Struct RunningStats 

pub struct RunningStats { /* private fields */ }

Available on crate feature statistics only.

Running statistics calculator using Welford’s algorithm

Computes mean, variance, standard deviation, min, and max in a single pass with O(1) memory. Uses Welford’s numerically stable algorithm to avoid catastrophic cancellation.

Example

use flowstats::statistics::RunningStats;

let mut stats = RunningStats::new();

for value in [2.0, 4.0, 4.0, 4.0, 5.0, 5.0, 7.0, 9.0] {
    stats.add(value);
}

assert!((stats.mean() - 5.0).abs() < 0.001);
assert!((stats.variance() - 4.0).abs() < 0.001);
assert!((stats.stddev() - 2.0).abs() < 0.001);
assert_eq!(stats.min(), Some(2.0));
assert_eq!(stats.max(), Some(9.0));

Distributed Usage

use flowstats::statistics::RunningStats;
use flowstats::traits::Sketch;

let mut stats1 = RunningStats::new();
let mut stats2 = RunningStats::new();

// Worker 1
for v in [1.0, 2.0, 3.0] {
    stats1.add(v);
}

// Worker 2
for v in [4.0, 5.0, 6.0] {
    stats2.add(v);
}

// Merge
stats1.merge(&stats2).unwrap();
assert!((stats1.mean() - 3.5).abs() < 0.001);

Implementations

impl RunningStats

pub fn new() -> Self

Create a new empty statistics accumulator

pub fn add(&mut self, value: f64)

Add a value to the statistics

Uses Welford’s online algorithm for numerical stability. NaN values are ignored to prevent poisoning the statistics.

pub fn len(&self) -> u64

Get the number of values

pub fn is_empty(&self) -> bool

Check if empty

pub fn mean(&self) -> f64

Get the mean (average)

pub fn variance(&self) -> f64

Get the population variance

This is the variance assuming the data represents the entire population. Use sample_variance() if the data is a sample.

pub fn sample_variance(&self) -> f64

Get the sample variance

This is the unbiased variance estimator (Bessel’s correction). Use variance() for population variance.

pub fn stddev(&self) -> f64

Get the population standard deviation

pub fn sample_stddev(&self) -> f64

Get the sample standard deviation

pub fn min(&self) -> Option<f64>

Get the minimum value

pub fn max(&self) -> Option<f64>

Get the maximum value

pub fn range(&self) -> Option<f64>

Get the range (max - min)

pub fn sum(&self) -> f64

Get the sum of all values

pub fn merge_stats(&mut self, other: &Self)

Merge with another RunningStats using parallel algorithm

Uses Chan et al.’s parallel algorithm for combining statistics.

Trait Implementations

impl Clone for RunningStats

fn clone(&self) -> RunningStats

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for RunningStats

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for RunningStats

fn default() -> Self

Returns the “default value” for a type.

impl Sketch for RunningStats

type Item = f64

The type of item this sketch processes

fn update(&mut self, item: &Self::Item)

Add an item to the sketch

fn merge(&mut self, other: &Self) -> Result<(), MergeError>

Merge another sketch into this one

fn clear(&mut self)

Reset sketch to empty state

fn size_bytes(&self) -> usize

Memory usage in bytes

fn count(&self) -> u64

Number of items processed

fn is_empty(&self) -> bool

Check if sketch is empty