blob: 9916b3ed7f717a82ca35a3b451e15ef0d921282a (
plain)
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
|
/* SPDX-License-Identifier: BSD-2-Clause */
/*
* Copyright (C) 2019, Raspberry Pi (Trading) Limited
*
* histogram.cpp - histogram calculations
*/
#include <math.h>
#include <stdio.h>
#include "histogram.hpp"
using namespace RPiController;
uint64_t Histogram::CumulativeFreq(double bin) const
{
if (bin <= 0)
return 0;
else if (bin >= Bins())
return Total();
int b = (int)bin;
return cumulative_[b] +
(bin - b) * (cumulative_[b + 1] - cumulative_[b]);
}
double Histogram::Quantile(double q, int first, int last) const
{
if (first == -1)
first = 0;
if (last == -1)
last = cumulative_.size() - 2;
assert(first <= last);
uint64_t items = q * Total();
while (first < last) // binary search to find the right bin
{
int middle = (first + last) / 2;
if (cumulative_[middle + 1] > items)
last = middle; // between first and middle
else
first = middle + 1; // after middle
}
assert(items >= cumulative_[first] && items <= cumulative_[last + 1]);
double frac = cumulative_[first + 1] == cumulative_[first] ? 0
: (double)(items - cumulative_[first]) /
(cumulative_[first + 1] - cumulative_[first]);
return first + frac;
}
double Histogram::InterQuantileMean(double q_lo, double q_hi) const
{
assert(q_hi > q_lo);
double p_lo = Quantile(q_lo);
double p_hi = Quantile(q_hi, (int)p_lo);
double sum_bin_freq = 0, cumul_freq = 0;
for (double p_next = floor(p_lo) + 1.0; p_next <= ceil(p_hi);
p_lo = p_next, p_next += 1.0) {
int bin = floor(p_lo);
double freq = (cumulative_[bin + 1] - cumulative_[bin]) *
(std::min(p_next, p_hi) - p_lo);
sum_bin_freq += bin * freq;
cumul_freq += freq;
}
// add 0.5 to give an average for bin mid-points
return sum_bin_freq / cumul_freq + 0.5;
}
|
