bool ClassificationData::merge(const ClassificationData &labelledData){
if( labelledData.getNumDimensions() != numDimensions ){
errorLog << "merge(const ClassificationData &labelledData) - The number of dimensions in the labelledData (" << labelledData.getNumDimensions() << ") does not match the number of dimensions of this dataset (" << numDimensions << ")" << endl;
return false;
}
//The dataset has changed so flag that any previous cross validation setup will now not work
crossValidationSetup = false;
crossValidationIndexs.clear();
//Reserve the memory
reserve( getNumSamples() + labelledData.getNumSamples() );
//Add the data from the labelledData to this instance
for(UINT i=0; i<labelledData.getNumSamples(); i++){
addSample(labelledData[i].getClassLabel(), labelledData[i].getSample());
}
//Set the class names from the dataset
vector< ClassTracker > classTracker = labelledData.getClassTracker();
for(UINT i=0; i<classTracker.size(); i++){
setClassNameForCorrespondingClassLabel(classTracker[i].className, classTracker[i].classLabel);
}
//Sort the class labels
sortClassLabels();
return true;
}
// Tests the learning algorithm on a basic dataset
TEST(BAG, TrainBasicDataset) {
BAG bag;
//Check the module is not trained
EXPECT_TRUE( !bag.getTrained() );
//Generate a basic dataset
const UINT numSamples = 10000;
const UINT numClasses = 10;
const UINT numDimensions = 100;
ClassificationData::generateGaussDataset( "gauss_data.csv", numSamples, numClasses, numDimensions, 10, 1 );
ClassificationData trainingData;
EXPECT_TRUE( trainingData.load( "gauss_data.csv" ) );
ClassificationData testData = trainingData.split( 50 );
//Add an adaptive naive bayes classifier to the BAG ensemble
bag.addClassifierToEnsemble( ANBC() );
//Add a MinDist classifier to the BAG ensemble, using two clusters
MinDist min_dist_two_clusters;
min_dist_two_clusters.setNumClusters(2);
bag.addClassifierToEnsemble( min_dist_two_clusters );
//Add a MinDist classifier to the BAG ensemble, using five clusters
MinDist min_dist_five_clusters;
min_dist_five_clusters.setNumClusters(5);
bag.addClassifierToEnsemble( min_dist_five_clusters );
//Train the classifier
EXPECT_TRUE( bag.train( trainingData ) );
EXPECT_TRUE( bag.getTrained() );
EXPECT_TRUE( bag.print() );
for(UINT i=0; i<testData.getNumSamples(); i++){
EXPECT_TRUE( bag.predict( testData[i].getSample() ) );
}
EXPECT_TRUE( bag.save( "bag_model.grt" ) );
bag.clear();
EXPECT_TRUE( !bag.getTrained() );
EXPECT_TRUE( bag.load( "bag_model.grt" ) );
EXPECT_TRUE( bag.getTrained() );
for(UINT i=0; i<testData.getNumSamples(); i++){
EXPECT_TRUE( bag.predict( testData[i].getSample() ) );
}
}
bool SVM::convertClassificationDataToLIBSVMFormat(ClassificationData &trainingData){
//clear any previous problems
deleteProblemSet();
const UINT numTrainingExamples = trainingData.getNumSamples();
numInputDimensions = trainingData.getNumDimensions();
//Compute the ranges encase the data should be scaled
ranges = trainingData.getRanges();
//Init the memory
prob.l = numTrainingExamples;
prob.x = new svm_node*[numTrainingExamples];
prob.y = new double[numTrainingExamples];
problemSet = true;
for(UINT i=0; i<numTrainingExamples; i++){
//Set the class ID
prob.y[i] = trainingData[i].getClassLabel();
//Assign the memory for this training example, note that a dummy node is needed at the end of the vector
prob.x[i] = new svm_node[numInputDimensions+1];
for(UINT j=0; j<numInputDimensions; j++){
prob.x[i][j].index = j+1;
prob.x[i][j].value = trainingData[i].getSample()[j];
}
prob.x[i][numInputDimensions].index = -1; //Assign the final node value
prob.x[i][numInputDimensions].value = 0;
}
return true;
}
bool SVM::train_(ClassificationData &trainingData){
//Clear any previous model
clear();
if( trainingData.getNumSamples() == 0 ){
errorLog << "train_(ClassificationData &trainingData) - Training data has zero samples!" << endl;
return false;
}
//Convert the labelled classification data into the LIBSVM data format
if( !convertClassificationDataToLIBSVMFormat(trainingData) ){
errorLog << "train_(ClassificationData &trainingData) - Failed To Convert Labelled Classification Data To LIBSVM Format!" << endl;
return false;
}
if( useAutoGamma ) param.gamma = 1.0/numInputDimensions;
//Train the model
bool trainingResult = trainSVM();
if(! trainingResult ){
errorLog << "train_(ClassificationData &trainingData) - Failed To Train SVM Model!" << endl;
return false;
}
return true;
}
bool BAG::train_(ClassificationData &trainingData){
//Clear any previous models
clear();
const unsigned int M = trainingData.getNumSamples();
const unsigned int N = trainingData.getNumDimensions();
const unsigned int K = trainingData.getNumClasses();
if( M == 0 ){
errorLog << "train_(ClassificationData &trainingData) - Training data has zero samples!" << endl;
return false;
}
numInputDimensions = N;
numClasses = K;
classLabels.resize(K);
ranges = trainingData.getRanges();
//Scale the training data if needed
if( useScaling ){
//Scale the training data between 0 and 1
trainingData.scale(0, 1);
}
UINT ensembleSize = (UINT)ensemble.size();
if( ensembleSize == 0 ){
errorLog << "train_(ClassificationData &trainingData) - The ensemble size is zero! You need to add some classifiers to the ensemble first." << endl;
return false;
}
for(UINT i=0; i<ensembleSize; i++){
if( ensemble[i] == NULL ){
errorLog << "train_(ClassificationData &trainingData) - The classifier at ensemble index " << i << " has not been set!" << endl;
return false;
}
}
//Train the ensemble
for(UINT i=0; i<ensembleSize; i++){
ClassificationData boostedDataset = trainingData.getBootstrappedDataset();
trainingLog << "Training ensemble " << i+1 << ". Ensemble type: " << ensemble[i]->getClassType() << endl;
//Train the classifier with the bootstrapped dataset
if( !ensemble[i]->train( boostedDataset ) ){
errorLog << "train_(ClassificationData &trainingData) - The classifier at ensemble index " << i << " failed training!" << endl;
return false;
}
}
//Set the class labels
classLabels = trainingData.getClassLabels();
//Flag that the model has been trained
trained = true;
return trained;
}
bool SwipeDetector::train_(ClassificationData &trainingData) {
//Clear any previous models
clear();
const unsigned int M = trainingData.getNumSamples();
const unsigned int N = trainingData.getNumDimensions();
if( M == 0 ) {
errorLog << "train_(trainingData &labelledTrainingData) - Training data has zero samples!" << std::endl;
return false;
}
numInputDimensions = N;
numClasses = 2; //This is always 2 for swipe detection [1 == swipe detected, everything else means no swipe detected]
classLabels.resize( 2 );
classLabels[0] = 1; //Swipe
classLabels[1] = 2; //No Swipe
nullRejectionThresholds.resize(2,0);
ranges = trainingData.getRanges();
//Scale the training data if needed
if( useScaling ) {
//Scale the training data between 0 and 1
trainingData.scale(0, 1);
}
//We currently have no way to automatically train the swipe detection, user needs to manually set thresholds, so just flag the model is trained
trained = true;
return true;
}
bool ANBC::setWeights(const ClassificationData &weightsData){
if( weightsData.getNumSamples() > 0 ){
weightsDataSet = true;
this->weightsData = weightsData;
return true;
}
return false;
}
// Tests the learning algorithm on a basic dataset
TEST(KNN, TrainBasicDataset) {
KNN knn;
