pmlpp/mlpp/mann/mann.cpp

//
//  MANN.cpp
//
//  Created by Marc Melikyan on 11/4/20.
//

#include "mann.h"
#include "../activation/activation.h"
#include "../cost/cost.h"
#include "../lin_alg/lin_alg.h"
#include "../regularization/reg.h"
#include "../utilities/utilities.h"

#include <iostream>

namespace MLPP {
MANN::MANN(std::vector<std::vector<double>> inputSet, std::vector<std::vector<double>> outputSet) :
		inputSet(inputSet), outputSet(outputSet), n(inputSet.size()), k(inputSet[0].size()), n_output(outputSet[0].size()) {
}

MANN::~MANN() {
	delete outputLayer;
}

std::vector<std::vector<double>> MANN::modelSetTest(std::vector<std::vector<double>> X) {
	if (!network.empty()) {
		network[0].input = X;
		network[0].forwardPass();

		for (int i = 1; i < network.size(); i++) {
			network[i].input = network[i - 1].a;
			network[i].forwardPass();
		}
		outputLayer->input = network[network.size() - 1].a;
	} else {
		outputLayer->input = X;
	}
	outputLayer->forwardPass();
	return outputLayer->a;
}

std::vector<double> MANN::modelTest(std::vector<double> x) {
	if (!network.empty()) {
		network[0].Test(x);
		for (int i = 1; i < network.size(); i++) {
			network[i].Test(network[i - 1].a_test);
		}
		outputLayer->Test(network[network.size() - 1].a_test);
	} else {
		outputLayer->Test(x);
	}
	return outputLayer->a_test;
}

void MANN::gradientDescent(double learning_rate, int max_epoch, bool UI) {
	class Cost cost;
	Activation avn;
	LinAlg alg;
	Reg regularization;

	double cost_prev = 0;
	int epoch = 1;
	forwardPass();

	while (true) {
		cost_prev = Cost(y_hat, outputSet);

		if (outputLayer->activation == "Softmax") {
			outputLayer->delta = alg.subtraction(y_hat, outputSet);
		} else {
			auto costDeriv = outputLayer->costDeriv_map[outputLayer->cost];
			auto outputAvn = outputLayer->activation_map[outputLayer->activation];
			outputLayer->delta = alg.hadamard_product((cost.*costDeriv)(y_hat, outputSet), (avn.*outputAvn)(outputLayer->z, 1));
		}

		std::vector<std::vector<double>> outputWGrad = alg.matmult(alg.transpose(outputLayer->input), outputLayer->delta);

		outputLayer->weights = alg.subtraction(outputLayer->weights, alg.scalarMultiply(learning_rate / n, outputWGrad));
		outputLayer->weights = regularization.regWeights(outputLayer->weights, outputLayer->lambda, outputLayer->alpha, outputLayer->reg);
		outputLayer->bias = alg.subtractMatrixRows(outputLayer->bias, alg.scalarMultiply(learning_rate / n, outputLayer->delta));

		if (!network.empty()) {
			auto hiddenLayerAvn = network[network.size() - 1].activation_map[network[network.size() - 1].activation];
			network[network.size() - 1].delta = alg.hadamard_product(alg.matmult(outputLayer->delta, alg.transpose(outputLayer->weights)), (avn.*hiddenLayerAvn)(network[network.size() - 1].z, 1));
			std::vector<std::vector<double>> hiddenLayerWGrad = alg.matmult(alg.transpose(network[network.size() - 1].input), network[network.size() - 1].delta);

			network[network.size() - 1].weights = alg.subtraction(network[network.size() - 1].weights, alg.scalarMultiply(learning_rate / n, hiddenLayerWGrad));
			network[network.size() - 1].weights = regularization.regWeights(network[network.size() - 1].weights, network[network.size() - 1].lambda, network[network.size() - 1].alpha, network[network.size() - 1].reg);
			network[network.size() - 1].bias = alg.subtractMatrixRows(network[network.size() - 1].bias, alg.scalarMultiply(learning_rate / n, network[network.size() - 1].delta));

