1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
|
import math
def Mean(xs):
"""
Calculate mean of dataset
"""
if len(xs)==0:
raise RuntimeError("Can't calculate mean of empty sequence")
return float(sum(xs))/len(xs)
def Median(xs):
"""
Calculate median of dataset
"""
if len(xs)==0:
raise RuntimeError("Can't calculate median of empty sequence")
sorted_xs=sorted(xs)
central_idx = int((len(xs)-1)/2)
if (len(xs) % 2)==0:
return (sorted_xs[central_idx]+sorted_xs[central_idx+1])/2.0
else:
return sorted_xs[central_idx]
def StdDev(xs):
"""
Calculate standard-deviation of dataset
| sum[xi-<x>]^2 |
sigma=sqrt|---------------|
| n |
"""
mean=Mean(xs)
return math.sqrt(sum([(x-mean)**2 for x in xs])/len(xs))
def Min(xs):
return min(xs)
def Max(xs):
return max(xs)
def Correl(xs, ys):
"""
Calculates the correlation coefficient between xs and ys as
sum[(xi-<x>)*(yi-<y>)]
r=----------------------
sx*sy
where <x>, <y> are the mean of dataset xs and ys, and, sx and sy are the
standard deviations.
"""
if len(xs)!=len(ys):
raise RuntimeError("Can't calculate correl. Sequence lengths do not match.")
if len(xs)==1:
raise RuntimeError("Can't calculate correl of sequences with length 1.")
mean_x=Mean(xs)
mean_y=Mean(ys)
sigma_x, sigma_y=(0.0, 0.0)
cross_term=0.0
for x, y in zip(xs, ys):
cross_term+=(x-mean_x)*(y-mean_y)
sigma_x+=(x-mean_x)**2
sigma_y+=(y-mean_y)**2
sigma_x=math.sqrt(sigma_x)
sigma_y=math.sqrt(sigma_y)
return cross_term/(sigma_x*sigma_y)
def Histogram(xs, bounds, num_bins):
bins=[0 for i in range(num_bins)]
d=1.0*num_bins/(bounds[1]-bounds[0])
for x in xs:
index=int((x-bounds[0])*d)
if index>num_bins-1 or index<0:
continue
bins[index]+=1
return bins
|