//Check the module is not trained
EXPECT_TRUE( !knn.getTrained() );
//Generate a basic dataset
const UINT numSamples = 1000;
const UINT numClasses = 10;
const UINT numDimensions = 10;
ClassificationData::generateGaussDataset( "gauss_data.csv", numSamples, numClasses, numDimensions, 10, 1 );
ClassificationData trainingData;
EXPECT_TRUE( trainingData.load( "gauss_data.csv" ) );
ClassificationData testData = trainingData.split( 50 );
//Train the classifier
EXPECT_TRUE( knn.train( trainingData ) );
EXPECT_TRUE( knn.getTrained() );
EXPECT_TRUE( knn.print() );
for(UINT i=0; i<testData.getNumSamples(); i++){
EXPECT_TRUE( knn.predict( testData[i].getSample() ) );
}
EXPECT_TRUE( knn.save( "knn_model.grt" ) );
knn.clear();
EXPECT_TRUE( !knn.getTrained() );
EXPECT_TRUE( knn.load( "knn_model.grt" ) );
EXPECT_TRUE( knn.getTrained() );
for(UINT i=0; i<testData.getNumSamples(); i++){
EXPECT_TRUE( knn.predict( testData[i].getSample() ) );
}
}
bool HierarchicalClustering::train_(ClassificationData &trainingData){
if( trainingData.getNumSamples() == 0 ){
return false;
}
//Convert the labelled training data to a training matrix
M = trainingData.getNumSamples();
N = trainingData.getNumDimensions();
MatrixFloat data(M,N);
for(UINT i=0; i<M; i++){
for(UINT j=0; j<N; j++){
data[i][j] = trainingData[i][j];
}
}
return train_( data );
}
int main (int argc, const char * argv[])
{
//Create a new gesture recognition pipeline
GestureRecognitionPipeline pipeline;
//Add an ANBC module
pipeline.setClassifier( ANBC() );
//Add a ClassLabelFilter as a post processing module with a minCount of 5 and a buffer size of 10
pipeline.addPostProcessingModule( ClassLabelFilter(5,10) );
//Load some training data to train and test the classifier
ClassificationData trainingData;
ClassificationData testData;
if( !trainingData.loadDatasetFromFile("ClassLabelFilterTrainingData.txt") ){
cout << "Failed to load training data!\n";
return EXIT_FAILURE;
}
if( !testData.loadDatasetFromFile("ClassLabelFilterTestData.txt") ){
cout << "Failed to load training data!\n";
return EXIT_FAILURE;
}
//Train the classifier
if( !pipeline.train( trainingData ) ){
cout << "Failed to train classifier!\n";
return EXIT_FAILURE;
}
//Use the test dataset to demonstrate the output of the ClassLabelFilter
for(UINT i=0; i<testData.getNumSamples(); i++){
VectorDouble inputVector = testData[i].getSample();
if( !pipeline.predict( inputVector ) ){
cout << "Failed to perform prediction for test sampel: " << i <<"\n";
return EXIT_FAILURE;
}
//Get the predicted class label (this will be the processed class label)
UINT predictedClassLabel = pipeline.getPredictedClassLabel();
//Get the unprocessed class label (i.e. the direct output of the classifier)
UINT unprocessedClassLabel = pipeline.getUnProcessedPredictedClassLabel();
//Also print the results to the screen
cout << "Processed Class Label: \t" << predictedClassLabel << "\tUnprocessed Class Label: \t" << unprocessedClassLabel << endl;
}
return EXIT_SUCCESS;
}
int main (int argc, const char * argv[])
{
//Load the example data
ClassificationData data;
if( !data.loadDatasetFromFile("WiiAccShakeData.txt") ){
cout << "ERROR: Failed to load data from file!\n";
return EXIT_FAILURE;
}
//The variables used to initialize the zero crossing counter feature extraction
UINT searchWindowSize = 20;
double deadZoneThreshold = 0.01;
UINT numDimensions = data.getNumDimensions();
UINT featureMode = ZeroCrossingCounter::INDEPENDANT_FEATURE_MODE; //This could also be ZeroCrossingCounter::COMBINED_FEATURE_MODE
//Create a new instance of the ZeroCrossingCounter feature extraction
ZeroCrossingCounter zeroCrossingCounter(searchWindowSize,deadZoneThreshold,numDimensions,featureMode);
//Loop over the accelerometer data, at each time sample (i) compute the features using the new sample and then write the results to a file
for(UINT i=0; i<data.getNumSamples(); i++){
//Compute the features using this new sample
zeroCrossingCounter.computeFeatures( data[i].getSample() );
//Write the data to the file
cout << "InputVector: ";
for(UINT j=0; j<data.getNumDimensions(); j++){
cout << data[i].getSample()[j] << "\t";
}
//Get the latest feature vector
VectorDouble featureVector = zeroCrossingCounter.getFeatureVector();
//Write the features to the file
cout << "FeatureVector: ";
for(UINT j=0; j<featureVector.size(); j++){
cout << featureVector[j];
if( j != featureVector.size()-1 ) cout << "\t";
}
cout << endl;
}
//Save the zero crossing counter settings to a file
zeroCrossingCounter.saveModelToFile("ZeroCrossingCounterSettings.txt");
//You can then load the settings again if you need them
zeroCrossingCounter.loadModelFromFile("ZeroCrossingCounterSettings.txt");
return EXIT_SUCCESS;
}
bool DecisionTreeClusterNode::computeSplit( const UINT &numSplittingSteps, const ClassificationData &trainingData, const Vector< UINT > &features, const Vector< UINT > &classLabels, UINT &featureIndex, Float &minError ){
const UINT M = trainingData.getNumSamples();
const UINT N = features.getSize();
const UINT K = classLabels.getSize();
if( N == 0 ) return false;
if( K == 0 ) return false;
minError = grt_numeric_limits< Float >::max();
Random random;
UINT bestFeatureIndex = 0;
Float bestThreshold = 0;
Float error = 0;
Vector< UINT > groupIndex(M);
Vector< MinMax > ranges = trainingData.getRanges();
MatrixDouble data(M,1); //This will store our temporary data for each dimension
//Randomly select which features we want to use
UINT numRandomFeatures = numSplittingSteps > N ? N : numSplittingSteps;
Vector< UINT > randomFeatures = random.getRandomSubset( 0, N, numRandomFeatures );
//Loop over each random feature and try and find the best split point
for(UINT n=0; n<numRandomFeatures; n++){
featureIndex = features[ randomFeatures[n] ];
//Use the data in this feature dimension to create a sum dataset
for(UINT i=0; i<M; i++){
data[i][0] = trainingData[i][featureIndex];
}
if( computeError( trainingData, data, classLabels, ranges, groupIndex, featureIndex, threshold, error ) ){
//Store the best threshold and feature index
if( error < minError ){
minError = error;
bestThreshold = threshold;
bestFeatureIndex = featureIndex;
}
}
}
//Set the best feature index that will be returned to the DecisionTree that called this function
featureIndex = bestFeatureIndex;
//Store the node size, feature index, best threshold and class probabilities for this node
set( M, featureIndex, bestThreshold, trainingData.getClassProbabilities(classLabels) );
return true;
}
bool KMeans::train_(ClassificationData &trainingData){
if( trainingData.getNumSamples() == 0 ){
errorLog << "train_(ClassificationData &trainingData) - The training data is empty!" << std::endl;
return false;
}
//Set the numClusters as the number of classes in the training data
numClusters = trainingData.getNumClasses();
//Convert the labelled training data to a training matrix
UINT M = trainingData.getNumSamples();
UINT N = trainingData.getNumDimensions();
MatrixFloat data(M,N);
for(UINT i=0; i<M; i++){
for(UINT j=0; j<N; j++){
data[i][j] = trainingData[i][j];
}
}
//Run the K-Means algorithm
return train_( data );
}
int main (int argc, const char * argv[])
{
//Load the example data
ClassificationData data;
if( !data.load("WiiAccShakeData.grt") ){
cout << "ERROR: Failed to load data from file!\n";
return EXIT_FAILURE;
}
//The variables used to initialize the MovementIndex feature extraction
