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

bugfixes for seasonal FTS

Browse Source
This commit is contained in:
Petrônio Cândido 2018-06-22 23:40:19 -03:00
parent f9d6538bae
commit 88d22a4482
7 changed files with 48 additions and 54 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
pyFTS/common/fts.py
View File

@ -37,7 +37,7 @@ class FTS(object):
self.sets = self.partitioner.sets
self.auto_update = False
self.benchmark_only = False
self.indexer = None
self.indexer = kwargs.get("indexer", None)
self.uod_clip = kwargs.get("uod_clip", True)
def fuzzy(self, data):

4
pyFTS/models/chen.py
View File

@ -22,12 +22,12 @@ class ConventionalFLRG(flrg.FLRG):
self.RHS.add(c)
def __str__(self):
tmp = self.LHS + " -> "
tmp = str(self.LHS) + " -> "
tmp2 = ""
for c in sorted(self.RHS, key=lambda s: s):
if len(tmp2) > 0:
tmp2 = tmp2 + ","
tmp2 = tmp2 + c
tmp2 = tmp2 + str(c)
return tmp + tmp2

24
pyFTS/models/seasonal/cmsfts.py
View File

@ -13,11 +13,11 @@ class ContextualSeasonalFLRG(sfts.SeasonalFLRG):
self.RHS = {}
def append_rhs(self, flr, **kwargs):
if flr.LHS in self.RHS:
self.RHS[flr.LHS].append_rhs(flr.RHS)
if flr.LHS.name in self.RHS:
self.RHS[flr.LHS.name].append_rhs(flr.RHS.name)
else:
self.RHS[flr.LHS] = chen.ConventionalFLRG(flr.LHS)
self.RHS[flr.LHS].append_rhs(flr.RHS)
self.RHS[flr.LHS.name] = chen.ConventionalFLRG(flr.LHS.name)
self.RHS[flr.LHS.name].append_rhs(flr.RHS.name)
def __str__(self):
tmp = str(self.LHS) + ": \n "
@ -31,17 +31,17 @@ class ContextualMultiSeasonalFTS(sfts.SeasonalFTS):
"""
Contextual Multi-Seasonal Fuzzy Time Series
"""
def __init__(self, name, indexer, **kwargs):
super(ContextualMultiSeasonalFTS, self).__init__("CMSFTS")
def __init__(self, **kwargs):
super(ContextualMultiSeasonalFTS, self).__init__(**kwargs)
self.name = "Contextual Multi Seasonal FTS"
self.shortname = "CMSFTS " + name
self.shortname = "CMSFTS "
self.detail = ""
self.seasonality = 1
self.has_seasonality = True
self.has_point_forecasting = True
self.is_high_order = True
self.is_multivariate = True
self.indexer = indexer
self.order = 1
self.flrgs = {}
def generate_flrg(self, flrs):
@ -61,11 +61,11 @@ class ContextualMultiSeasonalFTS(sfts.SeasonalFTS):
self.generate_flrg(flrs)
def get_midpoints(self, flrg, data):
if data in flrg.flrgs:
ret = np.array([self.sets[s].centroid for s in flrg.flrgs[data.name].RHS])
if data.name in flrg.RHS:
ret = np.array([self.sets[s].centroid for s in flrg.RHS[data.name].RHS])
return ret
else:
return np.array([self.sets[data].centroid])
return np.array([self.sets[data.name].centroid])
def forecast(self, data, **kwargs):
ordered_sets = FuzzySet.set_ordered(self.sets)
@ -97,3 +97,5 @@ class ContextualMultiSeasonalFTS(sfts.SeasonalFTS):
ret.append(sum(mp) / len(mp))
return ret

