bool RandomForests::combineModels( const RandomForests &forest ){
if( !getTrained() ){
errorLog << "combineModels( const RandomForests &forest ) - This instance has not been trained!" << endl;
return false;
}
if( !forest.getTrained() ){
errorLog << "combineModels( const RandomForests &forest ) - This external forest instance has not been trained!" << endl;
return false;
}
if( this->getNumInputDimensions() != forest.getNumInputDimensions() ) {
errorLog << "combineModels( const RandomForests &forest ) - The number of input dimensions of the external forest (";
errorLog << forest.getNumInputDimensions() << ") does not match the number of input dimensions of this instance (";
errorLog << this->getNumInputDimensions() << ")!" << endl;
return false;
}
//Add the trees in the other forest to this model
DecisionTreeNode *node;
for(UINT i=0; i<forest.getForestSize(); i++){
node = forest.getTree(i);
if( node ){
this->forest.push_back( node->deepCopy() );
forestSize++;
}
}
return true;
}
bool RandomForests::setDecisionTreeNode( const DecisionTreeNode &node ){
if( decisionTreeNode != NULL ){
delete decisionTreeNode;
decisionTreeNode = NULL;
}
this->decisionTreeNode = node.deepCopy();
return true;
}
RandomForests::RandomForests(const DecisionTreeNode &decisionTreeNode,const UINT forestSize,const UINT numRandomSplits,const UINT minNumSamplesPerNode,const UINT maxDepth,const UINT trainingMode,const bool removeFeaturesAtEachSpilt,const bool useScaling,const Float bootstrappedDatasetWeight)
{
this->decisionTreeNode = decisionTreeNode.deepCopy();
this->forestSize = forestSize;
this->numRandomSplits = numRandomSplits;
this->minNumSamplesPerNode = minNumSamplesPerNode;
this->maxDepth = maxDepth;
this->trainingMode = trainingMode;
this->removeFeaturesAtEachSpilt = removeFeaturesAtEachSpilt;
this->useScaling = useScaling;
this->bootstrappedDatasetWeight = bootstrappedDatasetWeight;
classType = "RandomForests";
classifierType = classType;
classifierMode = STANDARD_CLASSIFIER_MODE;
useNullRejection = false;
supportsNullRejection = false;
useValidationSet = true;
validationSetSize = 20;
debugLog.setProceedingText("[DEBUG RandomForests]");
errorLog.setProceedingText("[ERROR RandomForests]");
trainingLog.setProceedingText("[TRAINING RandomForests]");
warningLog.setProceedingText("[WARNING RandomForests]");
}
bool RandomForests::loadModelFromFile(fstream &file){
clear();
if(!file.is_open())
{
errorLog << "loadModelFromFile(string filename) - Could not open file to load model" << endl;
return false;
}
std::string word;
std::string treeNodeType;
file >> word;
//Find the file type header
if(word != "GRT_RANDOM_FOREST_MODEL_FILE_V1.0"){
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;
}
file >> word;
if(word != "DecisionTreeNodeType:"){
Classifier::errorLog << "loadModelFromFile(string filename) - Could not find the DecisionTreeNodeType!" << endl;
return false;
}
file >> treeNodeType;
if( treeNodeType != "NULL" ){
decisionTreeNode = dynamic_cast< DecisionTreeNode* >( DecisionTreeNode::createInstanceFromString( treeNodeType ) );
if( decisionTreeNode == NULL ){
Classifier::errorLog << "loadModelFromFile(string filename) - Could not create new DecisionTreeNode from type: " << treeNodeType << endl;
return false;
}
if( !decisionTreeNode->loadFromFile( file ) ){
Classifier::errorLog <<"loadModelFromFile(fstream &file) - Failed to load decisionTreeNode settings from file!" << endl;
return false;
}
}else{
Classifier::errorLog <<"loadModelFromFile(fstream &file) - Failed to load decisionTreeNode! DecisionTreeNodeType is NULL!" << endl;
return false;
}
file >> word;
if(word != "ForestSize:"){
errorLog << "loadModelFromFile(string filename) - Could not find the ForestSize!" << endl;
return false;
}
file >> forestSize;
file >> word;
if(word != "NumSplittingSteps:"){
errorLog << "loadModelFromFile(string filename) - Could not find the NumSplittingSteps!" << endl;
return false;
}
file >> numRandomSplits;
file >> word;
if(word != "MinNumSamplesPerNode:"){
errorLog << "loadModelFromFile(string filename) - Could not find the MinNumSamplesPerNode!" << endl;
return false;
}
file >> minNumSamplesPerNode;
file >> word;
if(word != "MaxDepth:"){
errorLog << "loadModelFromFile(string filename) - Could not find the MaxDepth!" << endl;
return false;
}
