Files
Platform/src/experimental_clfs/AdaBoost.h

58 lines
2.0 KiB
C++

// ***************************************************************
// SPDX-FileCopyrightText: Copyright 2024 Ricardo Montañana Gómez
// SPDX-FileType: SOURCE
// SPDX-License-Identifier: MIT
// ***************************************************************
#ifndef ADABOOST_H
#define ADABOOST_H
#include <vector>
#include <memory>
#include <torch/torch.h>
#include <bayesnet/ensembles/Ensemble.h>
namespace platform {
class AdaBoost : public bayesnet::Ensemble {
public:
explicit AdaBoost(int n_estimators = 100);
virtual ~AdaBoost() = default;
// Override base class methods
std::vector<std::string> graph(const std::string& title = "") const override;
// AdaBoost specific methods
void setNEstimators(int n_estimators) { this->n_estimators = n_estimators; }
int getNEstimators() const { return n_estimators; }
// Get the weight of each base estimator
std::vector<double> getEstimatorWeights() const { return alphas; }
// Override setHyperparameters from BaseClassifier
void setHyperparameters(const nlohmann::json& hyperparameters) override;
protected:
void buildModel(const torch::Tensor& weights) override;
void trainModel(const torch::Tensor& weights, const Smoothing_t smoothing) override;
private:
int n_estimators;
std::vector<double> alphas; // Weight of each base estimator
std::vector<double> training_errors; // Training error at each iteration
torch::Tensor sample_weights; // Current sample weights
// Train a single base estimator
std::unique_ptr<Classifier> trainBaseEstimator(const torch::Tensor& weights);
// Calculate weighted error
double calculateWeightedError(Classifier* estimator, const torch::Tensor& weights);
// Update sample weights based on predictions
void updateSampleWeights(Classifier* estimator, double alpha);
// Normalize weights to sum to 1
void normalizeWeights();
};
}
#endif // ADABOOST_H