https://github.com/wchresta/fastbinning

Non-uniform quantization with linear average-case computation time
https://github.com/wchresta/fastbinning

Last synced: 2 months ago
JSON representation

Non-uniform quantization with linear average-case computation time

• Host: GitHub
• URL: https://github.com/wchresta/fastbinning
• Owner: wchresta
• License: apache-2.0
• Created: 2021-08-19T10:01:49.000Z (almost 4 years ago)
• Default Branch: main
• Last Pushed: 2021-08-19T18:35:45.000Z (almost 4 years ago)
• Last Synced: 2025-01-31T16:41:16.633Z (4 months ago)
• Language: Go
• Size: 9.77 KB
• Stars: 0
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# fastbinning

Non-uniform quantization with linear average-case computation time

## Introduction

This is an implementation of the quantization or binning algorithm proposed in [a paper by Cadenas and M. Megson](https://arxiv.org/abs/2108.08228) in Go.

## Example

We use the example presented in the paper:

```go

bin, err := fastbinning.New([]float64{2, 11, 19, 20, 21, 27, 29, 30})

if err != nil {

    // This only happens if the Boundaries are not monotonically increasing.

    fmt.Printf("Creation of Bin failed: %s", err.Error())

}

// Search operations run in constant time and space on average

// Worst case is log(len(Boundaries)) time and constant space

bin.Search(4) // returns 1: 4 is between 2 and 11 which is the first bin

bin.Search(0) // returns 0: 0 is left of the first bin, so it is in bin 0

bin.Search(20.5) // returns 4: 20.5 is between 20 and 21

bin.Search(11) // returns 2: intervals are left-including, so 11 is in [11,19)

```