當前位置: 首頁>>代碼示例>>C++>>正文

C++ VectorXd::coeff方法代碼示例

本文整理匯總了C++中eigen::VectorXd::coeff方法的典型用法代碼示例。如果您正苦於以下問題：C++ VectorXd::coeff方法的具體用法？C++ VectorXd::coeff怎麽用？C++ VectorXd::coeff使用的例子？那麽, 這裏精選的方法代碼示例或許可以為您提供幫助。您也可以進一步了解該方法所在類eigen::VectorXd的用法示例。


在下文中一共展示了VectorXd::coeff方法的7個代碼示例，這些例子默認根據受歡迎程度排序。您可以為喜歡或者感覺有用的代碼點讚，您的評價將有助於係統推薦出更棒的C++代碼示例。

示例1: testEfficientPortfolio

/**
 * @covers EfficientRiskyAssetPortfolio::performanceIncreasingPortfolio
 */
void EfficientRiskyAssetPortfolioTest::testEfficientPortfolio()
{
	_createReturnStatistics();

	EfficientRiskyAssetPortfolio model(*_mean, *_variance);

	{
		// more than mvp return
		Eigen::VectorXd ep = model.efficientPortfolio(0.1);

		CPPUNIT_ASSERT_DOUBLES_EQUAL(28.0 / 97.0, ep.coeff(0), EPSILON);
		CPPUNIT_ASSERT_DOUBLES_EQUAL(41.0 / 97.0, ep.coeff(1), EPSILON);
		CPPUNIT_ASSERT_DOUBLES_EQUAL(28.0 / 97.0, ep.coeff(2), EPSILON);

		CPPUNIT_ASSERT_DOUBLES_EQUAL(0.1, model.portfolioReturn(ep), EPSILON);
		CPPUNIT_ASSERT_DOUBLES_EQUAL(192.0 / 2425.0, model.portfolioVariance(ep), EPSILON);
	}
	{
		// less than mvp return, i.e. mvp is the efficient portfolio
		Eigen::VectorXd ep = model.efficientPortfolio(0.01);

		CPPUNIT_ASSERT_DOUBLES_EQUAL(-24.0 / 70.0, ep.coeff(0), EPSILON);
		CPPUNIT_ASSERT_DOUBLES_EQUAL(51.0 / 70.0, ep.coeff(1), EPSILON);
		CPPUNIT_ASSERT_DOUBLES_EQUAL(43.0 / 70.0, ep.coeff(2), EPSILON);

		CPPUNIT_ASSERT_DOUBLES_EQUAL(73.0 / 1400.0, model.portfolioReturn(ep), EPSILON);
		CPPUNIT_ASSERT_DOUBLES_EQUAL(23.0 / 875.0, model.portfolioVariance(ep), EPSILON);
	}

	_destroyReturnStatistics();
}
開發者ID:regiomontanus，項目名稱:FastPortfolioOptimizer，代碼行數:34，代碼來源:EfficientRiskyAssetPortfolioTest.cpp

示例2: testPerformanceIncreasingPortfolio

/**
 * @covers EfficientRiskyAssetPortfolio::performanceIncreasingPortfolio
 */
void EfficientRiskyAssetPortfolioTest::testPerformanceIncreasingPortfolio()
{
	_createReturnStatistics();

	EfficientRiskyAssetPortfolio model(*_mean, *_variance);
	Eigen::VectorXd pip = model.performanceIncreasingPortfolio();

	CPPUNIT_ASSERT_DOUBLES_EQUAL(1280.0 / 97.0, pip.coeff(0), 1e-14);
	CPPUNIT_ASSERT_DOUBLES_EQUAL(-620.0 / 97.0, pip.coeff(1), 1e-14);
	CPPUNIT_ASSERT_DOUBLES_EQUAL(-660.0 / 97.0, pip.coeff(2), 1e-14);

	_destroyReturnStatistics();
}
開發者ID:regiomontanus，項目名稱:FastPortfolioOptimizer，代碼行數:16，代碼來源:EfficientRiskyAssetPortfolioTest.cpp

