https://github.com/jettify/streamhist

https://github.com/jettify/streamhist

Last synced: about 1 month ago
JSON representation

• Host: GitHub
• URL: https://github.com/jettify/streamhist
• Owner: jettify
• License: apache-2.0
• Created: 2020-12-20T22:22:47.000Z (almost 4 years ago)
• Default Branch: main
• Last Pushed: 2021-03-05T02:16:07.000Z (over 3 years ago)
• Last Synced: 2024-09-17T03:53:40.614Z (about 2 months ago)
• Language: Rust
• Size: 33.2 KB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# StreamHist

[![ci-badge](https://github.com/jettify/streamhist/workflows/CI/badge.svg)](https://github.com/jettify/streamhist/actions?query=workflow%3ACI)

[![Crates.io](https://img.shields.io/crates/v/streamhist.svg)](https://crates.io/crates/streamhist)

[![Documentation](https://docs.rs/streamhist/badge.svg)](https://docs.rs/streamhist/)

A rust implementation of a streaming centroid histogram algorithm found in

[Streaming Parallel Decision Trees](http://jmlr.org/papers/volume11/ben-haim10a/ben-haim10a.pdf)

 paper by Ben-Haim/Tom-Tov.

 # Example

 ```rust

 use rand::SeedableRng;

 use rand_distr::{Distribution, Normal};

 use rand_isaac::Isaac64Rng;

 use streamhist::StreamingHistogram;

 fn main() {

     let mut rng = Isaac64Rng::seed_from_u64(42);

     let dist = Normal::new(2.0, 3.0).unwrap();

     let mut hist = StreamingHistogram::new(32);

     let maxn = 10000;

     let vals: Vec = (0..maxn).map(|_| dist.sample(&mut rng)).collect();

     for v in vals.iter() {

         hist.insert_one(*v);

     }

     println!("------------------------------------------------");

     println!("Est Mean               {:?}", hist.mean().unwrap());

     println!("Est Var                {:?}", hist.var().unwrap());

     println!("Est Median             {:?}", hist.median().unwrap());

     println!("Est Count vals <= 2.0  {:?}", hist.count_less_then_eq(2.0));

     println!("Est quantile           {:?}", hist.quantile(0.75).unwrap());

     println!("Min                    {:?}", hist.min().unwrap());

     println!("Max                    {:?}", hist.max().unwrap());

     println!("Count                  {:?}", hist.count());

     println!("------------------------------------------------");

     assert_eq!(hist.count(), maxn);

 }

```

 # Lincese

  Licensed under the Apache License, Version 2.0