当前位置:
首页 资源下载
搜索资源 - regularization matlab
搜索资源列表
-
1下载:
matlab编写的图像处理正则化软件包。包含了基本的数字图像处理正则化-Matlab prepared by the image processing Regularization package. Contains a basic digital image processing Regularization
-
-
0下载:
基于正则化转置的LDA matlab源代码,可以应用在人脸识别中,详细说明见readme-based Regularization home to the LDA Matlab source code, can be applied to face recognition, detailed explanation see readme
-
-
0下载:
The MATLAB implementation of anistropic diffusion regularization used for smoothing image prior to interpolating
-
-
0下载:
This is Version 3.2 of Regularization Tools for Matlab 6.0 from Per Christian Hansen, IMM.
-
-
0下载:
L1General是一个求解各种L1-正则化问题的Matalb程序。详细说明可参考:http://www.di.ens.fr/~mschmidt/Software/L1General.html -L1General is a set of Matlab routines implementing several of the available strategies for solving L1-regularization problems.
-
-
0下载:
对图像进行正则化处理的matlab源程序,对学习正则化很有帮助-The image regularization process of matlab source code, learning regularization helpful
-
-
3下载:
利用L1范数TV正则化对影像进行超分辨率重建(Super resolution reconstruction of images using L1 norm TV regularization)
-
-
1下载:
正则化的目的:防止过拟合;正则化的本质:约束(限制)要优化的参数。(Regularization purpose: to prevent overfitting!
the nature of regularization: constraints (constraints) to optimize parameters.)
-
-
0下载:
解病态方程的正则化方法、逆问题不适定解的处理与正则化(Regularization method for solving ill conditioned equations)
-
-
0下载:
多种最优化方法的集合函数,包括吉洪诺夫正则化算法等。(mutilple method for inverse problem, mainly to solve ill-conditioned system of linear equations.including Tikhonov regularization method.)
-
-
0下载:
汉森的matlab正则化工具包,可用于求解正则化问题。(Hansen's Matlab regularization toolkit can be used to solve regularization problems.)
-
-
0下载:
simpson method and the other regularization method
-
-
0下载:
Regularization using MATLAB Toolbox
-
-
3下载:
吉洪诺夫正则化的matlab函数,可以自己选择参数值,调用即可(The matlab function of Tikhonov regularization)
-
-
1下载:
对多用户的mimo 规则化和非规则的预编码(Multi-user mimo regularization and non-regular pre-coding)
-
-
1下载:
深度神经网络在测试时面对如此大的网络是很难克服过拟合问题的。
Dropout能够很好地解决这个问题。通过阻止特征检测器的共同作用来提高神经网络的性能。这种方法的关键步骤在于训练时随机丢失网络的节点单元包括与之连接的网络权值。在训练的时候,Dropout方法可以使得网络变得更为简单紧凑。在测试阶段,通过Dropout训练得到的网络能够更准确地预测网络的输出。这种方式有效的减少了网络的过拟合问题,并且比其他正则化的方法有了更明显的提升。
本文通过一个简单的实验来比较使用Dropout方法前后网络
-
-
2下载:
这是用于l1正则化功能的bregman算法,主要用于图像去噪,去模糊,去卷积等(This is a Bregman algorithm for L1 regularization, which is mainly used for image denoising, blur, deconvolution, etc.)
-
-
7下载:
正则化代码,包括了常用的正则化方法以及L曲线求取最佳参数。(Matlab code for regularization)
-
-
0下载:
多核变换,稀疏变换,核变换自适应学习,图形图像变换(Descr iption
-----------
Multi-Kernel-Path package is a Matlab program that computes the entire regularization
path for multiple-kernel learning problems. More precisely, it solves any learning
problem with a smooth loss
-
-
5下载:
正则化工具,完整的正则化工具,包括各种正则化的处理,使用matlab编程语言(regularization tools)
-