UINT windowSize = 10;
UINT numDimensions = data.getNumDimensions();
//Create a new instance of the MovementIndex feature extraction
MovementIndex movementIndex(windowSize,numDimensions);
//Loop over the accelerometer data, at each time sample (i) compute the features using the new sample and then write the results to a file
for(UINT i=0; i<data.getNumSamples(); i++){
//Compute the features using this new sample
movementIndex.computeFeatures( data[i].getSample() );
//Write the data
cout << "InputVector: ";
for(UINT j=0; j<data.getNumDimensions(); j++){
cout << data[i].getSample()[j] << "\t";
}
//Get the latest feature vector
VectorFloat featureVector = movementIndex.getFeatureVector();
//Write the features
cout << "FeatureVector: ";
for(UINT j=0; j<featureVector.size(); j++){
cout << featureVector[j];
if( j != featureVector.size()-1 ) cout << "\t";
}
cout << endl;
}
//Save the MovementIndex settings to a file
movementIndex.save("MovementIndexSettings.grt");
//You can then load the settings again if you need them
movementIndex.load("MovementIndexSettings.grt");
return EXIT_SUCCESS;
}
bool Softmax::train_(ClassificationData &trainingData){
//Clear any previous model
clear();
const unsigned int M = trainingData.getNumSamples();
const unsigned int N = trainingData.getNumDimensions();
const unsigned int K = trainingData.getNumClasses();
if( M == 0 ){
errorLog << "train_(ClassificationData &labelledTrainingData) - Training data has zero samples!" << std::endl;
return false;
}
numInputDimensions = N;
numClasses = K;
models.resize(K);
classLabels.resize(K);
ranges = trainingData.getRanges();
//Scale the training data if needed
if( useScaling ){
//Scale the training data between 0 and 1
trainingData.scale(0, 1);
}
//Train a regression model for each class in the training data
for(UINT k=0; k<numClasses; k++){
//Set the class label
classLabels[k] = trainingData.getClassTracker()[k].classLabel;
//Train the model
if( !trainSoftmaxModel(classLabels[k],models[k],trainingData) ){
errorLog << "train(ClassificationData labelledTrainingData) - Failed to train model for class: " << classLabels[k] << std::endl;
return false;
}
}
//Flag that the algorithm has been trained
trained = true;
return trained;
}
int main(int argc, char * argv[])
{
if( argc < 3 ){
errorLog << "Not enough input arguments!" << endl;
printUsage();
return EXIT_FAILURE;
}
const string inputDirectory = argv[1];
const string outputFilename = argv[2];
//Parse the data directory for files
vector< string > filenames;
infoLog << "- Parsing data directory: " << inputDirectory << endl;
if( !Util::parseDirectory( inputDirectory, ".csv", filenames ) ){
errorLog << "Failed to parse input directory: " << inputDirectory << endl;
return EXIT_FAILURE;
}
if( filenames.size() == 0 ){
errorLog << "Failed to find any files in the input directory: " << inputDirectory << endl;
return EXIT_FAILURE;
}
ClassificationData data;
unsigned int numFiles = (unsigned int)filenames.size();
bool dataLoaded = false;
for(unsigned int i=0; i<numFiles; i++){
//Load the data
infoLog << "- Loading data " << i+1 << " of " << numFiles << endl;
ClassificationData tmp;
if( tmp.load( filenames[i] ) ){
if( i==0 ){
data.setNumDimensions( tmp.getNumDimensions() );
}
dataLoaded = true;
infoLog << "- Data loaded. Number of samples: " << tmp.getNumSamples() << endl;
data.merge( tmp );
}else{
warningLog << "- Failed to load data!" << endl;
}
}
if( dataLoaded ){
infoLog << "- Merged data to generate new dataset with " << data.getNumSamples() << " samples" << endl;
//Save the new datasets
infoLog << "- Saving main dataset to file: " << outputFilename << endl;
if( !data.save( outputFilename ) ){
errorLog << "Failed to save output data: " << outputFilename << endl;
return EXIT_FAILURE;
}
}else{
warningLog << "- Failed to load any data!" << endl;
return EXIT_FAILURE;
}
return EXIT_SUCCESS;
}
int main (int argc, const char * argv[])
{
//Create a new Softmax instance
Softmax softmax;
//Load some training data to train the classifier
ClassificationData trainingData;
if( !trainingData.loadDatasetFromFile("SoftmaxTrainingData.txt") ){
cout << "Failed to load training data!\n";
return EXIT_FAILURE;
}
//Use 20% of the training dataset to create a test dataset
ClassificationData testData = trainingData.partition( 80 );
//Train the classifier
if( !softmax.train( trainingData ) ){
cout << "Failed to train classifier!\n";
return EXIT_FAILURE;
}
//Save the Softmax model to a file
if( !softmax.saveModelToFile("SoftmaxModel.txt") ){
cout << "Failed to save the classifier model!\n";
return EXIT_FAILURE;
}
//Load the Softmax model from a file
if( !softmax.loadModelFromFile("SoftmaxModel.txt") ){
cout << "Failed to load the classifier model!\n";
return EXIT_FAILURE;
}
//Use the test dataset to test the softmax model
double accuracy = 0;
for(UINT i=0; i<testData.getNumSamples(); i++){
//Get the i'th test sample
UINT classLabel = testData[i].getClassLabel();
vector< double > inputVector = testData[i].getSample();
//Perform a prediction using the classifier
if( !softmax.predict( inputVector ) ){
cout << "Failed to perform prediction for test sample: " << i <<"\n";
return EXIT_FAILURE;
}
//Get the predicted class label
UINT predictedClassLabel = softmax.getPredictedClassLabel();
vector< double > classLikelihoods = softmax.getClassLikelihoods();
vector< double > classDistances = softmax.getClassDistances();
//Update the accuracy
if( classLabel == predictedClassLabel ) accuracy++;
cout << "TestSample: " << i << " ClassLabel: " << classLabel << " PredictedClassLabel: " << predictedClassLabel << endl;
}
cout << "Test Accuracy: " << accuracy/double(testData.getNumSamples())*100.0 << "%" << endl;
return EXIT_SUCCESS;
}
bool DecisionTreeClusterNode::computeError( const ClassificationData &trainingData, MatrixFloat &data, const Vector< UINT > &classLabels, Vector< MinMax > ranges, Vector< UINT > groupIndex, const UINT featureIndex, Float &threshold, Float &error ){
error = 0;
threshold = 0;
const UINT M = trainingData.getNumSamples();
const UINT K = (UINT)classLabels.size();
Float giniIndexL = 0;
Float giniIndexR = 0;
Float weightL = 0;
Float weightR = 0;
VectorFloat groupCounter(2,0);
MatrixFloat classProbabilities(K,2);
//Use this data to train a KMeans cluster with 2 clusters
KMeans kmeans;
kmeans.setNumClusters( 2 );
kmeans.setComputeTheta( true );
kmeans.setMinChange( 1.0e-5 );
kmeans.setMinNumEpochs( 1 );
kmeans.setMaxNumEpochs( 100 );
//Disable the logging to clean things up
kmeans.setTrainingLoggingEnabled( false );
if( !kmeans.train_( data ) ){
errorLog << __GRT_LOG__ << " Failed to train KMeans model for feature: " << featureIndex << std::endl;
return false;
}
//Set the split threshold as the mid point between the two clusters
const MatrixFloat &clusters = kmeans.getClusters();
threshold = 0;
for(UINT i=0; i<clusters.getNumRows(); i++){
threshold += clusters[i][0];
}
threshold /= clusters.getNumRows();
//Iterate over each sample and work out if it should be in the lhs (0) or rhs (1) group based on the current threshold
groupCounter[0] = groupCounter[1] = 0;
classProbabilities.setAllValues(0);
for(UINT i=0; i<M; i++){
groupIndex[i] = trainingData[ i ][ featureIndex ] >= threshold ? 1 : 0;
groupCounter[ groupIndex[i] ]++;
classProbabilities[ getClassLabelIndexValue(trainingData[i].getClassLabel(),classLabels) ][ groupIndex[i] ]++;
