- Several bugfixes in benchmarks methods and optimizations

benchmarks/arima.py

@@ -11,7 +11,7 @@ class ARIMA(fts.FTS):
     Façade for statsmodels.tsa.arima_model
     """
     def __init__(self, name, **kwargs):
-        super(ARIMA, self).__init__(1, "ARIMA")
+        super(ARIMA, self).__init__(1, "ARIMA"+name)
         self.name = "ARIMA"
         self.detail = "Auto Regressive Integrated Moving Average"
         self.is_high_order = True

benchmarks/distributed_benchmarks.py

@@ -8,6 +8,7 @@ python3 /usr/local/bin/dispynode.py -i [local IP] -d
 
 import datetime
 import time
+from copy import deepcopy
 
 import dispy
 import dispy.httpd
@@ -136,8 +137,9 @@ def point_sliding_window(data, windowsize, train=0.8, models=None, partitioners=
             pool.append(mfts)
 
     for count, model in enumerate(benchmark_models, start=0):
-        mfts = model("")
-        mfts.order = benchmark_models_parameters[count]
+        par = benchmark_models_parameters[count]
+        mfts = model(str(par if par is not None else ""))
+        mfts.order = par
         pool.append(mfts)
 
     experiments = 0

benchmarks/naive.py

@@ -7,7 +7,7 @@ from pyFTS import fts
 class Naive(fts.FTS):
     """Naïve Forecasting method"""
     def __init__(self, name, **kwargs):
-        super(Naive, self).__init__(1, "Naive " + name)
+        super(Naive, self).__init__(1, "Naive")
         self.name = "Naïve Model"
         self.detail = "Naïve Model"
         self.benchmark_only = True

benchmarks/quantreg.py

@@ -10,7 +10,7 @@ from pyFTS import fts
 class QuantileRegression(fts.FTS):
     """Façade for statsmodels.regression.quantile_regression"""
     def __init__(self, name, **kwargs):
-        super(QuantileRegression, self).__init__(1, "QR")
+        super(QuantileRegression, self).__init__(1, "QR"+name)
         self.name = "QR"
         self.detail = "Quantile Regression"
         self.is_high_order = True
@@ -43,8 +43,11 @@ class QuantileRegression(fts.FTS):
             self.upper_qt = [uqt.params[k] for k in uqt.params.keys()]
             self.lower_qt = [lqt.params[k] for k in lqt.params.keys()]
 
+        self.shortname = "QAR(" + str(self.order) + ")"
+
     def linearmodel(self,data,params):
-        return params[0] + sum([ data[k] * params[k+1] for k in np.arange(0, self.order) ])
+        #return params[0] + sum([ data[k] * params[k+1] for k in np.arange(0, self.order) ])
+        return sum([data[k] * params[k] for k in np.arange(0, self.order)])
 
     def forecast(self, data, **kwargs):
         ndata = np.array(self.doTransformations(data))

tests/general.py

@@ -57,15 +57,15 @@ from statsmodels.tsa.tsatools import lagmat
 from pyFTS.benchmarks import distributed_benchmarks as bchmk
 #from pyFTS.benchmarks import parallel_benchmarks as bchmk
 from pyFTS.benchmarks import Util
-#from pyFTS.benchmarks import arima
+from pyFTS.benchmarks import arima, quantreg
 
 #Util.cast_dataframe_to_sintetic_point("experiments/taiex_point_analitic.csv","experiments/taiex_point_sintetic.csv",11)
 
 #Util.plot_dataframe_point("experiments/taiex_point_sintetic.csv","experiments/taiex_point_analitic.csv",11)
 
 #tmp = arima.ARIMA("")
-#tmp.train(taiex[:1600],None,parameters=(2,0,1))
-#teste = tmp.forecast(taiex[1598:1601])
+#tmp.train(taiex[:1600],(2,1,1))
+#teste = tmp.forecast(taiex[1600:2000])
 
 #print(teste)