Bugfixes on seasonal and multivariate models

pyFTS/models/multivariate/common.py

@@ -3,7 +3,8 @@ import pandas as pd
 from pyFTS.common import FuzzySet
 
 def fuzzyfy_instance(data_point, var):
-    return FuzzySet.fuzzyfy(data_point, var.partitioner, mode='sets', method='fuzzy', alpha_cut=var.alpha_cut)
+    fsets = FuzzySet.fuzzyfy(data_point, var.partitioner, mode='sets', method='fuzzy', alpha_cut=var.alpha_cut)
+    return [(var.name, fs) for fs in fsets]
 
 
 

pyFTS/models/seasonal/common.py

@@ -88,10 +88,15 @@ class FuzzySet(FuzzySet.FuzzySet):
     """
 
     def __init__(self, datepart, name, mf, parameters, centroid, alpha=1.0, **kwargs):
-        super(FuzzySet, self).__init__(name, mf, parameters, centroid, alpha, type = 'datetime', **kwargs)
+        super(FuzzySet, self).__init__(name, mf, parameters, centroid, alpha,
+                                       type=kwargs.get('type', 'datetime'),
+                                       **kwargs)
         self.datepart = datepart
         self.type = 'seasonal'
 
     def membership(self, x):
-        dp = strip_datepart(x, self.datepart)
+        if self.type == 'datetime':
+            dp = strip_datepart(x, self.datepart)
+        else:
+            dp = x
         return self.mf(dp, self.parameters) * self.alpha

pyFTS/models/seasonal/partitioner.py

@@ -64,6 +64,17 @@ class TimeGridPartitioner(partitioner.Partitioner):
                                             **kwargs))
                     tmp.centroid = c
                     sets[set_name] = tmp
+                elif c == self.max - partlen:
+                    tmp = Composite(set_name, superset=True)
+                    tmp.append_set(FuzzySet(self.season, set_name, Membership.trimf,
+                                            [0.0000001, 0.0,
+                                             pl2], 0.0, alpha=.5,
+                                            **kwargs))
+                    tmp.append_set(FuzzySet(self.season, set_name, Membership.trimf,
+                                            [c - partlen, c, c + partlen], c,
+                                            **kwargs))
+                    tmp.centroid = c
+                    sets[set_name] = tmp
                 else:
                     sets[set_name] = FuzzySet(self.season, set_name, Membership.trimf,
                                          [c - partlen, c, c + partlen], c,

pyFTS/tests/general.py

@@ -16,21 +16,38 @@ from pyFTS.common import Transformations
 
 tdiff = Transformations.Differential(1)
 
-data =  pd.read_csv('/home/petronio/Downloads/priceHong').values
+dataset =  pd.read_csv('/home/petronio/Downloads/priceHong')
+dataset['hour'] = dataset.index.values % 24
 
 split = 24 * 800
-train = data[:split].flatten()
+#train = data[:split].flatten()
-test = data[split:].flatten()
+#test = data[split:].flatten()
 
-print(train)
+#print(train)
 
-fs_grid = Grid.GridPartitioner(data=train,npart=25)
+from pyFTS.models.multivariate import common, variable, mvfts
-#fs_entr.plot(ax[1])
+from pyFTS.partitioners import Grid
+from pyFTS.models.seasonal.common import DateTime
+from pyFTS.models.seasonal import partitioner as seasonal
 
-for method in [hofts.HighOrderFTS, pwfts.ProbabilisticWeightedFTS]:
+vhour = variable.Variable("Hour", data_label="hour", partitioner=seasonal.TimeGridPartitioner, npart=24,
-  for order in [2,3]:
+                            data=dataset,  partitioner_specific={'seasonality': DateTime.hour_of_day, 'type': 'common'})
-    model = method(partitioner=fs_grid, order=order)
+vprice = variable.Variable("Price", data_label="price", partitioner=Grid.GridPartitioner, npart=25,
-    model.fit(train)
+                            data=dataset)
+
+
+fig, ax = plt.subplots(nrows=2, ncols=1,figsize=[15,5])
+
+vhour.partitioner.plot(ax[0])
+vprice.partitioner.plot(ax[1])
+
+model = mvfts.MVFTS()
+#model.shortname += ' ' + key
+model.append_variable(vhour)
+model.append_variable(vprice)
+#  model.shortname +=  ' ' + w
+model.target_variable = vprice
+model.fit(dataset.iloc[:split])
 
 '''
 from pyFTS.data import TAIEX, SP500, NASDAQ