Interval forecasting on MVFTS and WMVFTS

pyFTS/models/multivariate/flrg.py

@@ -22,7 +22,6 @@ class FLRG(flg.FLRG):
                 self.LHS[var] = []
             self.LHS[var].append(fset)
 
-
     def append_rhs(self, fset, **kwargs):
         self.RHS.add(fset)
 
@@ -34,10 +33,22 @@ class FLRG(flg.FLRG):
 
         return np.nanmin(mvs)
 
+    def get_lower(self, sets):
+        if self.lower is None:
+            self.lower = min([sets[rhs].lower for rhs in self.RHS])
+
+        return self.lower
+
+    def get_upper(self, sets):
+        if self.upper is None:
+            self.upper = max([sets[rhs].upper for rhs in self.RHS])
+
+        return self.upper
+
     def __str__(self):
         _str = ""
         for k in self.RHS:
             _str += "," if len(_str) > 0 else ""
             _str += k
 
-        return self.get_key() + " -> " + _str
+        return self.get_key() + " -> " + _str

pyFTS/models/multivariate/mvfts.py

@@ -236,8 +236,8 @@ class MVFTS(fts.FTS):
                     los.append(_flrg.get_lower(self.target_variable.partitioner.sets))
 
             mv = np.array(mvs)
-            up = np.dot(mv, np.array(ups).T) / np.sum(mv)
-            lo = np.dot(mv, np.array(los).T) / np.sum(mv)
+            up = np.dot(mv, np.array(ups).T) / np.nansum(mv)
+            lo = np.dot(mv, np.array(los).T) / np.nansum(mv)
 
             ret.append([lo, up])
 

pyFTS/tests/multivariate.py

@@ -85,6 +85,7 @@ print(_s2-_s1)
 #model.fit(data, distributed='dispy', nodes=['192.168.0.110'])
 #'''
 
+'''
 from pyFTS.models.multivariate import common, variable, mvfts, wmvfts, cmvfts, grid
 from pyFTS.models.seasonal import partitioner as seasonal
 from pyFTS.models.seasonal.common import DateTime
@@ -99,7 +100,7 @@ from pyFTS.models.multivariate import common, variable, mvfts
 from pyFTS.models.seasonal import partitioner as seasonal
 from pyFTS.models.seasonal.common import DateTime
 
-#'''
+
 sp = {'seasonality': DateTime.minute_of_day, 'names': [str(k) for k in range(0,24)]}
 
 vhour = variable.Variable("Hour", data_label="date", partitioner=seasonal.TimeGridPartitioner, npart=24,
@@ -152,4 +153,38 @@ time_generator = lambda x : pd.to_datetime(x) + pd.to_timedelta(1, unit='h')
 
 
 forecasts = mload.predict(test_mv.iloc[:1], steps_ahead=48, generators={'date': time_generator,
-                                                                        'temperature': mtemp})
+                                                                        'temperature': mtemp})
+
+'''
+
+
+data = pd.read_csv('https://query.data.world/s/6xfb5useuotbbgpsnm5b2l3wzhvw2i', sep=';')
+
+train = data.iloc[:9000]
+test = data.iloc[9000:9200]
+
+from pyFTS.models.multivariate import common, variable, mvfts
+from pyFTS.models.seasonal import partitioner as seasonal
+from pyFTS.models.seasonal.common import DateTime
+from pyFTS.partitioners import Grid
+
+sp = {'seasonality': DateTime.day_of_year , 'names': ['Jan','Fev','Mar','Abr','Mai','Jun','Jul', 'Ago','Set','Out','Nov','Dez']}
+
+vmonth = variable.Variable("Month", data_label="data", partitioner=seasonal.TimeGridPartitioner, npart=12,
+                           data=train, partitioner_specific=sp)
+
+sp = {'seasonality': DateTime.minute_of_day, 'names': [str(k) for k in range(0,24)]}
+
+vhour = variable.Variable("Hour", data_label="data", partitioner=seasonal.TimeGridPartitioner, npart=24,
+                          data=train, partitioner_specific=sp)
+
+vavg = variable.Variable("Radiation", data_label="glo_avg", alias='rad',
+                         partitioner=Grid.GridPartitioner, npart=30, alpha_cut=.3,
+                         data=train)
+
+from pyFTS.models.multivariate import mvfts, wmvfts, cmvfts
+
+model = wmvfts.WeightedMVFTS(explanatory_variables=[vmonth, vhour, vavg], target_variable=vavg)
+model.fit(train)
+
+forecasts = model.predict(test, type='interval')