file >> maxDepth;
file >> word;
if(word != "RemoveFeaturesAtEachSpilt:"){
errorLog << "loadModelFromFile(string filename) - Could not find the RemoveFeaturesAtEachSpilt!" << endl;
return false;
}
file >> removeFeaturesAtEachSpilt;
file >> word;
if(word != "TrainingMode:"){
errorLog << "loadModelFromFile(string filename) - Could not find the TrainingMode!" << endl;
return false;
}
file >> trainingMode;
file >> word;
if(word != "ForestBuilt:"){
errorLog << "loadModelFromFile(string filename) - Could not find the ForestBuilt!" << endl;
return false;
}
file >> trained;
if( trained ){
//Find the forest header
file >> word;
if(word != "Forest:"){
errorLog << "loadModelFromFile(string filename) - Could not find the Forest!" << endl;
return false;
}
//Load each tree
UINT treeIndex;
forest.reserve( forestSize );
for(UINT i=0; i<forestSize; i++){
file >> word;
if(word != "Tree:"){
errorLog << "loadModelFromFile(string filename) - Could not find the Tree Header!" << endl;
cout << "WORD: " << word << endl;
cout << "Tree i: " << i << endl;
return false;
}
file >> treeIndex;
if( treeIndex != i+1 ){
errorLog << "loadModelFromFile(string filename) - Incorrect tree index: " << treeIndex << endl;
return false;
}
file >> word;
if(word != "TreeNodeType:"){
errorLog << "loadModelFromFile(string filename) - Could not find the TreeNodeType!" << endl;
cout << "WORD: " << word << endl;
cout << "i: " << i << endl;
return false;
}
file >> treeNodeType;
//Create a new DTree
DecisionTreeNode *tree = dynamic_cast< DecisionTreeNode* >( DecisionTreeNode::createInstanceFromString( treeNodeType ) );
if( tree == NULL ){
errorLog << "loadModelFromFile(fstream &file) - Failed to create new Tree!" << endl;
return false;
}
//Load the tree from the file
tree->setParent( NULL );
if( !tree->loadFromFile( file ) ){
errorLog << "loadModelFromFile(fstream &file) - Failed to load tree from file!" << endl;
return false;
}
//Add the tree to the forest
forest.push_back( tree );
}
}
return true;
}
void DecisionTree::setAntTreeAfraid(){
DecisionTreeNode defaultNode;
m_rootNode = new DecisionTreeNode(1, &defaultNode, "root -> State:VIEW");
DecisionTreeNode* lView = new DecisionTreeNode(2, &defaultNode, "lView -> T:HOME");
DecisionTreeNode* rView = new DecisionTreeNode(3, &defaultNode, "rView -> State:HUNGER");
DecisionTreeNode* lHunger = new DecisionTreeNode(4, &defaultNode, "lHunger -> T:FOOD");
DecisionTreeNode* rHunger = new DecisionTreeNode(5, &defaultNode, "rHunger -> T:HOME");
m_rootNode->setStateID(States::VIEW);
m_rootNode->setLeftNode(lView);
lView->setCondition(true);
lView->setLeaf(true);
lView->setTarget(TreeOutput::HOME);
m_rootNode->setRightNode(rView);
rView->setCondition(false);
rView->setStateID(States::HUNGER);
rView->setLeftNode(lHunger);
lHunger->setCondition(true);
lHunger->setLeaf(true);
lHunger->setTarget(TreeOutput::FOOD);
rView->setRightNode(rHunger);
rHunger->setCondition(false);
rHunger->setLeaf(true);
rHunger->setTarget(TreeOutput::HOME);
m_currentNode = m_rootNode;
}
void DecisionTree::setAntTreeAggressiv(){
DecisionTreeNode defaultNode;
m_rootNode = new DecisionTreeNode(1, &defaultNode);
DecisionTreeNode* lView = new DecisionTreeNode(2, &defaultNode);
DecisionTreeNode* rView = new DecisionTreeNode(3, &defaultNode);
DecisionTreeNode* lHunger = new DecisionTreeNode(4, &defaultNode);
DecisionTreeNode* rHunger = new DecisionTreeNode(5, &defaultNode);
m_rootNode->setStateID(States::VIEW);
m_rootNode->setLeftNode(lView);
lView->setCondition(true);
lView->setLeaf(true);
lView->setTarget(TreeOutput::PLAYER);
m_rootNode->setRightNode(rView);
rView->setCondition(false);
rView->setStateID(States::HUNGER);
rView->setLeftNode(lHunger);
lHunger->setCondition(true);
lHunger->setLeaf(true);
lHunger->setTarget(TreeOutput::FOOD);
rView->setRightNode(rHunger);
rHunger->setCondition(false);
rHunger->setLeaf(true);
rHunger->setTarget(TreeOutput::PATROL);
m_currentNode = m_rootNode;
}
bool RandomForests::loadModelFromFile(fstream &file){
clear();
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_RANDOM_FOREST_MODEL_FILE_V1.0"){
errorLog << "loadModelFromFile(string filename) - Could not find Model File Header" << endl;
return false;
}
file >> word;
if(word != "NumFeatures:"){
errorLog << "loadModelFromFile(string filename) - Could not find NumFeatures!" << endl;