}
//Compute the class probabilities for the lhs group and rhs group
for(UINT k=0; k<K; k++){
classProbabilities[k][0] = groupCounter[0]>0 ? classProbabilities[k][0]/groupCounter[0] : 0;
classProbabilities[k][1] = groupCounter[1]>0 ? classProbabilities[k][1]/groupCounter[1] : 0;
}
//Compute the Gini index for the lhs and rhs groups
giniIndexL = giniIndexR = 0;
for(UINT k=0; k<K; k++){
giniIndexL += classProbabilities[k][0] * (1.0-classProbabilities[k][0]);
giniIndexR += classProbabilities[k][1] * (1.0-classProbabilities[k][1]);
}
weightL = groupCounter[0]/M;
weightR = groupCounter[1]/M;
error = (giniIndexL*weightL) + (giniIndexR*weightR);
return true;
}
bool MinDist::train_(ClassificationData &labelledTrainingData){
//Clear any previous models
clear();
const unsigned int M = labelledTrainingData.getNumSamples();
const unsigned int N = labelledTrainingData.getNumDimensions();
const unsigned int K = labelledTrainingData.getNumClasses();
if( M == 0 ){
errorLog << "train_(ClassificationData &labelledTrainingData) - Training data has zero samples!" << endl;
return false;
}
if( M <= numClusters ){
errorLog << "train_(ClassificationData &labelledTrainingData) - There are not enough training samples for the number of clusters. Either reduce the number of clusters or increase the number of training samples!" << endl;
return false;
}
numInputDimensions = N;
numClasses = K;
models.resize(K);
classLabels.resize(K);
nullRejectionThresholds.resize(K);
ranges = labelledTrainingData.getRanges();
//Scale the training data if needed
if( useScaling ){
//Scale the training data between 0 and 1
labelledTrainingData.scale(0, 1);
}
//Train each of the models
for(UINT k=0; k<numClasses; k++){
//Get the class label for the kth class
UINT classLabel = labelledTrainingData.getClassTracker()[k].classLabel;
//Set the kth class label
classLabels[k] = classLabel;
//Get all the training data for this class
ClassificationData classData = labelledTrainingData.getClassData(classLabel);
MatrixDouble data(classData.getNumSamples(),N);
//Copy the training data into a matrix
for(UINT i=0; i<data.getNumRows(); i++){
for(UINT j=0; j<data.getNumCols(); j++){
data[i][j] = classData[i][j];
}
}
//Train the model for this class
models[k].setGamma( nullRejectionCoeff );
if( !models[k].train(classLabel,data,numClusters) ){
errorLog << "train_(ClassificationData &labelledTrainingData) - Failed to train model for class: " << classLabel;
errorLog << ". This is might be because this class does not have enough training samples! You should reduce the number of clusters or increase the number of training samples for this class." << endl;
models.clear();
return false;
}
//Set the null rejection threshold
nullRejectionThresholds[k] = models[k].getRejectionThreshold();
}
trained = true;
return true;
}
int main (int argc, const char * argv[])
{
//Create a new AdaBoost instance
AdaBoost adaBoost;
//Set the weak classifier you want to use
adaBoost.setWeakClassifier( DecisionStump() );
//Load some training data to train the classifier
ClassificationData trainingData;
if( !trainingData.load("AdaBoostTrainingData.grt") ){
cout << "Failed to load training data!\n";
return EXIT_FAILURE;
}
//Use 20% of the training dataset to create a test dataset
ClassificationData testData = trainingData.partition( 80 );
//Train the classifier
if( !adaBoost.train( trainingData ) ){
cout << "Failed to train classifier!\n";
return EXIT_FAILURE;
}
//Save the model to a file
if( !adaBoost.save("AdaBoostModel.grt") ){
cout << "Failed to save the classifier model!\n";
return EXIT_FAILURE;
}
//Load the model from a file
if( !adaBoost.load("AdaBoostModel.grt") ){
cout << "Failed to load the classifier model!\n";
return EXIT_FAILURE;
}
//Use the test dataset to test the AdaBoost model
double accuracy = 0;
for(UINT i=0; i<testData.getNumSamples(); i++){
//Get the i'th test sample
UINT classLabel = testData[i].getClassLabel();
vector< double > inputVector = testData[i].getSample();
//Perform a prediction using the classifier
if( !adaBoost.predict( inputVector ) ){
cout << "Failed to perform prediction for test sampel: " << i <<"\n";
return EXIT_FAILURE;
}
//Get the predicted class label
UINT predictedClassLabel = adaBoost.getPredictedClassLabel();
double maximumLikelhood = adaBoost.getMaximumLikelihood();
vector< double > classLikelihoods = adaBoost.getClassLikelihoods();
vector< double > classDistances = adaBoost.getClassDistances();
//Update the accuracy
if( classLabel == predictedClassLabel ) accuracy++;
cout << "TestSample: " << i << " ClassLabel: " << classLabel;
cout << " PredictedClassLabel: " << predictedClassLabel << " Likelihood: " << maximumLikelhood;
cout << endl;
}
cout << "Test Accuracy: " << accuracy/double(testData.getNumSamples())*100.0 << "%" << endl;
return EXIT_SUCCESS;
}
void prediction_axis_data(){
// Training and test data
ClassificationData trainingData;
ClassificationData testData;
string file_path = "../../../data/";
string class_name = "5";
if( !trainingData.loadDatasetFromFile(file_path + "train/grt/" + class_name + ".txt") ){
std::cout <<"Failed to load training data!\n";
}
if( !testData.loadDatasetFromFile(file_path + "test/grt/" + class_name + ".txt") ){
std::cout <<"Failed to load training data!\n";
}
// Pipeline setup
ANBC anbc;
anbc.setNullRejectionCoeff(1);
anbc.enableScaling(true);
anbc.enableNullRejection(true);
GestureRecognitionPipeline pipeline;
pipeline.setClassifier(anbc);
// Train the pipeline
if( !pipeline.train( trainingData ) ){
std::cout << "Failed to train classifier!\n";
}
// File stream
ofstream outputFileStream(class_name + ".csv");
// Evaluation
double accuracy = 0;
outputFileStream << "actualClass,predictedClass,maximumLikelihood,lZ,lY,lZ,rZ,rY,rZ \n";
for(UINT i=0; i<testData.getNumSamples(); i++){
UINT actualClassLabel = testData[i].getClassLabel();
vector< double > inputVector = testData[i].getSample();
if( !pipeline.predict( inputVector )){
std::cout << "Failed to perform prediction for test sampel: " << i <<"\n";
}
UINT predictedClassLabel = pipeline.getPredictedClassLabel();
double maximumLikelihood = pipeline.getMaximumLikelihood();
outputFileStream << actualClassLabel << "," << predictedClassLabel << "," << maximumLikelihood << ","
<< inputVector[0] << "," << inputVector[1] << "," << inputVector[2] << "," << inputVector[3] << "," << inputVector[4] << "," << inputVector[5] << "\n";
if( actualClassLabel == predictedClassLabel) accuracy++;
}
std::cout << "Test Accuracy testHandsUp : " << accuracy/double(testData.getNumSamples())*100.0 << " %\n";
}
bool Softmax::trainSoftmaxModel(UINT classLabel,SoftmaxModel &model,ClassificationData &data){
Float error = 0;
Float errorSum = 0;
Float lastErrorSum = 0;
Float delta = 0;
UINT N = data.getNumDimensions();
UINT M = data.getNumSamples();
UINT iter = 0;
bool keepTraining = true;
Random random;
VectorFloat y(M);
Vector< UINT > randomTrainingOrder(M);
//Init the model
model.init( classLabel, N );
//Setup the target vector, the input data is relabelled as positive samples (with label 1.0) and negative samples (with label 0.0)
for(UINT i=0; i<M; i++){
y[i] = data[i].getClassLabel()==classLabel ? 1.0 : 0;
}
//In most cases, the training data is grouped into classes (100 samples for class 1, followed by 100 samples for class 2, etc.)
