gpmp::ml::BayesGauss Class Reference

#include <bayes_clf.hpp>

Public Member Functions
	BayesGauss ()=default
	Constructor for BayesGauss class. More...
	~BayesGauss ()=default
	Destructor for BayesGauss class. More...
void	train (const std::vector< std::vector< double >> &data, const std::vector< std::string > &labels)
	Train the classifier with a set of labeled data. More...
std::string	predict (const std::vector< double > &newData) const
	Predict the class of a new data point. More...
void	display () const
	Display the learned probabilities. More...

Public Attributes
std::unordered_map< std::string, double >	class_probs
	Map storing the probabilities of each class. More...
std::unordered_map< std::string, std::vector< double > >	mean
	Map storing the mean values for each feature in each class. More...
std::unordered_map< std::string, std::vector< double > >	variance
	Map storing the variance values for each feature in each class. More...

Private Member Functions
void	mean_and_var (const std::vector< std::vector< double >> &data, const std::vector< std::string > &labels)
	Calculate the mean and variance for each class. More...

Detailed Description

Classification technique that assumes that each parameter (also called features or predictors) has an independent capacity of predicting the output variable.

Definition at line 172 of file bayes_clf.hpp.

Constructor & Destructor Documentation

BayesGauss()

gpmp::ml::BayesGauss::BayesGauss ( )

default

Constructor for BayesGauss class.

~BayesGauss()

gpmp::ml::BayesGauss::~BayesGauss ( )

default

Destructor for BayesGauss class.

Member Function Documentation

display()

void gpmp::ml::BayesGauss::display

(

)

const

Display the learned probabilities.

Note

This method is for debugging purposes

Definition at line 323 of file bayes_clf.cpp.

                                     {
    std::cout << "Class Probabilities:\n";
    for (const auto &entry : class_probs) {
        std::cout << entry.first << ": " << entry.second << "\n";
    }
 
    std::cout << "\nMean and Variance:\n";
    for (const auto &class_entry : mean) {
        std::cout << class_entry.first << ":\n";
        std::cout << "  Mean: ";
        for (size_t j = 0; j < class_entry.second.size(); ++j) {
            std::cout << class_entry.second[j] << " ";
        }
        std::cout << "\n  Variance: ";
        for (size_t j = 0; j < variance.at(class_entry.first).size(); ++j) {
            std::cout << variance.at(class_entry.first).at(j) << " ";
        }
        std::cout << "\n";
    }
}

Referenced by main().

mean_and_var()

void gpmp::ml::BayesGauss::mean_and_var ( const std::vector< std::vector< double >> &  data, const std::vector< std::string > &  labels  )

private

Calculate the mean and variance for each class.

Parameters

data	A vector of vectors representing the training instances
labels	A vector of strings representing the corresponding class labels

Definition at line 245 of file bayes_clf.cpp.

                                        {
    size_t num_features = data[0].size();
 
    for (size_t i = 0; i < data.size(); ++i) {
        const std::string &label = labels[i];
        const std::vector<double> &features = data[i];
 
        class_probs[label] += 1.0;
 
        // initialize mean[label] and variance[label] if not present
        if (mean.find(label) == mean.end()) {
            mean[label] = std::vector<double>(num_features, 0.0);
            variance[label] = std::vector<double>(num_features, 0.0);
        }
 
        // update mean
        for (size_t j = 0; j < num_features; ++j) {
            mean[label][j] += features[j];
        }
    }
 
    // calculate mean
    for (auto &entry : mean) {
        const std::string &label = entry.first;
        double class_count = class_probs[label];
 
        for (size_t j = 0; j < num_features; ++j) {
            entry.second[j] /= class_count;
        }
    }
 
    // calculate variance
    for (size_t i = 0; i < data.size(); ++i) {
        const std::string &label = labels[i];
        const std::vector<double> &features = data[i];
 
        for (size_t j = 0; j < num_features; ++j) {
            variance[label][j] += std::pow(features[j] - mean[label][j], 2);
        }
    }
 
    for (auto &entry : variance) {
        const std::string &label = entry.first;
        double class_count = class_probs[label];
 
        for (size_t j = 0; j < num_features; ++j) {
            entry.second[j] /= class_count;
        }
    }
}

predict()

std::string gpmp::ml::BayesGauss::predict

(

const std::vector< double > &

newData

)

const

Predict the class of a new data point.

Parameters

newData

A vector of doubles representing the features of the new data point

Returns

The predicted class label as a string

Definition at line 299 of file bayes_clf.cpp.

                                                                {
    double max_prob = -std::numeric_limits<double>::infinity();
    std::string predicted_class;
 
    for (const auto &entry : class_probs) {
        const std::string &label = entry.first;
        double probability = log(entry.second);
 
        for (size_t j = 0; j < newData.size(); ++j) {
            probability -=
                0.5 * (std::log(2 * M_PI * variance.at(label).at(j)) +
                       std::pow(newData[j] - mean.at(label).at(j), 2) /
                           (2 * variance.at(label).at(j)));
        }
 
        if (probability > max_prob) {
            max_prob = probability;
            predicted_class = label;
        }
    }
 
    return predicted_class;
}

Referenced by main().

train()

void gpmp::ml::BayesGauss::train	(	const std::vector< std::vector< double >> &	data,
		const std::vector< std::string > &	labels
	)

Train the classifier with a set of labeled data.

Parameters

data	A vector of vectors representing the training instances
labels	A vector of strings representing the corresponding class labels

Definition at line 228 of file bayes_clf.cpp.

                                                                     {
    // calculate class occurrences
    for (const auto &label : labels) {
        class_probs[label] += 1.0;
    }
 
    // calculate mean and variance for each feature in each class
    mean_and_var(data, labels);
 
    // calculate class probabilities
    double total_instances = static_cast<double>(data.size());
    for (auto &entry : class_probs) {
        entry.second /= total_instances;
    }
}

Referenced by main().

Member Data Documentation

class_probs

gpmp::ml::BayesGauss::class_probs

Map storing the probabilities of each class.

The key is the class label (string), and the value is the probability (double)

Definition at line 180 of file bayes_clf.hpp.

mean

gpmp::ml::BayesGauss::mean

Map storing the mean values for each feature in each class.

The key is the class label (string), and the value is a vector of mean values

Definition at line 187 of file bayes_clf.hpp.

variance

gpmp::ml::BayesGauss::variance

Map storing the variance values for each feature in each class.

The key is the class label (string), and the value is a vector of variance values

Definition at line 194 of file bayes_clf.hpp.

---

The documentation for this class was generated from the following files:

• include/openGPMP/ml/bayes_clf.hpp
• modules/ml/bayes_clf.cpp