Regressifier* Regressifier::deepCopy() const{
Regressifier *newInstance = createInstanceFromString( regressifierType );
if( newInstance == NULL ) return NULL;
if( !newInstance->deepCopyFrom( this ) ){
delete newInstance;
return NULL;
}
return newInstance;
}
Node* Node::createNewInstance() const{
return createInstanceFromString( type );
}
bool MultidimensionalRegression::loadLegacyModelFromFile( fstream &file ){
string word;
file >> word;
if(word != "NumFeatures:"){
errorLog << "loadModelFromFile(string filename) - Could not find NumFeatures!" << endl;
return false;
}
file >> numInputDimensions;
file >> word;
if(word != "NumOutputDimensions:"){
errorLog << "loadModelFromFile(string filename) - Could not find NumOutputDimensions!" << endl;
return false;
}
file >> numOutputDimensions;
file >> word;
if(word != "UseScaling:"){
errorLog << "loadModelFromFile(string filename) - Could not find UseScaling!" << endl;
return false;
}
file >> useScaling;
///Read the ranges if needed
if( useScaling ){
//Resize the ranges buffer
inputVectorRanges.resize(numInputDimensions);
targetVectorRanges.resize(numOutputDimensions);
//Load the ranges
file >> word;
if(word != "InputVectorRanges:"){
file.close();
errorLog << "loadModelFromFile(string filename) - Failed to find InputVectorRanges!" << endl;
return false;
}
for(UINT j=0; j<inputVectorRanges.size(); j++){
file >> inputVectorRanges[j].minValue;
file >> inputVectorRanges[j].maxValue;
}
file >> word;
if(word != "OutputVectorRanges:"){
file.close();
errorLog << "loadModelFromFile(string filename) - Failed to find OutputVectorRanges!" << endl;
return false;
}
for(UINT j=0; j<targetVectorRanges.size(); j++){
file >> targetVectorRanges[j].minValue;
file >> targetVectorRanges[j].maxValue;
}
}
file >> word;
if( word != "Regressifier:" ){
errorLog << "loadModelFromFile(string filename) - Failed to find Regressifier!" << endl;
return false;
}
//Load the regressifier type
string regressifierType;
file >> regressifierType;
if( regressifierType == "NOT_SET" ){
return true;
}
//Create the regressifer
regressifier = createInstanceFromString( regressifierType );
if( regressifier == NULL ){
errorLog << "loadModelFromFile(fstream &file) - Failed to create regression instance from string!" << endl;
return false;
}
if( !regressifier->loadModelFromFile( file ) ){
errorLog <<"loadModelFromFile(fstream &file) - Failed to load regressifier!" << endl;
return false;
}
if( numOutputDimensions > 0 ){
//Initialize the regression modules
regressionModules.resize(numOutputDimensions, NULL);
for(UINT i=0; i<regressionModules.size(); i++){
regressionModules[i] = createInstanceFromString( regressifierType );
if( !regressionModules[i]->loadModelFromFile( file ) ){
errorLog << "loadModelFromFile(fstream &file) - Failed to load regression module " << i << endl;
return false;
}
}
//Resize the regression data vector
regressionData.resize(numOutputDimensions,0);
//Flag that the model has been trained
trained = true;
}
return true;
}
PreProcessing* PreProcessing::createNewInstance() const{
return createInstanceFromString(preProcessingType);
}
Context* Context::createNewInstance() const{
return createInstanceFromString( contextType );
}
Regressifier* Regressifier::createNewInstance() const{
return createInstanceFromString( regressifierType );
}
bool BAG::loadModelFromFile(fstream &file){
clear();
UINT ensembleSize = 0;
if(!file.is_open())
{
errorLog << "loadModelFromFile(string filename) - Could not open file to load model" << endl;
return false;
}
std::string word;
file >> word;
//Check to see if we should load a legacy file
if( word == "GRT_BAG_MODEL_FILE_V1.0" ){
return loadLegacyModelFromFile( file );
}
//Find the file type header
if(word != "GRT_BAG_MODEL_FILE_V2.0"){
errorLog << "loadModelFromFile(string filename) - Could not find Model File Header" << endl;
return false;
}