示例3: testPerformanceIncreasingPortfolio

/**
 * @covers EfficientPortfolioWithRisklessAssetTest::performanceIncreasingPortfolio
 */
void EfficientPortfolioWithRisklessAssetTest::testPerformanceIncreasingPortfolio()
{
	_createReturnStatistics();

	EfficientPortfolioWithRisklessAsset model(0.04, *_mean, *_variance);
	Eigen::VectorXd pip = model.performanceIncreasingPortfolio();

	CPPUNIT_ASSERT_DOUBLES_EQUAL(0.0, pip.coeff(0), 1e-14);
	CPPUNIT_ASSERT_DOUBLES_EQUAL(0.0, pip.coeff(1), 1e-14);
	CPPUNIT_ASSERT_DOUBLES_EQUAL(0.0, pip.coeff(2), 1e-14);
	CPPUNIT_ASSERT_DOUBLES_EQUAL(0.0, pip.coeff(3), 1e-14);

	_destroyReturnStatistics();
}
開發者ID:regiomontanus，項目名稱:FastPortfolioOptimizer，代碼行數:17，代碼來源:EfficientPortfolioWithRisklessAssetTest.cpp

示例4: testMinimumVariancePortfolio

/**
 * @covers EfficientRiskyAssetPortfolio::minimumVariancePortfolio
 * @covers EfficientRiskyAssetPortfolio::portfolioReturn
 * @covers EfficientRiskyAssetPortfolio::portfolioVariance
 */
void EfficientRiskyAssetPortfolioTest::testMinimumVariancePortfolio()
{
	_createReturnStatistics();

	EfficientRiskyAssetPortfolio model(*_mean, *_variance);
	Eigen::VectorXd mvp = model.minimumVariancePortfolio();

	CPPUNIT_ASSERT_DOUBLES_EQUAL(-24.0 / 70.0, mvp.coeff(0), EPSILON);
	CPPUNIT_ASSERT_DOUBLES_EQUAL(51.0 / 70.0, mvp.coeff(1), EPSILON);
	CPPUNIT_ASSERT_DOUBLES_EQUAL(43.0 / 70.0, mvp.coeff(2), EPSILON);

	CPPUNIT_ASSERT_DOUBLES_EQUAL(73.0 / 1400.0, model.portfolioReturn(mvp), EPSILON);
	CPPUNIT_ASSERT_DOUBLES_EQUAL(23.0 / 875.0, model.portfolioVariance(mvp), EPSILON);

	_destroyReturnStatistics();
}
開發者ID:regiomontanus，項目名稱:FastPortfolioOptimizer，代碼行數:21，代碼來源:EfficientRiskyAssetPortfolioTest.cpp

示例5: testMinimumVariancePortfolio

/**
 * @covers EfficientPortfolioWithRisklessAssetTest::minimumVariancePortfolio
 * @covers EfficientPortfolioWithRisklessAssetTest::portfolioReturn
 * @covers EfficientPortfolioWithRisklessAssetTest::portfolioVariance
 */
void EfficientPortfolioWithRisklessAssetTest::testMinimumVariancePortfolio()
{
	_createReturnStatistics();

	EfficientPortfolioWithRisklessAsset model(0.04, *_mean, *_variance);
	Eigen::VectorXd mvp = model.minimumVariancePortfolio();

	CPPUNIT_ASSERT_DOUBLES_EQUAL(0.0, mvp.coeff(0), EPSILON);
	CPPUNIT_ASSERT_DOUBLES_EQUAL(0.0, mvp.coeff(1), EPSILON);
	CPPUNIT_ASSERT_DOUBLES_EQUAL(0.0, mvp.coeff(2), EPSILON);
	CPPUNIT_ASSERT_DOUBLES_EQUAL(1.0, mvp.coeff(3), EPSILON);

	CPPUNIT_ASSERT_DOUBLES_EQUAL(0.04, model.portfolioReturn(mvp), EPSILON);
	CPPUNIT_ASSERT_DOUBLES_EQUAL(0.0, model.portfolioVariance(mvp), EPSILON);

