perceptron

Documentation

Overview

Package perceptron Jiagu自然语言处理工具训练方法

Index

• Variables
• type AveragedPerceptron
  • func NewAveragedPerceptron() *AveragedPerceptron
  • func NewAveragedPerceptronFromJSON(aModel model.PerceptronJSONModel) *AveragedPerceptron
  • func NewAveragedPerceptronFromReader(r io.Reader) (*AveragedPerceptron, error)
  • func (p *AveragedPerceptron) AddClass(clas string)
  • func (p *AveragedPerceptron) AverageWeights()
  • func (p *AveragedPerceptron) Len() int
  • func (p AveragedPerceptron) Model() model.PerceptronJSONModel
  • func (p *AveragedPerceptron) Predict(features []model.Feature) model.Class
  • func (p *AveragedPerceptron) Update(truth string, guess string, features []model.Feature)
• type Perceptron
  • func New() *Perceptron
  • func NewFromModelFile(loc string) (*Perceptron, error)
  • func NewFromReader(r io.Reader) (*Perceptron, error)
  • func (p *Perceptron) Load(loc string) error
  • func (p *Perceptron) Predict(words []string) []model.Class
  • func (p *Perceptron) Save(loc string) error
  • func (p *Perceptron) Train(sentences []model.Sentence, iters int, shuf bool, showProgressBar bool) float64

Variables

var (
	DefaultStarts = []string{"-START-", "-START2-"}
	DefaultEnds   = []string{"-END-", "-END2-"}
)

Types

type AveragedPerceptron

type AveragedPerceptron struct {
	// contains filtered or unexported fields
}

AveragedPerceptron Averaged Perceptron for POS tagging

func NewAveragedPerceptron

func NewAveragedPerceptron() *AveragedPerceptron

NewAveragedPerceptron 新建Perceptron

func NewAveragedPerceptronFromJSON

func NewAveragedPerceptronFromJSON(aModel model.PerceptronJSONModel) *AveragedPerceptron

NewAveragedPerceptronFromJSON 从JSON Model创建Perceptron

func NewAveragedPerceptronFromReader

func NewAveragedPerceptronFromReader(r io.Reader) (*AveragedPerceptron, error)

NewAveragedPerceptronFromReader 从io.Reader创建Perceptron

func (*AveragedPerceptron) AddClass

func (p *AveragedPerceptron) AddClass(clas string)

AddClass add class

func (*AveragedPerceptron) AverageWeights

func (p *AveragedPerceptron) AverageWeights()

AverageWeights Average weights from all iterations.

func (*AveragedPerceptron) Len

func (p *AveragedPerceptron) Len() int

func (AveragedPerceptron) Model

func (p AveragedPerceptron) Model() model.PerceptronJSONModel

Model

func (*AveragedPerceptron) Predict

func (p *AveragedPerceptron) Predict(features []model.Feature) model.Class

Predict Dot-product the features and current weights and return the best label.

func (*AveragedPerceptron) Update

func (p *AveragedPerceptron) Update(truth string, guess string, features []model.Feature)

Update update the feature weights.

type Perceptron

type Perceptron struct {
	// contains filtered or unexported fields
}

Perceptron perceptron 核心类

func New

func New() *Perceptron

New 新建Perceptron

func NewFromModelFile

func NewFromModelFile(loc string) (*Perceptron, error)

NewFromModelFile 从model文件新建Perceptron

func NewFromReader

func NewFromReader(r io.Reader) (*Perceptron, error)

NewFromReader 从io.Reader新建Perceptron

func (*Perceptron) Load

func (p *Perceptron) Load(loc string) error

Load load trained model

func (*Perceptron) Predict

func (p *Perceptron) Predict(words []string) []model.Class

Predect 预测分类

func (*Perceptron) Save

func (p *Perceptron) Save(loc string) error

Save save trained model

func (*Perceptron) Train

func (p *Perceptron) Train(sentences []model.Sentence, iters int, shuf bool, showProgressBar bool) float64

Train train data