搜索资源列表
tsvd
- 截断奇异值分解技术是一种正则化方法,用于不适定问题的数值计算-Truncated Singular Value Decomposition is a regularizaiton method for ill-conditioned problem computation.
demoBagSVM
- 一种基于半监督的svm的图像分类方法。该方法通过聚类核的方法利用无标记样本局部正则化训练核的表达式。这种方法通过图像直接学习一个自适应的核。该程序仿真的是文章:Semi-supervised Remote Sensing Image Classification with Cluster Kernels。大家可以参考下。-A semi-supervised SVM is presented for the classification
borninv
- 本程序实现1DBorn近似反演,并且采用正则化,使得反演稳定。-This program implements 1DBorn approximate inversion, and using regularization, making inversion stability.
SpaRSA_2.0
- 求解优化问题中L1范数正则化函数和全变分正则化函数的流行算法-Sparse Reconstruction by Separable Approximation
reconstruction_algorithms
- 本代码主要给出了激光粒度仪颗粒散射光强分布以及4种粒度反演算法,以及4种算法之间的比较。四种反演算法为:TSVD、Chaine、Tikhonov和l1正则化。-The code gives the Zetasizer particle scattering intensity distribution, and four kinds of particle inversion algorithm, as well as a compar
BP_nerual_network
- 通过gdm、LM以及贝叶斯正则化,进行BP神将网络-By gdm, LM and Bayesian regularization, for God BP Network
Regularization-method
- 基于正则化方法的高频地波雷达海浪方向谱反演-Regularization method based on the HF radar wave directional spectrum inversion
moniPNAH
- 应用某旋转阵列进行声全息的等效源计算程序,加上正则化及相位计算部分就可以工程应用了。-A work acoustic array function. A mfile of NAH by ESM. please add the regulation function for actually work.
gcv
- 广义交叉验证函数(GCV) 广义交叉验证函数是用随机的方法对于一个给定的值A的正则化参数的计算。-Generalized cross-validation function (GCV) generalized cross validation function is stochastic method for a given value A regularization parameter calculations.
LinearRegression
- 利用numpy库实现线性回归 并且带有正则化之后的线性回归处理功能 还有用最小角回归(LARS)算法实现了lasso回归-Numpy library implements the use of linear regression and linear regression with regularization after processing algorithms as well as the minimum angle of reg
Low-patch-tv-denoising
- 采用低块秩的正则化方法,对规则图像进行去噪,能达到较好效果-denoising using low-patch regularizationl and TV leading to a good result
L1_norm_least_square_solver
- L1范数正则化最小二乘问题求解器,stanford大学Boyd教授出品。-Simple Matlab Solver for l1-regularized Least Squares Problems
regularization
- 正则化(regularization),是指在线性代数理论中,不适定问题通常是由一组线性代数方程定义的,而且这组方程组通常来源于有着很大的条件数的不适定反问题。大条件数意味着舍入误差或其它误差会严重地影响问题的结果。-Regularization, in mathematics and statistics and particularly in the fields of machine learning and inverse pr
BP_LM
- 采用贝叶斯正则化算法提高 BP 网络的推广能力。在本例中,我们采用两种训练方法,即 L-M 优化算法(trainlm)和贝叶斯正 -Bayesian regularization algorithm to improve the generalization ability of BP network. In this example, we use two training methods, namely LM optimizatio
Regularization-package
- 正则化方法求解病态问题工具包,非常强大~~~~~~~~其中包含直接解法和迭代解法,共十余种。-Regularization method for solving ill-posed problem toolkit, very powerful ~ ~ ~ ~ ~ ~ ~ ~
DRLSE_v0
- 本程序包采用距离正则化的水平集的方法,对多种图像进行处理和分割,实验结果表明,分割结果比传统水平集方法好很多-This package uses the distance regularized level set method, a variety of image processing and segmentation, experimental results show that the segmentation result i
beiyesizhengzehua
- 采用贝叶斯正则化算法提高 BP 网络的推广能力,采用两种训练方法-Bayesian regularization algorithm to improve the generalization ability of BP network, using two training methods
ROMP
- 该算法首先根据相关原子挑选多个原子作为候选集,然后从候选集中按照正则化原则挑选出部分原子,最后将其并入最终的支撑集,从而实现了原子的快速、有效选择-Firstly, a plurality of atoms selected on the basis of the candidate sets as atoms, and from the candidate set according to the principle of the s
Tikhonov-in-pMRI
- Tikhonov正则化在并行磁共振图像重建中的应用-Tikhonov regularization in parallel magnetic resonance image reconstruction
demo_sparse_tv
- 基于全变分空间正则化的稀疏超普解混算法的MATLAB程序-Matlab programs for “Total variation spatial regularization for sparse hyperspectral unmixing”