pep8 to sdm.utils

Author: msbrogli

sdm/utils.py

@@ -1,5 +1,6 @@
 from __future__ import print_function
-from math import log, exp
+from math import log
+
 
 def test_kernels(bits, sample, radius):
     import sdm as sdmlib
@@ -15,6 +16,7 @@ def test_kernels(bits, sample, radius):
             valid_kernels.append(kernel)
     return valid_kernels
 
+
 def div_pow2(x, a):
     try:
         r = x
@@ -24,6 +26,7 @@ def div_pow2(x, a):
     except OverflowError:
         return 2**(log(x)/log(2) - a)
 
+
 def calculate_probabilities(bits):
     from math import factorial
     comb = lambda a, b: factorial(a)//factorial(b)//factorial(a-b)
@@ -32,25 +35,29 @@ def calculate_probabilities(bits):
         acc.append(acc[-1] + comb(bits, i))
     return [div_pow2(x, bits) for x in acc[1:]]
 
+
 def __calculate_radius(bits, threshold=0.001):
     from math import factorial
     comb = lambda a, b: factorial(a)//factorial(b)//factorial(a-b)
     x = 0
     for i in range(bits+1):
         x += comb(bits, i)
         p = div_pow2(x, bits)
-        if p >= threadhold:
+        if p >= threshold:
             break
     return i
 
+
 def calculate_radius(bits, threshold=0.001):
     from scipy.stats import norm
     return int(bits/2.0 + norm.ppf(threshold)*(bits**0.5)/2.0)
-    if threadhold == 0.001:
+
+    if threshold == 0.001:
         # norminv(0.001) = 3.090232306167814
         return int(bits/2.0 - 3.090232306167814*(bits**0.5)/2.0)
     return __calculate_radius(bits, threshold)
 
+
 def calculate_probability(bits, radius):
     from scipy.stats import binom
     return binom.cdf(radius, bits, 0.5)