name: NGramModel

synopsis:

g++ [flags ...] file ... -l /isip/tools/lib/$ISIP_BINARY/lib_stat.a

#include <NGramModel.h>

~NGramModel();
NGramModel();
NGramModel(const NGramModel& copy_sym);

boolean load(Sof& sof, const long& tag, const Vector& symbol_table,
             const String& name = CLASS_NAME);
boolean getScore(const VectorLong& index_a, float& score_a);
boolean setOrder(long order);

quick start:

#include <NGramModel.h>

// configure ngram object and load a trigram file
//
NGramModel ngrm;
ngrm.setOrder(3);
ngrm.load(sof, 1, symbol_table);
sof.close();

// test getScore
//
String symb1(L"ONE"), symb2(L"TWO"), symb3(L"FOUR");
VectorLong index(2);

// get a bigram (ONE TWO) probability
//
index(0) = *symbol_hash.get(symb1);
index(1) = *symbol_hash.get(symb2);

float score = ngrm.getScore(index);

description:

NGramModel is a class that manages N-gram probabilities, such as bigram, trigram, etc. This class can be used to load an input N-gram file, such as NIST_ARPA grammar file, and then store it to HashTable for lookup.

dependencies:

public constants:

define the class name:

static const String CLASS_NAME = L"NGramModel";

define i/o related constants:

static const String NGramModel::PARAM_ORDER = L"order";

static const String NGramModel::PARAM_KEYS = L"keys";

static const String NGramModel::PARAM_VALUES = L"values";

static const String NGramModel::PARAM_GRAM_HASH = L"gram_hash_table";

define the default value(s) of the class data:

static const String DEF_PARAM = L"";

static const Long DEF_ORDER = 0;

static const Float DEF_ORDER = 0.0;

define default arguments to methods:

static const ulong DEF_REF_INCR = (ulong)1;

static const ulong DEF_REF_DECR = (ulong)1;

error codes:

error code indicating a general error:
```
static const long ERR = 00060600;
```

protected data:

define the order of Ngram:

Long order_d;
  
   define a hash table containing Ngrams:
       
       HashTable gram_hash_d;
       
  
 define a vector which mapping search symbols:

       SiVector symbol_table_d;

   debug level:
       
       static Integral::DEBUG debug_level_d;
       
  
 memory manager:
       
       static MemoryManager mgr_d;

required public methods:

static methods:

static const String& name();

static boolean diagnose(Integral::DEBUG debug_level);

debug methods:

static boolean setDebug(Integral::DEBUG debug_level);

boolean debug(const unichar* message) const;

destructor/constructor(s):

~NGramModel();

NGramModel(Long order = DEF_ORDER);

assign methods:

boolean assign(const NGramModel& copy_arg);

operator= methods:

Histogram& operator= (const NGramModel& arg);

i/o methods:

long sofSize() const;

boolean read(Sof& sof, long tag, const String& name = CLASS_NAME);

boolean write(Sof& sof, long tag, const String& name = CLASS_NAME) const;

boolean readData(Sof& sof, const String& pname = DEF_PARAM, long size = SofParser::FULL_OBJECT, boolean param = true, boolean nested = false);

boolean writeData(Sof& sof, const String& pname = DEF_PARAM) const;

equality methods:

boolean eq(const NGramModel& arg) const;

memory management methods:

static void* operator new(size_t size);

static void* operator new[](size_t size);

static void operator delete(void* ptr);

static void operator delete[](void* ptr);

static boolean setGrowSize(long grow_size);

boolean clear(Integral::CMODE ctype = Integral::DEF_CMODE);

class-specific public methods:

NGramModel set/get methods:

boolean setOrder(long order);

long getOrder() const;

HashTable<Long, NGramNode> getGramHash();

load/store methods:

boolean load(Sof& sof, const long& tag, const Vector& symbol_table, const String& name = CLASS_NAME);

boolean store(Sof& sof, const long& tag, const String& name = CLASS_NAME) const;

Ngram related methods:

float getScore(const VectorLong& symbol_indices);

boolean getScore(const VectorLong& index_a, float& score_a);

private methods:

none

examples:

none

notes:

none