blob: 0346def1fdd9c11b2b34facdebdbd14415fb5dbd (
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
|
// Copyright 2019 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
#include "tracing/tracing/value/running_statistics.h"
#include <algorithm>
#include <cassert>
#include <cmath>
namespace catapult {
void RunningStatistics::Add(double value) {
count_++;
max_ = std::max(max_, value);
min_ = std::min(min_, value);
sum_ += value;
if (std::islessequal(value, 0.0)) {
meanlogs_valid_ = false;
} else if (meanlogs_valid_) {
meanlogs_ += (std::log(std::abs(value)) - meanlogs_) / count_;
}
// The following uses Welford's algorithm for computing running mean and
// variance. See http://www.johndcook.com/blog/standard_deviation.
if (count_ == 1) {
mean_ = value;
variance_ = 0.0;
} else {
double old_mean = mean_;
double old_variance = variance_;
// Using the 2nd formula for updating the mean yields better precision but
// it doesn't work for the case old_mean is Infinity. Hence we handle that
// case separately.
if (std::isinf(old_mean)) {
mean_ = sum_ / count_;
} else {
mean_ = (old_mean + (value - old_mean) / count_);
}
variance_ = old_variance + ((value - old_mean) * (value - mean_));
}
}
double RunningStatistics::meanlogs() const {
assert(meanlogs_valid_);
return meanlogs_;
}
double RunningStatistics::variance() const {
if (count() == 0 || count() == 1) {
return 0;
}
// This returns the variance of the samples after Bessel's correction has
// been applied.
return variance_ / (count() - 1);
}
} // namespace catapult
|
