def require(self):
out = [ExtractInfoTask(self.graphs), GraphIndexTask()]
for g in self.graphs:
out.append(Optional(ExtractKernelBagTask(g, self.h, self.D)))
return out
def require(self):
task = []
if self.h > 0:
task.append(WLCollectorTask(self.graphs, self.h - 1, self.D))
for g in self.graphs:
task.append(Optional(PrepareKernelTask(g, self.h, self.D)))
return task
def require(self):
return [
Optional(
DatasetLabelTask(p, self.svcomp_name, self.directory, self.csv)
)
for p in self.paths
]
def require(self):
return [
Optional(
PescoGraphTask(graph)
)
for graph in self.paths
]
def require(self):
param_grid = {'h': self.h_Set, 'D': self.D_Set}
out = []
for params in ParameterGrid(param_grid):
out.append(
Optional(
CGridTask(self.graphs, self.train_index, params['h'],
params['D'])))
return out
def require(self):
loo = KFold(self.folds.value,
shuffle=True,
random_state=random.randint(0, 100))
out = []
for train_index, _ in loo.split(np.arange(len(self.graphs))):
out.append(
Optional(
hDGridTask(self.graphs, train_index.tolist(), self.h_Set,
self.D_Set)))
return out
def require(self):
index = np.array(self._index())
loo = KFold(self.k.value, shuffle=True, random_state=self.random_state.value)
return [
Optional(BagClassifierEvalutionTask(
self.clf_type,
self.clf_params,
self.h,
self.D,
self.scores,
train_index.tolist(),
test_index.tolist(),
self.category,
self.task_type,
self.kernel
))
for train_index, test_index in loo.split(index)
]
def require(self):
t = []
for category in self.categories.value:
t.append(Optional(CategoryLookupTask(category)))
return t
def require(self):
return [
Optional(MongoWLTask(g, self.h, self.maxDepth))
for g in self.graphs
]