//This can cause a problem for stochastic gradient descent algorithm. To avoid this issue, we randomly shuffle the order of the
//training samples. This random order is then used at each epoch.
for(UINT i=0; i<M; i++){
randomTrainingOrder[i] = i;
}
std::random_shuffle(randomTrainingOrder.begin(), randomTrainingOrder.end());
//Run the main stochastic gradient descent training algorithm
while( keepTraining ){
//Run one epoch of training using stochastic gradient descent
errorSum = 0;
for(UINT m=0; m<M; m++){
//Select the random sample
UINT i = randomTrainingOrder[m];
//Compute the error, given the current weights
error = y[i] - model.compute( data[i].getSample() );
errorSum += error;
//Update the weights
for(UINT j=0; j<N; j++){
model.w[j] += learningRate * error * data[i][j];
}
model.w0 += learningRate * error;
}
//Compute the error
delta = fabs( errorSum-lastErrorSum );
lastErrorSum = errorSum;
//Check to see if we should stop
if( delta <= minChange ){
keepTraining = false;
}
if( ++iter >= maxNumEpochs ){
keepTraining = false;
}
trainingLog << "Epoch: " << iter << " TotalError: " << errorSum << " Delta: " << delta << std::endl;
}
return true;
}
bool AdaBoost::train_(ClassificationData &trainingData){
//Clear any previous model
clear();
if( trainingData.getNumSamples() <= 1 ){
errorLog << "train_(ClassificationData &trainingData) - There are not enough training samples to train a model! Number of samples: " << trainingData.getNumSamples() << endl;
return false;
}
numInputDimensions = trainingData.getNumDimensions();
numClasses = trainingData.getNumClasses();
const UINT M = trainingData.getNumSamples();
const UINT POSITIVE_LABEL = WEAK_CLASSIFIER_POSITIVE_CLASS_LABEL;
const UINT NEGATIVE_LABEL = WEAK_CLASSIFIER_NEGATIVE_CLASS_LABEL;
double alpha = 0;
const double beta = 0.001;
double epsilon = 0;
TrainingResult trainingResult;
const UINT K = (UINT)weakClassifiers.size();
if( K == 0 ){
errorLog << "train_(ClassificationData &trainingData) - No weakClassifiers have been set. You need to set at least one weak classifier first." << endl;
return false;
}
classLabels.resize(numClasses);
models.resize(numClasses);
ranges = trainingData.getRanges();
//Scale the training data if needed
if( useScaling ){
trainingData.scale(ranges,0,1);
}
//Create the weights vector
VectorDouble weights(M);
//Create the error matrix
MatrixDouble errorMatrix(K,M);
for(UINT classIter=0; classIter<numClasses; classIter++){
//Get the class label for the current class
classLabels[classIter] = trainingData.getClassLabels()[classIter];
//Set the class label of the current model
models[ classIter ].setClassLabel( classLabels[classIter] );
//Setup the labels for this class, POSITIVE_LABEL == 1, NEGATIVE_LABEL == 2
ClassificationData classData;
classData.setNumDimensions(trainingData.getNumDimensions());
for(UINT i=0; i<M; i++){
UINT label = trainingData[i].getClassLabel()==classLabels[classIter] ? POSITIVE_LABEL : NEGATIVE_LABEL;
VectorDouble trainingSample = trainingData[i].getSample();
classData.addSample(label,trainingSample);
}
//Setup the initial training sample weights
std::fill(weights.begin(),weights.end(),1.0/M);
//Run the boosting loop
bool keepBoosting = true;
UINT t = 0;
while( keepBoosting ){
//Pick the classifier from the family of classifiers that minimizes the total error
UINT bestClassifierIndex = 0;
double minError = numeric_limits<double>::max();
for(UINT k=0; k<K; k++){
//Get the k'th possible classifier
WeakClassifier *weakLearner = weakClassifiers[k];
//Train the current classifier
if( !weakLearner->train(classData,weights) ){
errorLog << "Failed to train weakLearner!" << endl;
return false;
}
//Compute the weighted error for this clasifier
double e = 0;
double positiveLabel = weakLearner->getPositiveClassLabel();
double numCorrect = 0;
double numIncorrect = 0;
for(UINT i=0; i<M; i++){
//Only penalize errors
double prediction = weakLearner->predict( classData[i].getSample() );
if( (prediction == positiveLabel && classData[i].getClassLabel() != POSITIVE_LABEL) || //False positive
(prediction != positiveLabel && classData[i].getClassLabel() == POSITIVE_LABEL) ){ //False negative
e += weights[i]; //Increase the error proportional to the weight of the example
errorMatrix[k][i] = 1; //Flag that there was an error
numIncorrect++;
}else{
errorMatrix[k][i] = 0; //Flag that there was no error
numCorrect++;
}
}
trainingLog << "PositiveClass: " << classLabels[classIter] << " Boosting Iter: " << t << " Classifier: " << k << " WeightedError: " << e << " NumCorrect: " << numCorrect/M << " NumIncorrect: " <<numIncorrect/M << endl;
if( e < minError ){
minError = e;
bestClassifierIndex = k;
}
}
epsilon = minError;
//Set alpha, using the M1 weight value, small weights (close to 0) will receive a strong weight in the final classifier
alpha = 0.5 * log( (1.0-epsilon)/epsilon );
trainingLog << "PositiveClass: " << classLabels[classIter] << " Boosting Iter: " << t << " Best Classifier Index: " << bestClassifierIndex << " MinError: " << minError << " Alpha: " << alpha << endl;
if( isinf(alpha) ){ keepBoosting = false; trainingLog << "Alpha is INF. Stopping boosting for current class" << endl; }
if( 0.5 - epsilon <= beta ){ keepBoosting = false; trainingLog << "Epsilon <= Beta. Stopping boosting for current class" << endl; }
if( ++t >= numBoostingIterations ) keepBoosting = false;
trainingResult.setClassificationResult(t, minError, this);
trainingResults.push_back(trainingResult);
trainingResultsObserverManager.notifyObservers( trainingResult );
if( keepBoosting ){
//Add the best weak classifier to the committee
models[ classIter ].addClassifierToCommitee( weakClassifiers[bestClassifierIndex], alpha );
//Update the weights for the next boosting iteration
double reWeight = (1.0 - epsilon) / epsilon;
double oldSum = 0;
double newSum = 0;
for(UINT i=0; i<M; i++){
oldSum += weights[i];
//Only update the weights that resulted in an incorrect prediction
if( errorMatrix[bestClassifierIndex][i] == 1 ) weights[i] *= reWeight;
newSum += weights[i];
}
//Normalize all the weights
//This results to increasing the weights of the samples that were incorrectly labelled
//While decreasing the weights of the samples that were correctly classified
reWeight = oldSum/newSum;
for(UINT i=0; i<M; i++){
weights[i] *= reWeight;
}
}else{
trainingLog << "Stopping boosting training at iteration : " << t-1 << " with an error of " << epsilon << endl;
if( t-1 == 0 ){
//Add the best weak classifier to the committee (we have to add it as this is the first iteration)
if( isinf(alpha) ){ alpha = 1; } //If alpha is infinite then the first classifier got everything correct
models[ classIter ].addClassifierToCommitee( weakClassifiers[bestClassifierIndex], alpha );
}
}
}
}
//Normalize the weights
for(UINT k=0; k<numClasses; k++){
models[k].normalizeWeights();
}
//Flag that the model has been trained
trained = true;
//Setup the data for prediction
predictedClassLabel = 0;
maxLikelihood = 0;
classLikelihoods.resize(numClasses);
classDistances.resize(numClasses);
return true;
}
bool ANBC::train_(ClassificationData &labelledTrainingData){
//Clear any previous model
clear();
const unsigned int M = labelledTrainingData.getNumSamples();
const unsigned int N = labelledTrainingData.getNumDimensions();