	_destroyReturnStatistics();
}
開發者ID:regiomontanus，項目名稱:FastPortfolioOptimizer，代碼行數:22，代碼來源:EfficientPortfolioWithRisklessAssetTest.cpp

示例6: test

double test(const lbfgsfloatval_t* parameter,
			int featsize,
			boost::shared_ptr<Eigen::SparseMatrix<double, Eigen::RowMajor> >& samples,
			boost::shared_ptr<Eigen::VectorXi>& labels){
				Eigen::Map<Eigen::VectorXd> param((double*)parameter, featsize, 1);
				Eigen::VectorXd xw = (*(samples))*param;

				double allcnt=0, rightcnt  =0 ;
				for(int i=0; i < labels->rows(); ++i){
					if( xw.coeff(i)*labels->coeff(i) > 0){
						rightcnt += 1;
					}
					allcnt += 1;
				}
				return rightcnt /allcnt;
}
開發者ID:lacozhang，項目名稱:machinelearning，代碼行數:16，代碼來源:train.cpp

示例7: evaluate

static lbfgsfloatval_t evaluate(
	void *instance,
	const lbfgsfloatval_t *x,
	lbfgsfloatval_t *g,
	const int n,
	const lbfgsfloatval_t step
	){
		double fx = 0;
		timeutil t;

		optimization_instance_t* optInst = (optimization_instance_t*)instance;
		Eigen::Map<Eigen::VectorXd> param((double*)x, 
			n, 1);
		Eigen::Map<Eigen::VectorXd> grad((double*)g,
			n, 1);

		// price for sparse A*x
		t.tic();
		Eigen::VectorXd xw = (*(optInst->samples_))*param;

		if( (xw.cols() != 1) || (xw.rows() != optInst->samples_->rows())){
			std::cerr << "Error, column size must be 1" << std::endl;
			std::cerr << "product of Xw is " << xw.rows() 
				<< " by " << xw.cols() << std::endl;
			std::exit(-1);
		}

		int effective_cnt = 0;
		Eigen::ArrayXd xwy(xw.rows());

		xwy.setZero();
		for(int i=0; i<xw.rows(); ++i){
			double z = xw.coeff(i)*(optInst->labels_->coeff(i));
			xwy.coeffRef(i) = z;
			if( z < 1 ){
				fx += (1-z)*(1-z);
				++effective_cnt;
			}
		}

		// fx /= optInst->labels_->rows();
		// l2 regularization value
		for(int i = optInst->prog_params_->l2start;
			i <= optInst->prog_params_->l2end;
			++i){
				fx += 0.5*optInst->prog_params_->l2c * std::pow(param.coeff(i), 2);
		}

		grad.setZero();

		for (int k = 0; k < optInst->samples_->outerSize(); ++k){

			double z = xwy.coeff(k);
			if(z >= 1){
				continue;
			}
			double factor = -2 * optInst->labels_->coeff(k) * (1-z);
			for (Eigen::SparseMatrix<double, Eigen::RowMajor>::InnerIterator it( *(optInst->samples_) ,k); 
				it; ++it){
					int colIdx = it.col();
					grad.coeffRef(colIdx) += factor*it.value();
			}
		}

		// grad /= optInst->samples_->rows();

		//l2 regularization
		for(int i = optInst->prog_params_->l2start;
			i <= optInst->prog_params_->l2end;
			++i){
				grad.coeffRef(i) += optInst->prog_params_->l2c * param.coeff(i);
		}

		std::cout << "One computation of function value & gradient costs " << t.toc() << " seconds"
			<< std::endl;
		return fx;
}
開發者ID:lacozhang，項目名稱:machinelearning，代碼行數:77，代碼來源:train.cpp


注：本文中的eigen::VectorXd::coeff方法示例由純淨天空整理自Github/MSDocs等開源代碼及文檔管理平台，相關代碼片段篩選自各路編程大神貢獻的開源項目，源碼版權歸原作者所有，傳播和使用請參考對應項目的License；未經允許，請勿轉載。