[LNT] r239182 - Revert r238973 in favor of using future devision in stats routines

[Chris Matthews]

Author: cmatthews
Date: Fri Jun  5 14:24:31 2015
New Revision: 239182

URL: http://llvm.org/viewvc/llvm-project?rev=239182&view=rev
Log:
Revert r238973 in favor of using future devision in stats routines

We were incorrectly passing ints to these functions way too much.

Instead of asserting, use PEP238 devision operator. It is faster and
correct in all situations.

Modified:
    lnt/trunk/lnt/util/stats.py

Modified: lnt/trunk/lnt/util/stats.py
URL: http://llvm.org/viewvc/llvm-project/lnt/trunk/lnt/util/stats.py?rev=239182&r1=239181&r2=239182&view=diff
==============================================================================
--- lnt/trunk/lnt/util/stats.py (original)
+++ lnt/trunk/lnt/util/stats.py Fri Jun  5 14:24:31 2015
@@ -1,3 +1,4 @@
+from __future__ import division
 import math
 from lnt.external.stats.stats import mannwhitneyu as mannwhitneyu_large
 
@@ -18,32 +19,17 @@ def safe_max(l):
     else:
         return max(l)
 
-def check_floating(l):
-    """These math ops are totally wrong when they done on anything besides
-    floats.  I would just cast them, however that really slows them down.
-    So lets error on this """
-    for v in l:
-        assert type(v) == float, "Math op on non-floating point:" + str(v) + \
-            str(l)
-
-
 def mean(l):
-    check_floating(l)
     if l:
-        return float_mean(l)
+        return sum(l)/len(l)
     else:
         return None
 
 
-def float_mean(l):
-    return sum(l)/len(l)
-
-
 def median(l):
     if not l:
         return None
-    l = list(l)  # Could be a tuple.
-    check_floating(l)
+    l = list(l)
     l.sort()
     N = len(l)
     return (l[(N-1)//2] + l[N//2])*.5
@@ -52,13 +38,11 @@ def median(l):
 def median_absolute_deviation(l, med = None):
     if med is None:
         med = median(l)
-    check_floating(l)
     return median([abs(x - med) for x in l])
 
 
 def standard_deviation(l):
-    check_floating(l)
-    m = float_mean(l)
+    m = mean(l)
     means_sqrd = sum([(v - m)**2 for v in l]) / len(l)
     rms = math.sqrt(means_sqrd)
     return rms
@@ -68,8 +52,6 @@ def mannwhitneyu(a, b, sigLevel = .05):
     """
     Determine if sample a and b are the same at given significance level.
     """
-    check_floating(a)
-    check_floating(b)
     if len(a) <= 20 and len(b) <= 20:
         return mannwhitneyu_small(a, b, sigLevel)
     else:
@@ -89,7 +71,7 @@ def mannwhitneyu_small(a, b, sigLevel):
     assert len(a) <= 20, "Sample size must be less than 20."
     assert len(b) <= 20, "Sample size must be less than 20."
 
-    if sigLevel not in SIGN_TABLES:
+    if not sigLevel in SIGN_TABLES:
         raise ValueError("Do not have according significance table.")
 
     # Calculate U value for sample groups using method described on Wikipedia.