const unsigned int K = labelledTrainingData.getNumClasses();
if( M == 0 ){
errorLog << "train_(ClassificationData &labelledTrainingData) - Training data has zero samples!" << endl;
return false;
}
if( weightsDataSet ){
if( weightsData.getNumDimensions() != N ){
errorLog << "train_(ClassificationData &labelledTrainingData) - The number of dimensions in the weights data (" << weightsData.getNumDimensions() << ") is not equal to the number of dimensions of the training data (" << N << ")" << endl;
return false;
}
}
numInputDimensions = N;
numClasses = K;
models.resize(K);
classLabels.resize(K);
ranges = labelledTrainingData.getRanges();
//Scale the training data if needed
if( useScaling ){
//Scale the training data between 0 and 1
labelledTrainingData.scale(0, 1);
}
//Train each of the models
for(UINT k=0; k<numClasses; k++){
//Get the class label for the kth class
UINT classLabel = labelledTrainingData.getClassTracker()[k].classLabel;
//Set the kth class label
classLabels[k] = classLabel;
//Get the weights for this class
VectorDouble weights(numInputDimensions);
if( weightsDataSet ){
bool weightsFound = false;
for(UINT i=0; i<weightsData.getNumSamples(); i++){
if( weightsData[i].getClassLabel() == classLabel ){
weights = weightsData[i].getSample();
weightsFound = true;
break;
}
}
if( !weightsFound ){
errorLog << "train_(ClassificationData &labelledTrainingData) - Failed to find the weights for class " << classLabel << endl;
return false;
}
}else{
//If the weights data has not been set then all the weights are 1
for(UINT j=0; j<numInputDimensions; j++) weights[j] = 1.0;
}
//Get all the training data for this class
ClassificationData classData = labelledTrainingData.getClassData(classLabel);
MatrixDouble data(classData.getNumSamples(),N);
//Copy the training data into a matrix
for(UINT i=0; i<data.getNumRows(); i++){
for(UINT j=0; j<data.getNumCols(); j++){
data[i][j] = classData[i][j];
}
}
//Train the model for this class
models[k].gamma = nullRejectionCoeff;
if( !models[k].train(classLabel,data,weights) ){
errorLog << "train_(ClassificationData &labelledTrainingData) - Failed to train model for class: " << classLabel << endl;
//Try and work out why the training failed
if( models[k].N == 0 ){
errorLog << "train_(ClassificationData &labelledTrainingData) - N == 0!" << endl;
models.clear();
return false;
}
for(UINT j=0; j<numInputDimensions; j++){
if( models[k].mu[j] == 0 ){
errorLog << "train_(ClassificationData &labelledTrainingData) - The mean of column " << j+1 << " is zero! Check the training data" << endl;
models.clear();
return false;
}
}
models.clear();
return false;
}
}
//Store the null rejection thresholds
nullRejectionThresholds.resize(numClasses);
for(UINT k=0; k<numClasses; k++) {
nullRejectionThresholds[k] = models[k].threshold;
}
//Flag that the models have been trained
trained = true;
return trained;
}
bool Softmax::trainSoftmaxModel(UINT classLabel,SoftmaxModel &model,ClassificationData &data){
Float error = 0;
Float errorSum = 0;
Float lastErrorSum = 0;
Float delta = 0;
const UINT N = data.getNumDimensions();
const UINT M = data.getNumSamples();
UINT iter = 0;
bool keepTraining = true;
Random random;
VectorFloat y(M);
VectorFloat batchMean(N);
Vector< UINT > randomTrainingOrder(M);
Vector< VectorFloat > batchData(batchSize,VectorFloat(N));
//Init the model
model.init( classLabel, N );
//Setup the target vector, the input data is relabelled as positive samples (with label 1.0) and negative samples (with label 0.0)
for(UINT i=0; i<M; i++){
y[i] = data[i].getClassLabel()==classLabel ? 1.0 : 0;
}
//In most cases, the training data is grouped into classes (100 samples for class 1, followed by 100 samples for class 2, etc.)
//This can cause a problem for stochastic gradient descent algorithm. To avoid this issue, we randomly shuffle the order of the
//training samples. This random order is then used at each epoch.
for(UINT i=0; i<M; i++){
randomTrainingOrder[i] = i;
}
std::random_shuffle(randomTrainingOrder.begin(), randomTrainingOrder.end());
//Clear any previous training results
trainingResults.clear();
trainingResults.reserve( maxNumEpochs );
TrainingResult epochResult;
//Run the main stochastic gradient descent training algorithm
while( keepTraining ){
//Run one epoch of training using stochastic gradient descent
errorSum = 0;
UINT m=0;
while( m < M ){
//Get the batch data for this update
UINT roundSize = m+batchSize < M ? batchSize : M-m;
batchMean.fill(0.0);
for(UINT i=0; i<roundSize; i++){
for(UINT j=0; j<N; j++){
batchData[i][j] = data[ randomTrainingOrder[m+i] ][j];
batchMean[j] += batchData[i][j];
}
}
for(UINT j=0; j<N; j++) batchMean[j] /= roundSize;
//Compute the error on this batch, given the current weights
error = 0.0;
for(UINT i=0; i<roundSize; i++){
error += y[ randomTrainingOrder[m+i] ] - model.compute( batchData[i] );
}
error /= roundSize;
errorSum += error;
//Update the weights
for(UINT j=0; j<N; j++){
model.w[j] += learningRate * error * batchMean[j];
}
model.w0 += learningRate * error;
m += roundSize;
}
//Compute the error
delta = fabs( errorSum-lastErrorSum );
lastErrorSum = errorSum;
//Check to see if we should stop
if( delta <= minChange ){
keepTraining = false;
}
if( ++iter >= maxNumEpochs ){
keepTraining = false;
}
trainingLog << "Class: " << classLabel << " Epoch: " << iter << " TotalError: " << errorSum << " Delta: " << delta << std::endl;
epochResult.setClassificationResult( iter, errorSum, this );
trainingResults.push_back( epochResult );
}
return true;
}
void metrics_subset_data(){
ANBC anbc;
anbc.enableScaling(true);
anbc.enableNullRejection(true);
MinDist minDist;
minDist.setNumClusters(4);
minDist.enableScaling(true);
minDist.enableNullRejection(true);
// ofstream opRecall("anbc-recall-nr-0-10.csv");
// opRecall <<"nrCoeff,class0,class1,class2,class3,class4,class5\n";
//
// ofstream opInstanceRes("anbc-prediction-nr-2.csv");
// opInstanceRes <<"actualClass,predictedClass,maximumLikelihood,lZ,lY,lZ,rZ,rY,rZ\n";
//
// ofstream opMetrics("anbc-precision-recall-fmeasure-nr-2.csv");
// opMetrics <<"class1,class2,class3,class4,class5\n";
//
// ofstream opConfusion("anbc-confusion-nr-2.csv");
// opConfusion <<"class0,class1,class2,class3,class4,class5\n";
ofstream opRecall("mindist-recall-nr-0-10.csv");
opRecall <<"nrCoeff,class0,class1,class2,class3,class4,class5\n";
ofstream opInstanceRes("mindist-prediction-nr-2.csv");
opInstanceRes <<"actualClass,predictedClass,maximumLikelihood,lZ,lY,lZ,rZ,rY,rZ\n";
ofstream opMetrics("mindist-precision-recall-fmeasure-nr-2.csv");
opMetrics <<"class1,class2,class3,class4,class5\n";
ofstream opConfusion("mindist-confusion-nr-2.csv");
opConfusion <<"class0,class1,class2,class3,class4,class5\n";
// Training and test data
ClassificationData trainingData;
ClassificationData testData;
ClassificationData nullGestureData;
string file_path = "../../../data/";
if( !trainingData.loadDatasetFromFile(file_path + "train/grt/hri-training-dataset.txt") ){
std::cout <<"Failed to load training data!\n";
}
if( !nullGestureData.loadDatasetFromFile(file_path + "test/grt/0.txt") ){
std::cout <<"Failed to load null gesture data!\n";
}
testData = trainingData.partition(90);
testData.sortClassLabels();
// testData.saveDatasetToFile("anbc-validation-subset.txt");
