# HG changeset patch
# User Nicolas Saunier <nicolas.saunier@polymtl.ca>
# Date 1303169513 14400
# Node ID 7f1e54234f96cd2a7168a0e6b70eec672b0b1f57
# Parent  731df2fa00106395bbf6a1a6844313d6ed3affcd
added empirical discrete distribution, modified class organization and names

diff -r 731df2fa0010 -r 7f1e54234f96 python/utils.py
--- a/python/utils.py	Tue Mar 29 01:47:00 2011 -0400
+++ b/python/utils.py	Mon Apr 18 19:31:53 2011 -0400
@@ -23,7 +23,35 @@
         result += ((e-o)*(e-o))/e
     return result
 
-class empiricalDistribution:
+class EmpiricalDistribution:
+    def nSamples(self):
+        return sum(self.counts)
+
+
+class EmpiricalDiscreteDistribution(EmpiricalDistribution):
+    '''Class to represent a sample of a distribution for a discrete random variable
+    '''
+    def __init__(self, categories, counts):
+        self.categories = categories
+        self.counts = counts
+
+    def mean(self):
+        from numpy.core.fromnumeric import sum
+        result = [float(x*y) for x,y in zip(self.categories, self.counts)]
+        return sum(result)/self.nSamples()
+
+    def var(self, mean = None):
+        from numpy.core.fromnumeric import sum
+        if not mean:
+            m = self.mean()
+        else:
+            m = mean
+        result = 0.
+        squares = [float((x-m)*(x-m)*y) for x,y in zip(self.categories, self.counts)]
+        return sum(squares)/(self.nSamples()-1)
+
+
+class EmpiricalContinuousDistribution(EmpiricalDistribution):
     '''Class to represent a sample of a distribution for a continuous random variable
     with the number of observations for each interval
     intervals (categories variable) are defined by their left limits, the last one being the right limit
@@ -49,9 +77,6 @@
             result += self.counts[i]*(mid - m)*(mid - m)
         return result/(self.nSamples()-1)
 
-    def nSamples(self):
-        return sum(self.counts)
-
     def referenceCounts(self, cdf):
         '''cdf is a cumulative distribution function
         returning the probability of the variable being less that x'''