Joshua
open source statistical hierarchical phrase-based machine translation system
 All Classes Namespaces Functions Variables Typedefs Enumerations Enumerator Friends
joshua.pro.Optimizer Class Reference
Collaboration diagram for joshua.pro.Optimizer:
[legend]

List of all members.

Public Member Functions

 Optimizer (long _seed, int _sentNum, Vector< String > _output, double[] _initialTotalLambda, HashMap< String, String >[] _feat_hash, HashMap< String, String >[] _stats_hash, double _finalScore, EvaluationMetric _evalMetric, int _Tau, int _Xi, double _metricDiff, double[] _normalizationOptions, String _classifierAlg, String[] _classifierParam, String _trainMode, int _numSparseParam, int _numRegParam, String _nbestFormat)
double[] run_Optimizer ()
double computeCorpusMetricScore (double[] finalLambda)
Vector< String > process_Params ()

Private Member Functions

Vector< String > Sampler (int sentId)
double Alpha (double x)
HashMap< String, Double > compute_Score (int sentId, String[] cands)
void normalizeLambda (double[] origLambda)
double L_norm (double[] A, double pow)
void normalizeLambda_mode3 (double[] origLambda)

Private Attributes

EvaluationMetric evalMetric
Vector< String > output
double[] initialLambda
double[] finalLambda
double[] copyLambda
double finalScore
double[] normalizationOptions
HashMap< String, String >[] feat_hash
HashMap< String, String >[] stats_hash
Random randgen
int paramDim
int regParamDim
int sentNum
String trainMode
int Tau
int Xi
double metricDiff
String classifierAlg
String nbestFormat
String[] classifierParam

Static Private Attributes

static final double NegInf = (-1.0 / 0.0)
static final double PosInf = (+1.0 / 0.0)

Constructor & Destructor Documentation

joshua.pro.Optimizer.Optimizer ( long  _seed,
int  _sentNum,
Vector< String >  _output,
double[]  _initialTotalLambda,
HashMap< String, String >[]  _feat_hash,
HashMap< String, String >[]  _stats_hash,
double  _finalScore,
EvaluationMetric  _evalMetric,
int  _Tau,
int  _Xi,
double  _metricDiff,
double[]  _normalizationOptions,
String  _classifierAlg,
String[]  _classifierParam,
String  _trainMode,
int  _numSparseParam,
int  _numRegParam,
String  _nbestFormat 
)

Member Function Documentation

double joshua.pro.Optimizer.Alpha ( double  x) [private]

Here is the caller graph for this function:

HashMap<String, Double> joshua.pro.Optimizer.compute_Score ( int  sentId,
String[]  cands 
) [private]

Here is the call graph for this function:

Here is the caller graph for this function:

double joshua.pro.Optimizer.computeCorpusMetricScore ( double[]  finalLambda)

Here is the call graph for this function:

Here is the caller graph for this function:

double joshua.pro.Optimizer.L_norm ( double[]  A,
double  pow 
) [private]

Here is the caller graph for this function:

void joshua.pro.Optimizer.normalizeLambda ( double[]  origLambda) [private]

Here is the call graph for this function:

Here is the caller graph for this function:

void joshua.pro.Optimizer.normalizeLambda_mode3 ( double[]  origLambda) [private]

Here is the caller graph for this function:

Here is the call graph for this function:

Here is the caller graph for this function:

Here is the call graph for this function:

Here is the caller graph for this function:

Vector<String> joshua.pro.Optimizer.Sampler ( int  sentId) [private]

Here is the call graph for this function:

Here is the caller graph for this function:


Member Data Documentation

double [] joshua.pro.Optimizer.copyLambda [private]
HashMap<String, String> [] joshua.pro.Optimizer.feat_hash [private]
double [] joshua.pro.Optimizer.finalLambda [private]
final double joshua.pro.Optimizer.NegInf = (-1.0 / 0.0) [static, private]
Vector<String> joshua.pro.Optimizer.output [private]
final double joshua.pro.Optimizer.PosInf = (+1.0 / 0.0) [static, private]
Random joshua.pro.Optimizer.randgen [private]
HashMap<String, String> [] joshua.pro.Optimizer.stats_hash [private]
int joshua.pro.Optimizer.Tau [private]
int joshua.pro.Optimizer.Xi [private]