return false;
}
file >> numInputDimensions;
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(numInputDimensions);
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;
}
}
file >> word;
if(word != "ForestSize:"){
errorLog << "loadModelFromFile(string filename) - Could not find the ForestSize!" << endl;
return false;
}
file >> forestSize;
file >> word;
if(word != "NumSplittingSteps:"){
errorLog << "loadModelFromFile(string filename) - Could not find the NumSplittingSteps!" << endl;
return false;
}
file >> numRandomSplits;
file >> word;
if(word != "MinNumSamplesPerNode:"){
errorLog << "loadModelFromFile(string filename) - Could not find the MinNumSamplesPerNode!" << endl;
return false;
}
file >> minNumSamplesPerNode;
file >> word;
if(word != "MaxDepth:"){
errorLog << "loadModelFromFile(string filename) - Could not find the MaxDepth!" << endl;
return false;
}
file >> maxDepth;
file >> word;
if(word != "ForestBuilt:"){
errorLog << "loadModelFromFile(string filename) - Could not find the ForestBuilt!" << endl;
return false;
}
file >> trained;
if( trained ){
file >> word;
if(word != "Forest:"){
errorLog << "loadModelFromFile(string filename) - Could not find the Forest!" << endl;
return false;
}
UINT treeIndex;
for(UINT i=0; i<forestSize; i++){
file >> word;
if(word != "Tree:"){
errorLog << "loadModelFromFile(string filename) - Could not find the Tree Header!" << endl;
return false;
}
file >> treeIndex;
if( treeIndex != i+1 ){
errorLog << "loadModelFromFile(string filename) - Incorrect tree index: " << treeIndex << endl;
return false;
}
//Create a new DTree
DecisionTreeNode *tree = new DecisionTreeNode;
if( tree == NULL ){
errorLog << "loadModelFromFile(fstream &file) - Failed to create new Tree!" << endl;
return false;
}
tree->setParent( NULL );
if( !tree->loadFromFile( file ) ){
errorLog << "loadModelFromFile(fstream &file) - Failed to load tree from file!" << endl;
return false;
}
//Add the tree to the forest
forest.push_back( tree );
}
}
return true;
}
bool computeFeatureWeights( CommandLineParser &parser ){
infoLog << "Computing feature weights..." << endl;
string resultsFilename = "";
string modelFilename = "";
bool combineWeights = false;
//Get the model filename
if( !parser.get("model-filename",modelFilename) ){
errorLog << "Failed to parse filename from command line! You can set the model filename using the --model." << endl;
printUsage();
return false;
}
//Get the results filename
if( !parser.get("filename",resultsFilename) ){
errorLog << "Failed to parse results filename from command line! You can set the results filename using the -f." << endl;
printUsage();
return false;
}
//Get the results filename
parser.get("combine-weights",combineWeights);
//Load the model
GestureRecognitionPipeline pipeline;
if( !pipeline.load( modelFilename ) ){
errorLog << "Failed to load model from file: " << modelFilename << endl;
printUsage();
return false;
}
//Make sure the pipeline contains a random forest model and that it is trained
RandomForests *forest = pipeline.getClassifier< RandomForests >();
if( !forest ){
errorLog << "Model loaded, but the pipeline does not contain a RandomForests classifier!" << endl;
printUsage();
return false;
}
if( !forest->getTrained() ){
errorLog << "Model loaded, but the RandomForests classifier is not trained!" << endl;
printUsage();
return false;
}
//Compute the feature weights
if( combineWeights ){
VectorFloat weights = forest->getFeatureWeights();
if( weights.getSize() == 0 ){
errorLog << "Failed to compute feature weights!" << endl;
printUsage();
return false;
}
//Save the results to a file
fstream file;
file.open( resultsFilename.c_str(), fstream::out );
const unsigned int N = weights.getSize();
for(unsigned int i=0; i<N; i++){
file << weights[i] << endl;
}
file.close();
}else{
double norm = 0.0;
const unsigned int K = forest->getForestSize();
const unsigned int N = forest->getNumInputDimensions();
VectorFloat tmp( N, 0.0 );
MatrixDouble weights(K,N);
for(unsigned int i=0; i<K; i++){
DecisionTreeNode *tree = forest->getTree(i);
tree->computeFeatureWeights( tmp );
norm = 1.0 / Util::sum( tmp );
for(unsigned int j=0; j<N; j++){
tmp[j] *= norm;
weights[i][j] = tmp[j];
tmp[j] = 0;
}
}
//Save the results to a file
weights.save( resultsFilename );
}
return true;
}