25
pyFTS/models/seasonal/sfts.py
View File

@ -7,13 +7,13 @@ S.-M. Chen, “Forecasting enrollments based on fuzzy time series,” Fuzzy Sets
"""
import numpy as np
from pyFTS.common import FuzzySet, FLR, fts
from pyFTS.common import FuzzySet, FLR, flrg, fts
class SeasonalFLRG(FLR.FLRG):
class SeasonalFLRG(flrg.FLRG):
"""First Order Seasonal Fuzzy Logical Relationship Group"""
def __init__(self, seasonality):
super(SeasonalFLRG, self).__init__(None,None)
super(SeasonalFLRG, self).__init__(1)
self.LHS = seasonality
self.RHS = []
@ -26,10 +26,10 @@ class SeasonalFLRG(FLR.FLRG):
def __str__(self):
tmp = str(self.LHS) + " -> "
tmp2 = ""
for c in sorted(self.RHS, key=lambda s: s.name):
for c in sorted(self.RHS, key=lambda s: str(s)):
if len(tmp2) > 0:
tmp2 = tmp2 + ","
tmp2 = tmp2 + c.name
tmp2 = tmp2 + str(c)
return tmp + tmp2
def __len__(self):
@ -38,10 +38,11 @@ class SeasonalFLRG(FLR.FLRG):
class SeasonalFTS(fts.FTS):
"""First Order Seasonal Fuzzy Time Series"""
def __init__(self, name, **kwargs):
super(SeasonalFTS, self).__init__(1, "SFTS", **kwargs)
def __init__(self, **kwargs):
super(SeasonalFTS, self).__init__(**kwargs)
self.name = "Seasonal FTS"
self.detail = "Chen"
self.shortname = "SFTS"
self.order = 1
self.seasonality = 1
self.has_seasonality = True
self.has_point_forecasting = True
@ -62,11 +63,15 @@ class SeasonalFTS(fts.FTS):
#print(season)
self.flrgs[ss].append_rhs(flr.RHS)
def get_midpoints(self, flrg):
ret = np.array([self.sets[s].centroid for s in flrg.RHS])
return ret
def train(self, data, **kwargs):
if kwargs.get('sets', None) is not None:
self.sets = kwargs.get('sets', None)
tmpdata = FuzzySet.fuzzyfy_series_old(data, self.sets)
flrs = FLR.generate_recurrent_flrs(tmpdata)
flrs = FLR.generate_non_recurrent_flrs(tmpdata)
self.generate_flrg(flrs)
def forecast(self, data, **kwargs):
@ -81,7 +86,7 @@ class SeasonalFTS(fts.FTS):
flrg = self.flrgs[str(season)]
mp = self.getMidpoints(flrg)
mp = self.get_midpoints(flrg)
ret.append(np.percentile(mp, 50))

41
pyFTS/tests/sfts.py
View File

@ -8,43 +8,30 @@ import pandas as pd
from pyFTS.common import Util
from pyFTS.benchmarks import benchmarks as bchmk
os.chdir("/home/petronio/dados/Dropbox/Doutorado/Codigos/")
os.chdir("/home/petronio/Downloads")
sonda = pd.read_csv("DataSets/SONDA_BSB_MOD.csv", sep=";")
data = pd.read_csv("dress_data.csv", sep=",")
sonda['data'] = pd.to_datetime(sonda['data'])
#sonda['data'] = pd.to_datetime(sonda['data'])
sonda = sonda[:][527041:]
#data.index = np.arange(0,len(data.index))
sonda.index = np.arange(0,len(sonda.index))
data = data["a"].tolist()
sonda_treino = sonda[:1051200]
sonda_teste = sonda[1051201:]
from pyFTS.models.seasonal import sfts, cmsfts, SeasonalIndexer
ix = SeasonalIndexer.LinearSeasonalIndexer([7],[1])
#res = bchmk.simpleSearch_RMSE(sonda_treino, sonda_teste,
# sfts.SeasonalFTS,np.arange(3,30),[1],parameters=1440,
# tam=[15,8], plotforecasts=False,elev=45, azim=40,
# save=False,file="pictures/sonda_sfts_error_surface", intervals=False)
from pyFTS.partitioners import Grid
partitions = ['grid','entropy']
fs = Grid.GridPartitioner(data=data,npart=10)
indexers = ['m15','Mh','Mhm15']
model = sfts.SeasonalFTS(indexer=ix, partitioner=fs)
#model = cmsfts.ContextualMultiSeasonalFTS(indexer=ix, partitioner=fs)
models = []
ixs = []
model.fit(data)
sample = sonda_teste[0:4300]
print(model)
for max_part in [10, 20, 30, 40, 50]:
for part in partitions:
for ind in indexers:
ix = Util.load_obj("models/sonda_ix_" + ind + ".pkl")
model = Util.load_obj("models/sonda_msfts_" + part + "_" + str(max_part) + "_" + ind + ".pkl")
model.shortname = part + "_" + str(max_part) + "_" + ind
models.append(model)
ixs.append(ix)
print(bchmk.print_point_statistics(sample, models, indexers=ixs))
print(model.predict(data))