sam/pyFTS
1
0
Fork 0
Code Issues 3 Pull Requests Actions Packages Projects Releases Wiki Activity

Bugfixes in High Order NSFTS

Browse Source
This commit is contained in:
Petrônio Cândido 2019-11-27 08:59:40 -03:00
parent 2b214e561b
commit 9c9553e71e
2 changed files with 26 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

39
pyFTS/models/nonstationary/honsfts.py
View File

@ -9,7 +9,6 @@ class HighOrderNonStationaryFLRG(flrg.NonStationaryFLRG):
"""First Order NonStationary Fuzzy Logical Relationship Group""" """First Order NonStationary Fuzzy Logical Relationship Group"""
def __init__(self, order, **kwargs): def __init__(self, order, **kwargs):
super(HighOrderNonStationaryFLRG, self).__init__(order, **kwargs) super(HighOrderNonStationaryFLRG, self).__init__(order, **kwargs)
self.LHS = [] self.LHS = []
self.RHS = {} self.RHS = {}
self.count = 0.0 self.count = 0.0
@ -101,10 +100,10 @@ class HighOrderNonStationaryFTS(nsfts.NonStationaryFTS):
def generate_flrg(self, data, **kwargs): def generate_flrg(self, data, **kwargs):
l = len(data) l = len(data)
for k in np.arange(self.order, l): for k in np.arange(self.max_lag, l):
if self.dump: print("FLR: " + str(k)) if self.dump: print("FLR: " + str(k))
sample = data[k - self.order: k] sample = data[k - self.max_lag: k]
rhs = [key for key in self.partitioner.ordered_sets rhs = [key for key in self.partitioner.ordered_sets
if self.partitioner.sets[key].membership(data[k], [0,1]) > 0.0] if self.partitioner.sets[key].membership(data[k], [0,1]) > 0.0]
@ -145,18 +144,17 @@ class HighOrderNonStationaryFTS(nsfts.NonStationaryFTS):
lags = [] lags = []
for ct, dat in enumerate(sample): for ct, dat in enumerate(sample):
affected_sets = [[ct, self.partitioner.sets[key].membership(dat, perturb[ct])] affected_sets = [key for ct, key in enumerate(self.partitioner.ordered_sets)
for ct, key in enumerate(self.partitioner.ordered_sets)
if self.partitioner.sets[key].membership(dat, perturb[ct]) > 0.0] if self.partitioner.sets[key].membership(dat, perturb[ct]) > 0.0]
if len(affected_sets) == 0: if len(affected_sets) == 0:
if dat < self.partitioner.lower_set().get_lower(perturb[0]): if dat < self.partitioner.lower_set().get_lower(perturb[0]):
affected_sets.append([0, 1]) affected_sets.append(self.partitioner.lower_set().name)
elif dat > self.partitioner.upper_set().get_upper(perturb[-1]): elif dat > self.partitioner.upper_set().get_upper(perturb[-1]):
affected_sets.append([self.partitioner.partitions - 1, 1]) affected_sets.append(self.partitioner.upper_set().name)
lags.append([a[0] for a in affected_sets]) lags.append(affected_sets)
# Build the tree with all possible paths # Build the tree with all possible paths
@ -166,7 +164,7 @@ class HighOrderNonStationaryFTS(nsfts.NonStationaryFTS):
flrg = HighOrderNonStationaryFLRG(self.order) flrg = HighOrderNonStationaryFLRG(self.order)
for kk in path: for kk in path:
flrg.append_lhs(self.partitioner.ordered_sets[kk]) flrg.append_lhs(kk)
affected_flrgs.append(flrg) affected_flrgs.append(flrg)
mv = [] mv = []
@ -185,16 +183,12 @@ class HighOrderNonStationaryFTS(nsfts.NonStationaryFTS):
explain = kwargs.get('explain', False) explain = kwargs.get('explain', False)
fuzzyfied = kwargs.get('fuzzyfied', False)
time_displacement = kwargs.get("time_displacement", 0) time_displacement = kwargs.get("time_displacement", 0)
window_size = kwargs.get("window_size", 1) window_size = kwargs.get("window_size", 1)
no_update = kwargs.get("no_update", False) no_update = kwargs.get("no_update", False)
mode = kwargs.get('mode', 'mean')
ret = [] ret = []
l = len(ndata) if not explain else self.max_lag + 1 l = len(ndata) if not explain else self.max_lag + 1
@ -214,8 +208,7 @@ class HighOrderNonStationaryFTS(nsfts.NonStationaryFTS):
if no_update: if no_update:
perturb = [[0, 1] for k in np.arange(self.partitioner.partitions)] perturb = [[0, 1] for k in np.arange(self.partitioner.partitions)]
else: else:
perturb = self.conditional_perturbation_factors(sample[0]) perturb = self.conditional_perturbation_factors(sample[-1])
affected_flrgs, affected_flrgs_memberships = self._affected_flrgs(sample, perturb) affected_flrgs, affected_flrgs_memberships = self._affected_flrgs(sample, perturb)
@ -246,4 +239,20 @@ class HighOrderNonStationaryFTS(nsfts.NonStationaryFTS):
ret.append(pto) ret.append(pto)
if self.method == 'conditional' and not no_update:
self.forecasts.append(pto)
self.residuals.append(self.inputs[-1] - self.forecasts[-1])
self.inputs.extend(sample)
for g in range(self.order):
self.inputs.pop(0)
self.forecasts.pop(0)
self.residuals.pop(0)
return ret return ret
def __str__(self):
tmp = self.name + ":\n"
for r in self.flrgs:
tmp = "{0}{1}\n".format(tmp, str(self.flrgs[r]))
return tmp

3
pyFTS/tests/nonstationary.py
View File

@ -33,5 +33,6 @@ print(fs)
model = honsfts.HighOrderNonStationaryFTS(partitioner=fs, order=2) model = honsfts.HighOrderNonStationaryFTS(partitioner=fs, order=2)
#model = nsfts.NonStationaryFTS(partitioner=fs) #model = nsfts.NonStationaryFTS(partitioner=fs)
model.fit(train) model.fit(train)
#print(model)
forecasts = model.predict(test) forecasts = model.predict(test)
print(forecasts) #print(forecasts)