classification

Documentation

Overview

Package classification implements a classifier for use as a visModel in the vision service

Index

• func Overlay(img image.Image, classifications Classifications) (image.Image, error)
• type Classification
  • func NewClassification(score float64, label string) Classification
• type Classifications
  • func (cc Classifications) TopN(n int) (Classifications, error)
• type Classifier
• type Postprocessor
  • func NewLabelConfidenceFilter(labels map[string]float64) Postprocessor
  • func NewLabelFilter(labels map[string]interface{}) Postprocessor
  • func NewScoreFilter(conf float64) Postprocessor

Functions

func Overlay

func Overlay(img image.Image, classifications Classifications) (image.Image, error)

Overlay returns a color image with the classification labels and confidence scores overlaid on the original image.

Types

type Classification

type Classification interface {
	Score() float64
	Label() string
}

Classification returns a confidence score of the classification and a label of the class.

func NewClassification

func NewClassification(score float64, label string) Classification

NewClassification creates a simple 2D classification.

type Classifications

type Classifications []Classification

Classifications is a list of the Classification object.

func (Classifications) TopN

func (cc Classifications) TopN(n int) (Classifications, error)

TopN finds the N Classifications with the highest confidence scores.

type Classifier

type Classifier func(context.Context, image.Image) (Classifications, error)

A Classifier is defined as a function from an image to a list of Classifications.

type Postprocessor

type Postprocessor func(Classifications) Classifications

Postprocessor defines a function that filters/modifies on an incoming array of Classifications.

func NewLabelConfidenceFilter

func NewLabelConfidenceFilter(labels map[string]float64) Postprocessor

NewLabelConfidenceFilter returns a function that filters out classifications based on label map. Does not filter when input is empty.

func NewLabelFilter

func NewLabelFilter(labels map[string]interface{}) Postprocessor

NewLabelFilter returns a function that filters out classifications without one of the chosen labels. Does not filter when input is empty.

func NewScoreFilter

func NewScoreFilter(conf float64) Postprocessor

NewScoreFilter returns a function that filters out classifications below a certain confidence score.