andreacavalli
/
rust-average
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Reduce code duplication

This commit is contained in:
Vinzent Steinberg 2017-05-22 13:55:11 +02:00
parent 77fa8b4ed2
commit 9bb501116d
1 changed files with 29 additions and 51 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

80
src/weighted_average.rs
View File

@ -1,6 +1,6 @@
use core;
use conv::ApproxFrom;
use super::Average;
/// Estimate the weighted and unweighted arithmetic mean and the unweighted
/// variance of a sequence of numbers ("population").
@ -26,12 +26,8 @@ pub struct WeightedAverage {
/// Weighted average value.
weighted_avg: f64,
/// Number of samples.
n: u64,
/// Unweighted average value.
unweighted_avg: f64,
/// Intermediate sum of squares for calculating the *unweighted* variance.
v: f64,
/// Estimator of unweighted average and its variance.
unweighted_avg: Average,
}
impl WeightedAverage {
@ -39,7 +35,7 @@ impl WeightedAverage {
pub fn new() -> WeightedAverage {
WeightedAverage {
weight_sum: 0., weight_sum_sq: 0., weighted_avg: 0.,
n: 0, unweighted_avg: 0., v: 0.,
unweighted_avg: Average::new(),
}
}
@ -58,15 +54,12 @@ impl WeightedAverage {
let prev_avg = self.weighted_avg;
self.weighted_avg = prev_avg + (weight / self.weight_sum) * (sample - prev_avg);
self.n += 1;
let delta = sample - self.unweighted_avg;
self.unweighted_avg += delta / f64::approx_from(self.n).unwrap();
self.v += delta * (sample - self.unweighted_avg);
self.unweighted_avg.add(sample);
}
/// Determine whether the sample is empty.
pub fn is_empty(&self) -> bool {
self.n == 0
self.unweighted_avg.is_empty()
}
/// Return the sum of the weights.
@ -86,12 +79,12 @@ impl WeightedAverage {
/// Estimate the unweighted mean of the sequence.
pub fn unweighted_mean(&self) -> f64 {
self.unweighted_avg
self.unweighted_avg.mean()
}
/// Return sample size.
pub fn len(&self) -> u64 {
self.n
self.unweighted_avg.len()
}
/// Calculate the effective sample size.
@ -106,20 +99,14 @@ impl WeightedAverage {
///
/// This is a biased estimator of the variance of the population.
pub fn population_variance(&self) -> f64 {
if self.n < 2 {
return 0.;
}
self.v / f64::approx_from(self.n).unwrap()
self.unweighted_avg.population_variance()
}
/// Calculate the *unweighted* sample variance.
///
/// This is an unbiased estimator of the variance of the population.
pub fn sample_variance(&self) -> f64 {
if self.n < 2 {
return 0.;
}
self.v / f64::approx_from(self.n - 1).unwrap()
self.unweighted_avg.sample_variance()
}
/// Estimate the standard error of the *weighted* mean of the sequence.
@ -160,28 +147,14 @@ impl WeightedAverage {
/// assert!((avg_total.error() - avg_left.error()).abs() < 1e-15);
/// ```
pub fn merge(&mut self, other: &WeightedAverage) {
// This is similar to the algorithm proposed by Chan et al. in 1979.
//
// See https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance.
{
let total_weight_sum = self.weight_sum + other.weight_sum;
self.weighted_avg = (self.weight_sum * self.weighted_avg
+ other.weight_sum * other.weighted_avg)
/ (self.weight_sum + other.weight_sum);
self.weight_sum = total_weight_sum;
self.weight_sum_sq += other.weight_sum_sq;
}
{
let delta = other.unweighted_avg - self.unweighted_avg;
let len_self = f64::approx_from(self.n).unwrap();
let len_other = f64::approx_from(other.n).unwrap();
let len_total = len_self + len_other;
self.n += other.n;
self.unweighted_avg = (len_self * self.unweighted_avg
+ len_other * other.unweighted_avg)
/ len_total;
self.v += other.v + delta*delta * len_self * len_other / len_total;
}
let total_weight_sum = self.weight_sum + other.weight_sum;
self.weighted_avg = (self.weight_sum * self.weighted_avg
+ other.weight_sum * other.weighted_avg)
/ total_weight_sum;
self.weight_sum = total_weight_sum;
self.weight_sum_sq += other.weight_sum_sq;
self.unweighted_avg.merge(&other.unweighted_avg);
}
}
@ -216,12 +189,15 @@ mod tests {
let mut avg_left: WeightedAverage = left.iter().map(|x| (*x, 1.)).collect();
let avg_right: WeightedAverage = right.iter().map(|x| (*x, 1.)).collect();
avg_left.merge(&avg_right);
assert_eq!(avg_total.n, avg_left.n);
assert_eq!(avg_total.weight_sum, avg_left.weight_sum);
assert_eq!(avg_total.weight_sum_sq, avg_left.weight_sum_sq);
assert_eq!(avg_total.weighted_avg, avg_left.weighted_avg);
assert_eq!(avg_total.unweighted_avg, avg_left.unweighted_avg);
assert_eq!(avg_total.v, avg_left.v);
assert_eq!(avg_total.unweighted_avg.len(), avg_left.unweighted_avg.len());
assert_eq!(avg_total.unweighted_avg.mean(), avg_left.unweighted_avg.mean());
assert_eq!(avg_total.unweighted_avg.sample_variance(),
avg_left.unweighted_avg.sample_variance());
}
}
@ -236,12 +212,14 @@ mod tests {
let mut avg_left: WeightedAverage = left.iter().map(|&(x, w)| (x, w)).collect();
let avg_right: WeightedAverage = right.iter().map(|&(x, w)| (x, w)).collect();
avg_left.merge(&avg_right);
assert_eq!(avg_total.n, avg_left.n);
assert_eq!(avg_total.unweighted_avg.len(), avg_left.unweighted_avg.len());
assert_almost_eq!(avg_total.weight_sum, avg_left.weight_sum, 1e-15);
assert_eq!(avg_total.weight_sum_sq, avg_left.weight_sum_sq);
assert_almost_eq!(avg_total.weighted_avg, avg_left.weighted_avg, 1e-15);
assert_almost_eq!(avg_total.unweighted_avg, avg_left.unweighted_avg, 1e-15);
assert_almost_eq!(avg_total.v, avg_left.v, 1e-14);
assert_almost_eq!(avg_total.unweighted_avg.mean(),
avg_left.unweighted_avg.mean(), 1e-15);
assert_almost_eq!(avg_total.unweighted_avg.sample_variance(),
avg_left.unweighted_avg.sample_variance(), 1e-14);
}
}
}