			for (int i = network.size() - 2; i >= 0; i--) {
				auto hiddenLayerAvn = network[i].activation_map[network[i].activation];
				network[i].delta = alg.hadamard_product(alg.matmult(network[i + 1].delta, network[i + 1].weights), (avn.*hiddenLayerAvn)(network[i].z, 1));
				std::vector<std::vector<double>> hiddenLayerWGrad = alg.matmult(alg.transpose(network[i].input), network[i].delta);
				network[i].weights = alg.subtraction(network[i].weights, alg.scalarMultiply(learning_rate / n, hiddenLayerWGrad));
				network[i].weights = regularization.regWeights(network[i].weights, network[i].lambda, network[i].alpha, network[i].reg);
				network[i].bias = alg.subtractMatrixRows(network[i].bias, alg.scalarMultiply(learning_rate / n, network[i].delta));
			}
		}

		forwardPass();

		if (UI) {
			Utilities::CostInfo(epoch, cost_prev, Cost(y_hat, outputSet));
			std::cout << "Layer " << network.size() + 1 << ": " << std::endl;
			Utilities::UI(outputLayer->weights, outputLayer->bias);
			if (!network.empty()) {
				std::cout << "Layer " << network.size() << ": " << std::endl;
				for (int i = network.size() - 1; i >= 0; i--) {
					std::cout << "Layer " << i + 1 << ": " << std::endl;
					Utilities::UI(network[i].weights, network[i].bias);
				}
			}
		}

		epoch++;
		if (epoch > max_epoch) {
			break;
		}
	}
}

double MANN::score() {
	Utilities util;
	forwardPass();
	return util.performance(y_hat, outputSet);
}

void MANN::save(std::string fileName) {
	Utilities util;
	if (!network.empty()) {
		util.saveParameters(fileName, network[0].weights, network[0].bias, 0, 1);
		for (int i = 1; i < network.size(); i++) {
			util.saveParameters(fileName, network[i].weights, network[i].bias, 1, i + 1);
		}
		util.saveParameters(fileName, outputLayer->weights, outputLayer->bias, 1, network.size() + 1);
	} else {
		util.saveParameters(fileName, outputLayer->weights, outputLayer->bias, 0, network.size() + 1);
	}
}

void MANN::addLayer(int n_hidden, std::string activation, std::string weightInit, std::string reg, double lambda, double alpha) {
	if (network.empty()) {
		network.push_back(HiddenLayer(n_hidden, activation, inputSet, weightInit, reg, lambda, alpha));
		network[0].forwardPass();
	} else {
		network.push_back(HiddenLayer(n_hidden, activation, network[network.size() - 1].a, weightInit, reg, lambda, alpha));
		network[network.size() - 1].forwardPass();
	}
}

void MANN::addOutputLayer(std::string activation, std::string loss, std::string weightInit, std::string reg, double lambda, double alpha) {
	if (!network.empty()) {
		outputLayer = new MultiOutputLayer(n_output, network[0].n_hidden, activation, loss, network[network.size() - 1].a, weightInit, reg, lambda, alpha);
	} else {
		outputLayer = new MultiOutputLayer(n_output, k, activation, loss, inputSet, weightInit, reg, lambda, alpha);
	}
}

double MANN::Cost(std::vector<std::vector<double>> y_hat, std::vector<std::vector<double>> y) {
	Reg regularization;
	class Cost cost;
	double totalRegTerm = 0;

	auto cost_function = outputLayer->cost_map[outputLayer->cost];
	if (!network.empty()) {
		for (int i = 0; i < network.size() - 1; i++) {
			totalRegTerm += regularization.regTerm(network[i].weights, network[i].lambda, network[i].alpha, network[i].reg);
		}
	}
	return (cost.*cost_function)(y_hat, y) + totalRegTerm + regularization.regTerm(outputLayer->weights, outputLayer->lambda, outputLayer->alpha, outputLayer->reg);
}

void MANN::forwardPass() {
	if (!network.empty()) {
		network[0].input = inputSet;
		network[0].forwardPass();

		for (int i = 1; i < network.size(); i++) {
			network[i].input = network[i - 1].a;
			network[i].forwardPass();
		}
		outputLayer->input = network[network.size() - 1].a;
	} else {
		outputLayer->input = inputSet;
	}
	outputLayer->forwardPass();
	y_hat = outputLayer->a;
}
} //namespace MLPP