From 01cdf092eb52630aede29bf550a61d294975c99b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Petr=C3=B4nio=20C=C3=A2ndido=20de=20Lima=20e=20Silva?=
 <petronio.candido@gmail.com>
Date: Thu, 20 Oct 2016 11:57:59 -0200
Subject: [PATCH] =?UTF-8?q?IFTS=20implementa=C3=A7=C3=A3o=20inicial=20dife?=
 =?UTF-8?q?rente=20da=20rfts,=20com=20suporte=20=C3=A0=20high=20order?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 ifts.py | 103 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 103 insertions(+)
 create mode 100644 ifts.py

diff --git a/ifts.py b/ifts.py
new file mode 100644
index 0000000..f116bb1
--- /dev/null
+++ b/ifts.py
@@ -0,0 +1,103 @@
+import numpy as np
+from pyFTS import *
+
+
+class IntervalFTS(hofts.HighOrderFTS):
+	def __init__(self,name):
+		super(IntervalFTS, self).__init__(name)
+		self.flrgs = {}
+    
+	def getUpper(self,flrg):
+		ret = np.array([s.upper for s in flrg.RHS])
+		return ret
+		
+	def getLower(self,flrg):
+		ret = np.array([s.lower for s in flrg.RHS])
+		return ret
+		
+	def getSequenceMembership(self, data, fuzzySets):
+		mb = [ fuzzySets[k].membership( data[k] )  for k in np.arange(0,len(data))  ]
+		return mb
+    	
+	def forecast(self,data):
+		
+		ndata = np.array(data)
+		
+		l = len(ndata)
+		
+		ret = []
+		
+		for k in np.arange(self.order+1,l):
+			
+			up = []
+			lo = []
+			
+			# Achar os conjuntos que tem pert > 0 para cada lag
+			count = 0
+			lags = {}
+			combinations = 1
+			for instance in ndata[k-self.order : k]:
+				mb = common.fuzzyInstance(instance, self.sets)
+				tmp = np.argwhere( mb )
+				idx = np.ravel(tmp) #flatten 
+				
+				lag = {}
+				lag["id"] = count
+				lag["sets"] = idx
+				lag["memberships"] = [mb[ k ] for k in idx]
+				lag["count"] = len(idx)
+				lags[count] = lag
+				combinations = combinations * lag["count"]
+				count = count + 1				
+				
+				print(combinations)
+				
+				
+			# Build a tree exploring all possibilities
+			
+			# Trace each path from leaf to roots and reverse path
+			
+			# -------
+				
+			#return lags
+				
+			wflrgs = {}
+			
+			# Gerar as permutações possíveis e as FLRG's correspondentes
+			lag_inc = [0 for k in np.arange(0,self.order) ]
+			isComplete = False
+			while (isComplete):
+				flrg = hofts.HighOrderFLRG(self.order)
+				
+				flrg.appendLHS( self.sets[ lag_inc[0]  ] )
+				
+				for lag_count in np.arange(1,self.order):
+					if lag_count > 1: lag_inc[ lag_count - 1 ] = 0
+					
+#					for 
+					
+				#lag_count = lag_count + 1
+			
+			# Achar a pert geral de cada FLRG
+			
+			# achar o os bounds de cada FLRG
+			
+			# gerar o intervalo
+			
+#			tmpdata = common.fuzzySeries(XX,self.sets)
+#			tmpflrg = HighOrderFLRG(self.order)
+		
+#			for s in tmpdata: tmpflrg.appendLHS(s)
+			
+#			if tmpflrg.strLHS() not in self.flrgs:
+#				ret.append(tmpdata[-1].centroid)
+#			else:
+#				flrg = self.flrgs[tmpflrg.strLHS()]
+#				mp = self.getMidpoints(flrg)
+				
+#				ret.append(sum(mp)/len(mp))
+			
+#		return ret
+		
+	
+