搜索资源列表
mixing-matrix-estimate
- 二阶稀疏信号欠定盲分离,基于超平面法矢量的混合矩阵的估计-Second-order sparse underdetermined blind signal separation method based on hyperplane vector mixing matrix estimate
SparseVector
- Fast and memory efficient implementation of a sparse vector class in C++, including routines for platform-independent binary persistence.
RVM_matlabToolBox
- 相关向量机(RVM)的matlab源程序,包含快速算法,内含代码使用说明。 RVM采取是与支持向量机相同的函数形式稀疏概率模型,对未知函数进行预测或分类。 优点: (1) 不仅仅输出预测目标量的点估计值,还可以输出预测值的分布. (2) 使用更少数量的支持向量,从而显著减少输出目标量预测值的计算时间. (3) RVM不需要估计过多的参数. (4) RVM对是否满足Mercer 定理的核函数没有限制,适应性更好. -Relevance Vector Ma
AR_SKFA
- 在ar人脸库上的sparse kernel feature analysis程序,该程序挨个计算kernel空间鉴别向量,最后组成鉴别向量集进行分类-In the ar face database on the sparse kernel feature analysis program, which identified one by one calculation kernel space vector, the final set of vectors to classify the co
sparselib_1_6
- SparseLib + +是一个c++类库,为在不同的计算机平台进行有效稀疏矩阵的计算。这个软件组成对象包含了目前正在使用的几种稀疏矩阵存储格式(在这个版本:压缩行、压缩列和协调格式),提供基本的功能,节约稀疏矩阵存储空间和计算时间。-SparseLib++ is a C++ class library for efficient sparse matrix computations across various computational platforms. The software
vectprod
- helpful code for sparse matrix vector product
smvppp
- Fast sparse matrix vector product
CUDA-imization
- CUDA稀疏矩阵与矢量乘法的优化CUDA-based sparse matrix vector multiplication optimization-CUDA-based sparse matrix vector multiplication optimization
powermethod-andinversepowermethod
- 对于大型稀疏矩阵(501乘501),用幂法求其按模最大特征值和最大特征值,用反幂法求其按模最小特征值和距离给定数字最近的特征值,求得了矩阵的条件数和行列式,讨论迭代初始向量的选取对计算结果的影响。-For large sparse matrix (501 x 501), with the power method for the maximum modulus of its eigenvalues and by the largest eigenvalue, with
Sparse-Bayesian-Learning-and-rvm
- 稀疏贝叶斯学习和相关向量回归,大家可以看一下,人工智能中的应用-Relevance vector machine, new applications in the field of artificial intelligence, I hope you look at...
mi-(2)
- 幂法是一种计算矩阵主特征值(矩阵按模最大的特征值)及对应特征向量的迭代方法, 特别是用于大型稀疏矩阵。 -The power method is a method of calculating the matrix eigenvalue (matrix largest characteristic value) and the iterative method of the corresponding feature vector, particularly for large spars
Sp
- 这是一个建立在OMP算法基础上,笔者自己提出来的一个稀疏分解的DOA估计算法,利用阵列流形矢量间的不相关进行匹配追踪选择,选出与信源位置流形最接近的,以便求得DOA-This is a building based on the algorithm of OMP, the author himself a sparse decomposition algorithm for DOA estimation, using the array manifold a vector between the
achvectors
- Delphi 的矩阵库,可以实际动态数组,矩阵产实现搜索等功能-Vectors is a class library for Object Pascal (Delphi 1, 3-7, Borland C++ Builder, Borland Kylix, Free Pascal v.1.0 and higher) which implements vectors (dynamically sized arrays) based on different Object Pascal s
Sparse_matrix_and_vector
- Sparse matrix and vector product operation
spmv_csr
- 稀疏矩阵的DIA/ELLPACK/COO/CSR/HYB表示形式,以及各表示形式下的稀疏矩阵乘法(稀疏大矩阵*矢量)的CUDA实现。对于矩阵中每一行稀疏元素个数较统一的情况,ELLPACK表示最佳,其次是HYB(ELL+COO)。 CUDA™ 是一种由NVIDIA推出的通用并行计算架构,该架构使GPU能够解决复杂的计算问题。 它包含了CUDA指令集架构(ISA)以及GPU内部的并行计算引擎。 开发人员现在可以使用C语言来为CUDA™ 架构编写程序-Sparse matri
CSL1.0
- A Comprehensive Sparse Learning Package-Descr iptions: Six sparse algorithms are implemented in this package. They are 1) SVM 2) L1-norm SVM a) Linear Programming SVM b) SC-SVM 3) L0-norm SVM 4) RVM (Relevance Vector Machine) 5)
gmres
- gmres 方法。视图求解线性方程组A*x=b的解x。nXn的稀疏矩阵A必须是方程且应是大型稀疏矩阵。列向量b的长度必须为n。-gmres method. View solving linear equations A* x = b the solution x. nXn sparse matrix A must be equation and should be large sparse matrix. The length of the column vector b must be to
mul
- 稀疏矩阵的矢量乘方法,主要用于数值计算,矩阵方法中-Sparse matrix vector multiplication method
myomp
- 应用正交匹配追踪求解等式y=Ax,要求: 待求x是稀疏向量,A为高斯随机矩阵 调用形式:x = myomp(A,y,err) A -线性投影矩阵; y -投影向量 err -所需精度-apply Orthogonal matching pursuit to solve the equation y = Ax, requirements: the unknown x is sparse vector, A is a Gaussian random. ca
SPGL1: A solver for large-scale sparse
- % SPGL1: A solver for large-scale sparse reconstruction % Version 1.7 (May 20, 2009) % % Files % spg_bp - Solve the basis pursuit (BP) problem % spg_bpdn - Solve the basis pursuit denoise (BPDN) problem % spg_lasso