X-Git-Url: https://git.njae.me.uk/?a=blobdiff_plain;f=segment.py;h=dd0b2a8347ee800c4addf996f369ea0293b47bb7;hb=2f33e16ccc84ddb0023f3621dd6ad545c1bb3251;hp=712895b6b0d7f1563ee4149fe7d94445a3931233;hpb=abcf8d713890f19f2dbf066ba92a0e679432d351;p=cipher-tools.git

diff --git a/segment.py b/segment.py
index 712895b..dd0b2a8 100644
--- a/segment.py
+++ b/segment.py
@@ -1,4 +1,3 @@
-# import re, string, random, glob, operator, heapq
 import string
 import collections
 from math import log10
@@ -24,7 +23,7 @@ def splits(text, L=20):
 def Pwords(words): 
     """The Naive Bayes log probability of a sequence of words.
     """
-    return sum(Pw[w] for w in words)
+    return sum(Pw[w.lower()] for w in words)
 
 class Pdist(dict):
     """A probability distribution estimated from counts in datafile.