testData.saveDatasetToFile("mindist-validation-subset.txt");
for(double nullRejectionCoeff = 0; nullRejectionCoeff <= 10; nullRejectionCoeff=nullRejectionCoeff+0.2){
// anbc.setNullRejectionCoeff(nullRejectionCoeff);
// GestureRecognitionPipeline pipeline;
// pipeline.setClassifier(anbc);
minDist.setNullRejectionCoeff(nullRejectionCoeff);
GestureRecognitionPipeline pipeline;
pipeline.setClassifier(minDist);
pipeline.train(trainingData);
pipeline.test(testData);
TestResult testRes = pipeline.getTestResults();
opRecall << nullRejectionCoeff << ",";
//null rejection prediction
double accuracy = 0;
for(UINT i=0; i<nullGestureData.getNumSamples(); i++){
vector< double > inputVector = nullGestureData[i].getSample();
if( !pipeline.predict( inputVector )){
std::cout << "Failed to perform prediction for test sampel: " << i <<"\n";
}
UINT predictedClassLabel = pipeline.getPredictedClassLabel();
if(predictedClassLabel == 0 ) accuracy++;
}
opRecall << accuracy/double(nullGestureData.getNumSamples()) << ",";
// other classes prediction
for(int cl = 0; cl < testRes.recall.size(); cl++ ){
opRecall << testRes.recall[cl];
if(cl < testRes.recall.size() - 1){
opRecall << ",";
}
}
opRecall<< endl;
// Calculate instance prediction, precision, recall, fmeasure and confusion matrix for nullRejection 2.0
if(AreDoubleSame(nullRejectionCoeff, 2.0))
{
//instance prediction
for(UINT i=0; i<testData.getNumSamples(); i++){
UINT actualClassLabel = testData[i].getClassLabel();
vector< double > inputVector = testData[i].getSample();
if( !pipeline.predict( inputVector )){
std::cout << "Failed to perform prediction for test sampel: " << i <<"\n";
}
UINT predictedClassLabel = pipeline.getPredictedClassLabel();
double maximumLikelihood = pipeline.getMaximumLikelihood();
opInstanceRes << actualClassLabel << "," << predictedClassLabel << "," << maximumLikelihood << ","
<< inputVector[0] << "," << inputVector[1] << "," << inputVector[2] << "," << inputVector[3] << "," << inputVector[4] << "," << inputVector[5] << "\n";
}
//precision, recall, fmeasure
for(int cl = 0; cl < testRes.precision.size(); cl++ ){
opMetrics << testRes.precision[cl];
if(cl < testRes.precision.size() - 1){
opMetrics << ",";
}
}
opMetrics<< endl;
for(int cl = 0; cl < testRes.recall.size(); cl++ ){
opMetrics << testRes.recall[cl];
if(cl < testRes.recall.size() - 1){
opMetrics << ",";
}
}
opMetrics<< endl;
for(int cl = 0; cl < testRes.fMeasure.size(); cl++ ){
opMetrics << testRes.fMeasure[cl];
if(cl < testRes.fMeasure.size() - 1){
opMetrics << ",";
}
}
opMetrics<< endl;
//confusion matrix
MatrixDouble matrix = testRes.confusionMatrix;
for(UINT i=0; i<matrix.getNumRows(); i++){
for(UINT j=0; j<matrix.getNumCols(); j++){
opConfusion << matrix[i][j];
if(j < matrix.getNumCols() - 1){
opConfusion << ",";
}
}
opConfusion << endl;
}
opConfusion << endl;
}
}
cout << "Done\n";
}
bool Softmax::train_(ClassificationData &trainingData){
//Clear any previous model
clear();
const unsigned int M = trainingData.getNumSamples();
const unsigned int N = trainingData.getNumDimensions();
const unsigned int K = trainingData.getNumClasses();
if( M == 0 ){
errorLog << __GRT_LOG__ << " Training data has zero samples!" << std::endl;
return false;
}
numInputDimensions = N;
numOutputDimensions = K;
numClasses = K;
models.resize(K);
classLabels.resize(K);
ranges = trainingData.getRanges();
ClassificationData validationData;
//Scale the training data if needed
if( useScaling ){
//Scale the training data between 0 and 1
trainingData.scale(0, 1);
}
if( useValidationSet ){
validationData = trainingData.split( 100-validationSetSize );
}
//Train a regression model for each class in the training data
for(UINT k=0; k<numClasses; k++){
//Set the class label
classLabels[k] = trainingData.getClassTracker()[k].classLabel;
//Train the model
if( !trainSoftmaxModel(classLabels[k],models[k],trainingData) ){
errorLog << __GRT_LOG__ << " Failed to train model for class: " << classLabels[k] << std::endl;
return false;
}
}
//Flag that the models have been trained
trained = true;
converged = true;
//Compute the final training stats
trainingSetAccuracy = 0;
validationSetAccuracy = 0;
//If scaling was on, then the data will already be scaled, so turn it off temporially so we can test the model accuracy
bool scalingState = useScaling;
useScaling = false;
if( !computeAccuracy( trainingData, trainingSetAccuracy ) ){
trained = false;
converged = false;
errorLog << __GRT_LOG__ << " Failed to compute training set accuracy! Failed to fully train model!" << std::endl;
return false;
}
if( useValidationSet ){
if( !computeAccuracy( validationData, validationSetAccuracy ) ){
trained = false;
converged = false;
errorLog << __GRT_LOG__ << " Failed to compute validation set accuracy! Failed to fully train model!" << std::endl;
return false;
}
}
trainingLog << "Training set accuracy: " << trainingSetAccuracy << std::endl;
if( useValidationSet ){
trainingLog << "Validation set accuracy: " << validationSetAccuracy << std::endl;
}
//Reset the scaling state for future prediction
useScaling = scalingState;
return trained;
}
void metrics_separate_data(){
// Training and test data
ClassificationData trainingData;
ClassificationData testData;
string file_path = "../../../data/";
if( !trainingData.loadDatasetFromFile(file_path + "train/grt/12345.txt") ){
std::cout <<"Failed to load training data!\n";
}
ANBC anbc;
anbc.enableScaling(true);
anbc.enableNullRejection(true);
SVM svm(SVM::RBF_KERNEL);
svm.enableScaling(true);
svm.enableNullRejection(true);
MinDist minDist;
minDist.setNumClusters(4);
minDist.enableScaling(true);
minDist.enableNullRejection(true);
ofstream outputFileStream("accuracy-mindist.csv");
outputFileStream << "classLabel,nullRejectionCoeff,accuracy, \n";
for(int class_name = 1; class_name<=5; class_name++){
if( !testData.loadDatasetFromFile(file_path + "test/grt/" + to_string(class_name) + ".txt") ){
std::cout <<"Failed to load training data!\n";
}
for(double nullRejectionCoeff = 0; nullRejectionCoeff <= 10; nullRejectionCoeff=nullRejectionCoeff+0.2){
// anbc.setNullRejectionCoeff(nullRejectionCoeff);
// svm.setNullRejectionCoeff(nullRejectionCoeff);
minDist.setNullRejectionCoeff(nullRejectionCoeff);
GestureRecognitionPipeline pipeline;
// pipeline.setClassifier(anbc);
// pipeline.setClassifier(svm);
pipeline.setClassifier(minDist);
// Train the pipeline
if( !pipeline.train( trainingData ) ){
std::cout << "Failed to train classifier!\n";
}
// Evaluation
double accuracy = 0;
for(UINT i=0; i<testData.getNumSamples(); i++){
UINT actualClassLabel = testData[i].getClassLabel();
vector< double > inputVector = testData[i].getSample();
if( !pipeline.predict( inputVector )){
std::cout << "Failed to perform prediction for test sampel: " << i <<"\n";
}
UINT predictedClassLabel = pipeline.getPredictedClassLabel();
if( actualClassLabel == predictedClassLabel) accuracy++;
}
outputFileStream << class_name << ',' << nullRejectionCoeff << ',' << accuracy/double(testData.getNumSamples())*100.0 << '\n';
cout<< "Done" << endl;
}
}
//---------------------- Null Gesture Test -----------------//
int class_name = 0;
if( !testData.loadDatasetFromFile(file_path + "test/grt/" + to_string(class_name) + ".txt") ){
