Mercurial Hosting > traffic-intelligence

diff python/utils.py @ 372:349eb1e09f45

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Cleaned the methods/functions indicating if a point is in a polygon In general, shapely should be used, especially for lots of points: from shapely.geometry import Polygon, Point poly = Polygon(array([[0,0],[0,1],[1,1],[1,0]])) p = Point(0.5,0.5) poly.contains(p) -> returns True poly.contains(Point(-1,-1)) -> returns False You can convert a moving.Point to a shapely point: p = moving.Point(1,2) p.asShapely() returns the equivalent shapely point If you have several points to test, use moving.pointsInPolygon(points, polygon) where points are moving.Point and polygon is a shapely polygon.
author Nicolas Saunier <nicolas.saunier@polymtl.ca>
date Tue, 16 Jul 2013 17:00:17 -0400
parents 924e38c9f70e
children d0b86ed50f32
line wrap: on
line diff
--- a/python/utils.py	Tue Jul 16 01:36:50 2013 -0400
+++ b/python/utils.py	Tue Jul 16 17:00:17 2013 -0400
@@ -218,13 +218,14 @@
         self.lengthFunc = lengthFunc
         self.alignmentShift = 0
 
-    def similarities(self, l1, l2):
+    def similarities(self, l1, l2, shift=0):
         from numpy import zeros, int as npint
         n1 = len(l1)
         n2 = len(l2)
         self.similarityTable = zeros((n1+1,n2+1), dtype = npint)
         for i in xrange(1,n1+1):
-            for j in xrange(max(1,i-self.delta),min(n2+1,i+self.delta+1)):
+            for j in xrange(max(1,i-shift-self.delta),min(n2+1,i-shift+self.delta+1)):
+                #print max(1,i-shift-self.delta),min(n2+1,i-shift+self.delta+1)
                 if self.similarityFunc(l1[i-1], l2[j-1]):
                     self.similarityTable[i,j] = self.similarityTable[i-1,j-1]+1
                 else:
@@ -249,51 +250,61 @@
 
         eg distance(p1, p2) < epsilon
         '''
-        from numpy import argmax
+        #from numpy import argmax
         self.similarities(l1, l2)
-        self.similarityTable = self.similarityTable[:, :min(len(l2), len(l1)+self.delta)+1]
+        #self.similarityTable = self.similarityTable[:, :min(len(l2), len(l1)+self.delta)+1]
         if computeSubSequence:
             self.subSequenceIndices = self.subSequence(len(l1), len(l2))
-        return self.similarityTable[-1,-1]
+        return self.similarityTable.max()
 
     def _compute(self, _l1, _l2, computeSubSequence = False):
         '''returns the best matching if using a finite delta by shiftinig the series alignments'''
+        if len(_l2) < len(_l1): # l1 is the shortest
+            l1 = _l2
+            l2 = _l1
+            revertIndices = True
+        else:
+            l1 = _l1
+            l2 = _l2
+            revertIndices = False
+        n1 = len(l1)
+        n2 = len(l2)
+
         if self.aligned:
-            from numpy import argmax
-            if len(_l2) < len(_l1): # l1 is the shortest
-                l1 = _l2
-                l2 = _l1
-                revertIndices = True
-            else:
-                l1 = _l1
-                l2 = _l2
-                revertIndices = False
-            n1 = len(l1)
-            n2 = len(l2)
             # for i in xrange(min(delta,n1), max(n1+n2-delta, n2+1)): # i is the alignment of the end of l1 in l2
             #     print l1[min(-i-1,n1):] # min(n1+n2-i,n1)
             #     print l2[max(0,i-n1):]
             #     print LCSS(l1[min(-i-1,n1):], l2[max(0,i-n1):], similarityFunc, delta)
             lcssValues = {}
             similarityTables = {}
-            for i in xrange(min(self.delta,n1), max(n1+n2-self.delta, n2+1)):
+            #for i in xrange(min(self.delta,n1), max(n1+n2-self.delta, n2+1)):
+            for i in xrange(-max(n1,n2)-self.delta, +max(n1,n2)+self.delta):
                 #print l1[min(-i-1,n1):] # min(n1+n2-i,n1)
                 #print l2[max(0,i-n1):]
-                lcssValues[i] = self.computeLCSS(l1[min(-i-1,n1):], l2[max(0,i-n1):])
+                #lcssValues[i] = self.computeLCSS(l1[min(-i-1,n1):], l2[max(0,i-n1):])
                 #print i, lcssValues[i]
+                lcssValues[i] = self.similarities(l1, l2, i)
                 similarityTables[i] = self.similarityTable
-            imax = argMaxDict(lcssValues)
-            self.similarityTable = similarityTables[imax]
-            self.subSequenceIndices = self.subSequence(self.similarityTable.shape[0]-1, self.similarityTable.shape[1]-1) 
+                #print i
+                print self.similarityTable
+            self.alignmentShift = argMaxDict(lcssValues)
+            self.similarityTable = similarityTables[self.alignmentShift]
+            # do the subsequence computation here, once similarityTable is set
+            #self.subSequenceIndices = self.subSequence(self.similarityTable.shape[0]-1, self.similarityTable.shape[1]-1) 
+            #lcss = lcssValues[imax]
+        else:
+            self.alignmentShift = 0
+            self.similarities(l1, l2)
+        self.similarityTable = self.similarityTable[:, :min(len(l2), len(l1)+self.delta)+1]
+        if computeSubSequence:
+            self.subSequenceIndices = self.subSequence(self.similarityTable.shape[0]-1, self.similarityTable.shape[1]-1)
             if revertIndices:
-                self.subSequenceIndices = [(j+imax-n1,i) for i,j in self.subSequenceIndices]
-                self.alignmentShift = imax-n1
+                self.subSequenceIndices = [(j+self.alignmentShift,i) for i,j in self.subSequenceIndices]
+            #self.alignmentShift = imax-n1
             else:
-                self.subSequenceIndices = [(i+n1-imax,j) for i,j in self.subSequenceIndices]
-                self.alignmentShift = n1-imax
-            return lcssValues[imax]
-        else:
-            return self.computeLCSS(_l1, _l2, computeSubSequence)
+                self.subSequenceIndices = [(i+self.alignmentShift,j) for i,j in self.subSequenceIndices]
+            #self.alignmentShift = n1-imax
+        return self.similarityTable[-1,-1]
 
     def compute(self, l1, l2, computeSubSequence = False):
         '''get methods are to be shadowed in child classes '''