tesseract  v4.0.0-17-g361f3264
Open Source OCR Engine
tesseract::LanguageModelNgramInfo Struct Reference

#include <lm_state.h>

Collaboration diagram for tesseract::LanguageModelNgramInfo:

Public Member Functions

 LanguageModelNgramInfo (const char *c, int l, bool p, float nc, float ncc)
 

Public Attributes

STRING context
 
int context_unichar_step_len
 
bool pruned
 
float ngram_cost
 -ln(P_ngram_model(path)) More...
 
float ngram_and_classifier_cost
 -[ ln(P_classifier(path)) + scale_factor * ln(P_ngram_model(path)) ] More...
 

Detailed Description

Struct for storing additional information used by Ngram language model component.

Constructor & Destructor Documentation

◆ LanguageModelNgramInfo()

tesseract::LanguageModelNgramInfo::LanguageModelNgramInfo ( const char *  c,
int  l,
bool  p,
float  nc,
float  ncc 
)
inline

Member Data Documentation

◆ context

STRING tesseract::LanguageModelNgramInfo::context

◆ context_unichar_step_len

int tesseract::LanguageModelNgramInfo::context_unichar_step_len

Length of the context measured by advancing using UNICHAR::utf8_step() (should be at most the order of the character ngram model used).

◆ ngram_and_classifier_cost

float tesseract::LanguageModelNgramInfo::ngram_and_classifier_cost

-[ ln(P_classifier(path)) + scale_factor * ln(P_ngram_model(path)) ]

◆ ngram_cost

float tesseract::LanguageModelNgramInfo::ngram_cost

-ln(P_ngram_model(path))

◆ pruned

bool tesseract::LanguageModelNgramInfo::pruned

The paths with pruned set are pruned out from the perspective of the character ngram model. They are explored further because they represent a dictionary match or a top choice. Thus ngram_info is still computed for them in order to calculate the combined cost.


The documentation for this struct was generated from the following file: