normalizer/histogram_old.py at master · tinchoa/normalizer · GitHub

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
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
import scipy.stats as st
import math
from copy import copy
import numpy


d = {}
for i in range (0,10000):
	k = float(i)/10000.
	d[k] = st.norm.ppf(k)
d[1] = numpy.Inf

def ppf(x):
	x = int(x*10000)/10000.0
	try:
		return d[x]
	except KeyError:
		return numpy.NAN


class Histograma:
	def __init__(self, valores=[], bins=0):
		self.total = 0
		self.createHistograma(valores, bins)

	def createHistograma(self, valores=[], bins=0):
		#print "valores", valores
		if bins == 0:
			self.bins = math.sqrt(len(valores))
		if len(valores) == 0:
			self.valores = []
		else:
			self.valores = copy(valores)


		self.valores.sort()
		min = float(self.valores[0])
		#print "min", min
		max = float(self.valores[-1])
		#print "max", max

		if min == max:
			self.bins = 1
			pivo = 1
		else:
			pivo = (max-min)/(self.bins-1)

		self.pivo = pivo
		#print self.pivo, max, min

		self.hist = []

		tmp = min
		i = 0
		while len(self.hist) < self.bins:
			self.hist.append([tmp,tmp+pivo,0])
			for i in range(i, len(valores)):
				if self.valores[i] >= tmp and self.valores[i] < tmp+pivo:
					self.hist[-1][2] += 1
					self.total += 1
				else:
					break
			tmp+= pivo

	def updateHistograma(self, valores):
		self.valores = copy(valores)

		self.valores.sort()

		if self.valores[0] < self.hist[0][0]:
			min = self.valores[0]
		else:
			min = self.hist[0][0]
		if self.valores[-1] > self.hist[-1][1]:
			max = self.valores[-1]
		else:
			max = self.hist[-1][1]

		self.bins = math.sqrt(self.total)

		if min == max:
			self.bins = 1
			pivo = 1
		else:
			pivo = float(max-min)/float(self.bins-1)

		self.pivo = pivo

		while min < self.hist[0][0]:
			self.hist = [[self.hist[0][0]-self.pivo,self.hist[0][0],0]]+self.hist
		while max >= self.hist[-1][1]:
			self.hist = self.hist + [[self.hist[-1][1], self.hist[-1][1]+self.pivo, 0]]

		i = 0
		b = 0
		while b < len(self.hist):
			tmp = self.hist[b]
			for i in range(i, len(self.valores)):
				if self.valores[i] >= tmp[0]:
					if self.valores[i] < tmp[1]:
						tmp[2] += 1
						self.total += 1
				else:
					break
			b +=1

	def normalizedHistograma (self):
		normalizedHist = []
		k = 0
		for i in range(len(self.hist)):
			k += self.hist[i][-1]
			normalizedHist.append(copy(self.hist[i]))
			normalizedHist[-1][-1] = float(k)
		for j in normalizedHist:
			#j[-1] = st.norm.ppf(float(j[-1])/float(self.total))
			j[-1] = ppf(float(j[-1])/float(self.total))

		return normalizedHist

	def getNormalizedValues(self, valores):
		normalizedHist = self.normalizedHistograma()

		#print "hist", self.hist
		#print "norm hist", normalizedHist

		if type(valores) != type([]):
			valores = [valores]

		results = []
		for i in valores:
			for j in normalizedHist:
				if i >= j[0]:
					if i < j[1]:

						if j[2] < -2.5: final = -3.09
						elif j[2] > 2.5: final = 3.09
						else: final = j[2]

						results.append(final)
						break
		#print results, normalizedHist
		return results