std::cout <<"Failed to load training data!\n";
}
for(double nullRejectionCoeff = 0; nullRejectionCoeff <= 10; nullRejectionCoeff=nullRejectionCoeff+0.2){
// anbc.setNullRejectionCoeff(nullRejectionCoeff);
// svm.setNullRejectionCoeff(nullRejectionCoeff);
minDist.setNullRejectionCoeff(nullRejectionCoeff);
GestureRecognitionPipeline pipeline;
// pipeline.setClassifier(anbc);
// pipeline.setClassifier(svm);
pipeline.setClassifier(minDist);
// Train the pipeline
if( !pipeline.train( trainingData ) ){
std::cout << "Failed to train classifier!\n";
}
// Evaluation
double accuracy = 0;
for(UINT i=0; i<testData.getNumSamples(); i++){
vector< double > inputVector = testData[i].getSample();
if( !pipeline.predict( inputVector )){
std::cout << "Failed to perform prediction for test sampel: " << i <<"\n";
}
UINT predictedClassLabel = pipeline.getPredictedClassLabel();
if(predictedClassLabel == 0 ) accuracy++;
}
outputFileStream << class_name << ',' << nullRejectionCoeff << ',' << accuracy/double(testData.getNumSamples())*100.0 << '\n';
cout<< "Done" << endl;
}
}
bool LDA::train(ClassificationData trainingData){
errorLog << "SORRY - this module is still under development and can't be used yet!" << std::endl;
return false;
//Reset any previous model
numInputDimensions = 0;
numClasses = 0;
models.clear();
classLabels.clear();
trained = false;
if( trainingData.getNumSamples() == 0 ){
errorLog << "train(LabelledClassificationData trainingData) - There is no training data to train the model!" << std::endl;
return false;
}
numInputDimensions = trainingData.getNumDimensions();
numClasses = trainingData.getNumClasses();
//Calculate the between scatter matrix
MatrixFloat SB = computeBetweenClassScatterMatrix( trainingData );
//Calculate the within scatter matrix
MatrixFloat SW = computeWithinClassScatterMatrix( trainingData );
/*
//Counters and stat containers
vector< UINT > groupLabels(numClasses);
VectorDouble groupCounters(numClasses);
VectorDouble priorProb(numClasses);
MatrixFloat groupMeans(numClasses,numFeatures);
MatrixFloat pCov(numFeatures,numFeatures);
MatrixFloat pCovInv(numFeatures,numFeatures);
MatrixFloat modelCoeff(numClasses,numFeatures+1);
pCov.setAllValues(0);
modelCoeff.setAllValues(0);
//Set the class labels and counters
for(UINT k=0; k<numClasses; k++){
groupLabels[k] = trainingData.getClassTracker()[k].classLabel;
groupCounters[k] = trainingData.getClassTracker()[k].counter;
}
//Loop over the classes to compute the group stats
for(UINT k=0; k<numClasses; k++){
LabelledClassificationData classData = trainingData.getClassData( groupLabels[k] );
MatrixFloat cov(numFeatures,numFeatures);
//Compute class mu
for(UINT j=0; j<numFeatures; j++){
groupMeans[k][j] = 0;
for(UINT i=0; i<classData.getNumSamples(); i++){
groupMeans[k][j] += classData[i][j];
}
groupMeans[k][j] /= Float(classData.getNumSamples());
}
//Compute the class covariance
for(UINT m=0; m<numFeatures; m++){
for(UINT n=0; n<numFeatures; n++){
cov[m][n] = 0;
for(UINT i=0; i<classData.getNumSamples(); i++){
cov[m][n] += (classData[i][m]-groupMeans[k][m]) * (classData[i][n]-groupMeans[k][n]);
}
cov[m][n] /= Float(classData.getNumSamples()-1);
}
}
debugLog << "Group Cov:\n";
for(UINT m=0; m<numFeatures; m++){
for(UINT n=0; n<numFeatures; n++){
debugLog << cov[m][n] << "\t";
}debugLog << "\n";
}debugLog << std::endl;
//Set the prior probability for this class (which is just 1/numClasses)
priorProb[k] = 1.0/Float(numClasses);
//Update the main covariance matrix
Float weight = ((classData.getNumSamples() - 1) / Float(trainingData.getNumSamples() - numClasses) );
debugLog << "Weight: " << weight << std::endl;
for(UINT m=0; m<numFeatures; m++){
for(UINT n=0; n<numFeatures; n++){
pCov[m][n] += weight * cov[m][n];
}
}
}
for(UINT k=0; k<numClasses; k++){
debugLog << "GroupMu: " << groupLabels[k] << "\t";
for(UINT j=0; j<numFeatures; j++){
debugLog << groupMeans[k][j] << "\t";
}debugLog << std::endl;
}
debugLog << "pCov:\n";
for(UINT m=0; m<numFeatures; m++){
for(UINT n=0; n<numFeatures; n++){
debugLog << pCov[m][n] << "\t";
}debugLog << "\n";
}debugLog << std::endl;
//Invert the pCov matrix
LUDecomposition matrixInverter(pCov);
if( !matrixInverter.inverse(pCovInv) ){
errorLog << "Failed to invert pCov Matrix!" << std::endl;
return false;
}
//Loop over classes to calculate linear discriminant coefficients
Float sum = 0;
vector< Float > temp(numFeatures);
for(UINT k=0; k<numClasses; k++){
//Compute the temporary vector
for(UINT j=0; j<numFeatures; j++){
temp[j] = 0;
for(UINT m=0; m<numFeatures; m++){
temp[j] += groupMeans[k][m] * pCovInv[m][j];
}
}
//Compute the model coefficients
sum = 0;
for(UINT j=0; j<numFeatures; j++){
sum += temp[j]*groupMeans[k][j];
}
modelCoeff[k][0] = -0.5 * sum + log( priorProb[k] );
for(UINT j=0; j<numFeatures; j++){
modelCoeff[k][j+1] = temp[j];
}
}
//Setup the models for realtime prediction
models.resize(numClasses);
classLabels.resize(numClasses);
for(UINT k=0; k<numClasses; k++){
classLabels[k] = groupLabels[k];
models[k].classLabel = groupLabels[k];
models[k].priorProb = priorProb[k];
models[k].weights = modelCoeff.getRowVector(k);
}
//Flag that the models were successfully trained
trained = true;
*/
return true;
}
bool RandomForests::train_(ClassificationData &trainingData){
//Clear any previous model
clear();
const unsigned int M = trainingData.getNumSamples();
const unsigned int N = trainingData.getNumDimensions();
const unsigned int K = trainingData.getNumClasses();
if( M == 0 ){
errorLog << "train_(ClassificationData &trainingData) - Training data has zero samples!" << endl;
return false;
}
if( bootstrappedDatasetWeight <= 0.0 || bootstrappedDatasetWeight > 1.0 ){
errorLog << "train_(ClassificationData &trainingData) - Bootstrapped Dataset Weight must be [> 0.0 and <= 1.0]" << endl;
return false;
}
numInputDimensions = N;
numClasses = K;
classLabels = trainingData.getClassLabels();
ranges = trainingData.getRanges();
//Scale the training data if needed
if( useScaling ){
//Scale the training data between 0 and 1
trainingData.scale(0, 1);
}
//Flag that the main algorithm has been trained encase we need to trigger any callbacks
trained = true;
//Train the random forest
forest.reserve( forestSize );
for(UINT i=0; i<forestSize; i++){
//Get a balanced bootstrapped dataset
UINT datasetSize = (UINT)(trainingData.getNumSamples() * bootstrappedDatasetWeight);
ClassificationData data = trainingData.getBootstrappedDataset( datasetSize, true );
DecisionTree tree;
tree.setDecisionTreeNode( *decisionTreeNode );
tree.enableScaling( false ); //We have already scaled the training data so we do not need to scale it again
tree.setTrainingMode( trainingMode );
tree.setNumSplittingSteps( numRandomSplits );
tree.setMinNumSamplesPerNode( minNumSamplesPerNode );
tree.setMaxDepth( maxDepth );
tree.enableNullRejection( useNullRejection );
tree.setRemoveFeaturesAtEachSpilt( removeFeaturesAtEachSpilt );
trainingLog << "Training forest " << i+1 << "/" << forestSize << "..." << endl;
//Train this tree
if( !tree.train( data ) ){
errorLog << "train_(ClassificationData &labelledTrainingData) - Failed to train tree at forest index: " << i << endl;
clear();
return false;
}
//Deep copy the tree into the forest
forest.push_back( tree.deepCopyTree() );
}
return true;
}
