 7 relations, epsilon: ' + str(epsilon) +
', max gap: ' + str(maxGap))
sti1 = SymbolicTimeInterval(1, 3, 'A', 345)
sti2 = SymbolicTimeInterval(2, 6, 'B', 567)
checkRelation(rh, sti1, sti2, epsilon, maxGap,
AllenSevenRelationEngine.OVERLAP)
def checkRealtion7Contains(epsilon, maxGap):
rh = RelationHandler(RelationHandler.RELATION_ALLEN_7)
logger.debug('Testing CONTAINS in 7 relations, epsilon: ' + str(epsilon) +
', max gap: ' + str(maxGap))
sti1 = SymbolicTimeInterval(1, 5, 'A', 345)
sti2 = SymbolicTimeInterval(2, 4, 'B', 567)
checkRelation(rh, sti1, sti2, epsilon, maxGap,
AllenSevenRelationEngine.CONTAIN)
def checkNotInGap(epsilon, maxGap):
rh = RelationHandler(RelationHandler.RELATION_ALLEN_7)
logger.debug('Testing Not in gap, epsilon: ' + str(epsilon) +
', max gap: ' + str(maxGap))
sti1 = SymbolicTimeInterval(1, 3, 'A', 345)
sti2 = SymbolicTimeInterval(3 + maxGap + 1, 3 + maxGap + 8, 'B', 567)
checkRelation(rh, sti1, sti2, epsilon, maxGap,
AllenSevenRelationEngine.NOT_DEFINED)
def test_is_pair_exist():
suportVec = 0.5
logger.debug('loading file for karma with support vec value: ' +
str(suportVec))
karma = Karma(suportVec)
karma.fit('../KarmaLego_TestsData/single_entity_2_bins.csv')
sti_a = SymbolicTimeInterval(0, 2, 1)
sti_b = SymbolicTimeInterval(4, 6, 2)
if karma.is_pair_exist(0, 1, sti_a, sti_b):
logger.info('[PASSED] pair exist ' + str(sti_a) + ' ' + str(sti_b))
else:
logger.info('[FAILED] pair dont exist ' + str(sti_a) + ' ' +
str(sti_b))
def parse_file_into_symbolic_time_intervals(self,
file_path,
num_comma,
start_index=0):
"""
parse the file to a map of entities to detect with their symbolic time intervals
sort the symbolic time intervals in a lexicographic order
:param file_path: string, the path to the symbolic time intervals file
:param start_index: int, the row fro mwhich to parse the symbolic time intervals
:return: nothing
"""
self._entities_map_to_detect = {}
with open(file_path) as f:
lines = f.readlines()[start_index:]
for i in range(0, len(lines), 2):
entity_id = int(lines[i].replace(";", ""))
sym_ti_ls = lines[i + 1].split(";")
for sym_ti in sym_ti_ls[:-1]:
parts = sym_ti.split(",")
st = int(parts[0])
et = int(parts[1])
symbol = parts[num_comma]
sym_ti_obj = SymbolicTimeInterval(start_time=st,
end_time=et,
symbol=symbol)
if entity_id not in self._entities_map_to_detect:
self._entities_map_to_detect[entity_id] = [sym_ti_obj]
else:
self._entities_map_to_detect[entity_id].append(sym_ti_obj)
def parse_file_into_symbolic_time_intervals_parts(self, file_path,
start_index, symbol_type,
num_comma):
"""
parse the file to a map of entities with their symbolic time intervals
remove relevant symbols
sort the symbolic time intervals in a lexicographic order
:param file_path: string, the path to the symbolic time intervals file
:param start_index: int, the row from which to parse the symbolic time intervals
:return: nothing
"""
symbols_map = {}
with open(file_path) as f:
lines = f.readlines()[start_index:]
for i in range(0, len(lines), 2):
entity_symbols = []
entity_id = int(lines[i].replace(";", ""))
sym_ti_ls = lines[i + 1].split(";")
for sym_ti in sym_ti_ls[:-1]:
parts = sym_ti.split(",")
st = int(parts[0])
et = int(parts[1])
symbol = parts[num_comma]
sym_ti_obj = SymbolicTimeInterval(start_time=st,
end_time=et,
symbol=symbol)
if entity_id not in self._entities_map:
self._entities_map[entity_id] = [sym_ti_obj]
else:
self._entities_map[entity_id].append(sym_ti_obj)
if symbol not in symbols_map:
symbols_map[symbol] = 1
entity_symbols.append(symbol)
elif symbol not in entity_symbols:
symbols_map[symbol] = symbols_map[symbol] + 1
entity_symbols.append(symbol)
num_of_entitites = len(self._entities_map)
unrelevent_symbols = [
sym for sym in symbols_map
if symbols_map[sym] < (num_of_entitites * self._min_ver_support)
]
self._frequent_symbols = [
sym for sym in symbols_map if sym not in unrelevent_symbols
]
if symbol_type == 'int':
for entity_id in self._entities_map:
self._entities_map[entity_id] = sorted([
sym_ti for sym_ti in self._entities_map[entity_id]
if sym_ti._symbol not in unrelevent_symbols
],
key=self.sort_str_int)
else:
for entity_id in self._entities_map:
self._entities_map[entity_id] = sorted(
[
sym_ti for sym_ti in self._entities_map[entity_id]
if sym_ti._symbol not in unrelevent_symbols
],
key=operator.attrgetter('_start_time', '_end_time',
'_symbol'))
def test_get_pairs():
suportVec = 0.5
logger.debug('loading file for karma with support vec value: ' +
str(suportVec))
karma = Karma(suportVec)
karma.fit('../KarmaLego_TestsData/single_entity_2_bins.csv')
sti = SymbolicTimeInterval(0, 2, 1)
pairs = karma.get_pairs(1, 2, 0, 1, sti)
if 2 == len(pairs):
logger.info('[PASSED] pairs fetched ' + str(pairs))
else:
logger.info('[FAILED] pairs not fetched - ' + str(pairs))