//Load the base settings from the file
if( !Classifier::loadBaseSettingsFromFile(file) ){
errorLog << "loadModelFromFile(string filename) - Failed to load base settings from file!" << endl;
return false;
}
if( trained ){
//Load the ensemble size
file >> word;
if(word != "EnsembleSize:"){
errorLog << "loadModelFromFile(string filename) - Could not find the EnsembleSize!" << endl;
return false;
}
file >> ensembleSize;
//Load the weights
weights.resize( ensembleSize );
file >> word;
if(word != "Weights:"){
errorLog << "loadModelFromFile(string filename) - Could not find the Weights!" << endl;
return false;
}
for(UINT i=0; i<ensembleSize; i++){
file >> weights[i];
}
//Load the classifier types
vector< string > classifierTypes( ensembleSize );
file >> word;
if(word != "ClassifierTypes:"){
errorLog << "loadModelFromFile(string filename) - Could not find the ClassifierTypes!" << endl;
return false;
}
for(UINT i=0; i<ensembleSize; i++){
file >> classifierTypes[i];
}
//Load the ensemble
file >> word;
if(word != "Ensemble:"){
errorLog << "loadModelFromFile(string filename) - Could not find the Ensemble!" << endl;
return false;
}
ensemble.resize(ensembleSize,NULL);
for(UINT i=0; i<ensembleSize; i++){
ensemble[i] = createInstanceFromString( classifierTypes[i] );
if( ensemble[i] == NULL ){
errorLog << "loadModelFromFile(string filename) - Could not create a new classifier instance from the classifierType: " << classifierTypes[i] << endl;
clearEnsemble();
return false;
}
if( !ensemble[i]->loadModelFromFile( file ) ){
errorLog << "loadModelFromFile(string filename) - Failed to load ensemble classifier: " << i << endl;
clearEnsemble();
return false;
}
}
//Recompute the null rejection thresholds
recomputeNullRejectionThresholds();
//Resize the prediction results to make sure it is setup for realtime prediction
maxLikelihood = DEFAULT_NULL_LIKELIHOOD_VALUE;
bestDistance = DEFAULT_NULL_DISTANCE_VALUE;
classLikelihoods.resize(numClasses,DEFAULT_NULL_LIKELIHOOD_VALUE);
classDistances.resize(numClasses,DEFAULT_NULL_DISTANCE_VALUE);
}
return true;
}
Classifier* Classifier::createNewInstance() const{
return createInstanceFromString( classifierType );
}
bool MultidimensionalRegression::loadModelFromFile( std::fstream &file ){
trained = false;
numInputDimensions = 0;
deleteAll();
if(!file.is_open())
{
errorLog << "loadModelFromFile(string filename) - Could not open file to load model" << std::endl;
return false;
}
std::string word;
//Find the file type header
file >> word;
//Check to see if we should load a legacy file
if( word == "GRT_MULTIDIMENSIONAL_REGRESSION_MODEL_FILE_V1.0" ){
return loadLegacyModelFromFile( file );
}
if( word != "GRT_MULTIDIMENSIONAL_REGRESSION_MODEL_FILE_V2.0" ){
errorLog << "loadModelFromFile( fstream &file ) - Could not find Model File Header" << std::endl;
return false;
}
//Load the regressifier settings from the file
if( !Regressifier::loadBaseSettingsFromFile(file) ){
errorLog <<"loadModelFromFile( fstream &file ) - Failed to save Regressifier base settings to file!" << std::endl;
return false;
}
file >> word;
if( word != "Regressifier:" ){
errorLog << "loadModelFromFile(string filename) - Failed to find Regressifier!" << std::endl;
return false;
}
//Load the regressifier type
std::string regressifierType;
file >> regressifierType;
if( regressifierType == "NOT_SET" ){
return true;
}
//Create the regressifer
regressifier = createInstanceFromString( regressifierType );
if( regressifier == NULL ){
errorLog << "loadModelFromFile(fstream &file) - Failed to create regression instance from string!" << std::endl;
return false;
}
if( !regressifier->loadModelFromFile( file ) ){
errorLog <<"loadModelFromFile(fstream &file) - Failed to load regressifier!" << std::endl;
return false;
}
if( numOutputDimensions > 0 ){
//Initialize the regression modules
regressionModules.resize(numOutputDimensions, NULL);
