result

Documentation

Overview

This package contains data types for the training results representation. Each training for Bayes algorithm provides these data:

• Classes frequencies
• Features per classes frequencies
• Total number of samples processed during the training

This package contains data types for the training results representation. Each training for Bayes algorithm provides these data:

• Classes frequencies
• Features per classes frequencies
• Total number of samples processed during the training

Index

• type ClassFrequency
• type TextClassFrequency
• type TextTokenFrequency
• type TextTrainingResult
  • func NewTextTrainingResult() *TextTrainingResult
  • func (t *TextTrainingResult) IncClassCount(class string)
  • func (t *TextTrainingResult) IncSamplesCount()
  • func (t *TextTrainingResult) IncTokenCount(class, token string)
• type TokenFrequency
• type TrainingResult
  • func NewTrainingResult() *TrainingResult
  • func (t *TrainingResult) IncClassCount(classID int64)
  • func (t *TrainingResult) IncSamplesCount()
  • func (t *TrainingResult) IncTokenCount(classID, tokenID int64)

Types

type ClassFrequency

type ClassFrequency map[int64]int64

This map contains classes frequencies in the training dataset. The key of map is a Class ID and values is a count of samples with this class.

type TextClassFrequency

type TextClassFrequency map[string]int64

This map contains classes frequencies in the training dataset. The key of map is a Class ID and values is a count of samples with this class.

type TextTokenFrequency

type TextTokenFrequency map[string]map[string]int64

This map contains tokens (or features) frequencies for classes in the training dataset. The first-level key is a Class ID, the second-level key is Token ID, and the value is a number of sample which have the corresponding class and token.

type TextTrainingResult

type TextTrainingResult struct {
	SamplesCount       int64 // The total number of samples seen in dataset
	TextClassFrequency TextClassFrequency
	TextTokenFrequency TextTokenFrequency
}

func NewTextTrainingResult

func NewTextTrainingResult() *TextTrainingResult

func (*TextTrainingResult) IncClassCount

func (t *TextTrainingResult) IncClassCount(class string)

func (*TextTrainingResult) IncSamplesCount

func (t *TextTrainingResult) IncSamplesCount()

func (*TextTrainingResult) IncTokenCount

func (t *TextTrainingResult) IncTokenCount(class, token string)

type TokenFrequency

type TokenFrequency map[int64]map[int64]int64

This map contains tokens (or features) frequencies for classes in the training dataset. The first-level key is a Class ID, the second-level key is Token ID, and the value is a number of sample which have the corresponding class and token.

type TrainingResult

type TrainingResult struct {
	SamplesCount   int64 // The total number of samples seen in dataset
	ClassFrequency ClassFrequency
	TokenFrequency TokenFrequency
}

func NewTrainingResult

func NewTrainingResult() *TrainingResult

func (*TrainingResult) IncClassCount

func (t *TrainingResult) IncClassCount(classID int64)

func (*TrainingResult) IncSamplesCount

func (t *TrainingResult) IncSamplesCount()

func (*TrainingResult) IncTokenCount

func (t *TrainingResult) IncTokenCount(classID, tokenID int64)