for(UINT i=0; i<regressionModules.getSize(); i++){
regressionModules[i] = createInstanceFromString( regressifierType );
if( !regressionModules[i]->loadModelFromFile( file ) ){
errorLog << "loadModelFromFile(fstream &file) - Failed to load regression module " << i << std::endl;
return false;
}
}
}
return true;
}
FeatureExtraction* FeatureExtraction::createNewInstance() const{
return createInstanceFromString(featureExtractionType);
}
bool BAG::loadModelFromFile(fstream &file){
trained = false;
numFeatures = 0;
numClasses = 0;
classLabels.clear();
clearEnsemble();
UINT ensembleSize = 0;
if(!file.is_open())
{
errorLog << "loadModelFromFile(string filename) - Could not open file to load model" << endl;
return false;
}
std::string word;
//Find the file type header
file >> word;
if(word != "GRT_BAG_MODEL_FILE_V1.0"){
errorLog << "loadModelFromFile(string filename) - Could not find Model File Header" << endl;
return false;
}
file >> word;
if(word != "Trained:"){
errorLog << "loadModelFromFile(string filename) - Could not find Trained Header" << endl;
return false;
}
file >> trained;
file >> word;
if(word != "NumFeatures:"){
errorLog << "loadModelFromFile(string filename) - Could not find NumFeatures!" << endl;
return false;
}
file >> numFeatures;
file >> word;
if(word != "NumClasses:"){
errorLog << "loadModelFromFile(string filename) - Could not find NumClasses!" << endl;
return false;
}
file >> numClasses;
file >> word;
if(word != "UseScaling:"){
errorLog << "loadModelFromFile(string filename) - Could not find UseScaling!" << endl;
return false;
}
file >> useScaling;
file >> word;
if(word != "UseNullRejection:"){
errorLog << "loadModelFromFile(string filename) - Could not find UseNullRejection!" << endl;
return false;
}
file >> useNullRejection;
///Read the ranges if needed
if( useScaling ){
//Resize the ranges buffer
ranges.resize(numFeatures);
file >> word;
if(word != "Ranges:"){
errorLog << "loadModelFromFile(string filename) - Could not find the Ranges!" << endl;
return false;
}
for(UINT n=0; n<ranges.size(); n++){
file >> ranges[n].minValue;
file >> ranges[n].maxValue;
}
}
//Resize the buffer
classLabels.resize(numClasses);
//Load the class labels
file >> word;
if(word != "ClassLabels:"){
errorLog << "loadModelFromFile(string filename) - Could not find the ClassLabels!" << endl;
return false;
}
for(UINT i=0; i<numClasses; i++){
file >> classLabels[i];
}
//Load the ensemble size
file >> word;
if(word != "EnsembleSize:"){
errorLog << "loadModelFromFile(string filename) - Could not find the EnsembleSize!" << endl;
return false;
}
file >> ensembleSize;
if( ensembleSize > 0 ){
//Load the weights
weights.resize( ensembleSize );
file >> word;
if(word != "Weights:"){
errorLog << "loadModelFromFile(string filename) - Could not find the Weights!" << endl;
return false;
}
for(UINT i=0; i<ensembleSize; i++){
file >> weights[i];
}
//Load the classifier types
vector< string > classifierTypes( ensembleSize );
file >> word;
if(word != "ClassifierTypes:"){
errorLog << "loadModelFromFile(string filename) - Could not find the ClassifierTypes!" << endl;
return false;
}
for(UINT i=0; i<ensembleSize; i++){
file >> classifierTypes[i];
}
//Load the ensemble
file >> word;
if(word != "Ensemble:"){
errorLog << "loadModelFromFile(string filename) - Could not find the Ensemble!" << endl;
return false;
}
ensemble.resize(ensembleSize,NULL);
for(UINT i=0; i<ensembleSize; i++){
ensemble[i] = createInstanceFromString( classifierTypes[i] );
if( ensemble[i] == NULL ){
errorLog << "loadModelFromFile(string filename) - Could not create a new classifier instance from the classifierType: " << classifierTypes[i] << endl;
clearEnsemble();
return false;
}
if( !ensemble[i]->loadModelFromFile( file ) ){
errorLog << "loadModelFromFile(string filename) - Failed to load ensemble classifier: " << i << endl;
clearEnsemble();
return false;
}
}
}