## SDPT3 version 4.0: MATLAB/Octave software for semidefinite-quadratic-linear programming
### [Kim-Chuan Toh](http://www.math.nus.edu.sg/~mattohkc/index.html), [Michael J. Todd](http://people.orie.cornell.edu/miketodd/todd.html), and [Reha H. Tütüncü](http://www.math.cmu.edu/~reha/)
SDPT3 is a Matlab package for solving convex optimization problems involving linear equations and inequalities, second-order cone constraints, and semidefinite constraints (linear matrix inequalities).
This is an *unofficial* repository for SDPT3. The [official SDPT3 site](http://www.math.nus.edu.sg/~mattohkc/sdpt3.html) is administered by [Kim-Chuan Toh](http://www.math.nus.edu.sg/~mattohkc/index.html), co-author of SDPT3 and professor of mathematics at the National University of Singapore.
This repo is administered by [Michael Grant](http://cvxr.com/bio), the developer of [CVX](http://cvxr.com/cvx), a modeling framework for convex optimization that uses SDPT3 as a solver. *It is not our intent for this repo to become an independent fork of SDPT3*. To that end:
+ We will periodically monitor the [official site](http://www.math.nus.edu.sg/~mattohkc/sdpt3.html) for updates, and incorporate any we find into this repo.
+ Any improvements that we make will be submitted to the authors for inclusion in the official release.
Having said that, this distribution *does* differ from the release currently posted on the main site in several ways:
+ *Preliminary* support for Octave 3.8.0 and later has been added. While we have found that it works well,
you should expect an occasional error or incompatibility simple due to a lack of testing.
+ The installer has been replaced. The new installer will build the MEX files only if they are not already
present (or unless a "-rebuild" flag is given). It will also set the MATLAB paths, if necessary.
*At this point, Octave users will need to build their own MEX files.*
+ The ".dll" files have been removed. Those using very old versions of MATLAB will need to rebuild them.
The maintainer no longer keeps a version of MATLAB that old, so they can no longer be verified.
+ The code has been reformatted and cleaned up to eliminate all warnings from MATLAB's Code Analyzer.
+ Old files that are no longer in use have been removed. (That's what a repo is for!)
While you are welcome to submit bug reports on the [GitHub issue page](https://github.com/mcg1969/SDPT3/issues) for this repo, we cannot guarantee that they will be addressed in a timely fashion.
This version of SDPT3 is distributed under the GNU General Public License 2.0. For more details, please see the included files [Copyright](https://github.com/mcg1969/SDPT3/blob/master/Copyright) and [GNU\_General\_Public\_License\_v2](https://github.com/mcg1969/SDPT3/blob/master/GNU_General_Public_License_v2).
Michael C. Grant
[CVX Research, Inc.](http://cvxr.com)
没有合适的资源?快使用搜索试试~ 我知道了~
温馨提示
最小二乘法简单求解, 最小二乘法是回归分析中的一种标准方法,通过最小化残差的平方和(残差是观察值和模型提供的拟合值)在每个单独方程的结果中得出。 最重要的应用是数据拟合。当问题在自变量(x变量)中有很大的不确定性时,简单回归和最小二乘法就会出现问题;在这种情况下,可以考虑拟合变量误差模型所需的方法,而不是最小二乘法。 最小二乘问题分为两类:线性或普通最小二乘和非线性最小二乘,这取决于残差在所有未知数中是否是线性的。线性最小二乘问题出现在统计回归分析中;它有一个封闭形式的解决方案。非线性问题通常通过迭代细化来解决;在每次迭代中,系统都近似为线性系统,因此两种情况下的核心计算都是相似的。 多项式最小二乘法将因变量预测中的方差描述为自变量的函数以及与拟合曲线的偏差。 当观测来自一个指数族,其自然充分统计量和温和条件得到满足(例如,对于正态分布、指数分布、泊松分布和二项分布),标准化最小二乘估计和最大似然估计是相同的。[1]最小二乘法也可以作为矩估计法推导出来。 以下讨论主要是根据线性函数提出的,但最小二乘法的使用对于更一般的函数族是有效和实用的。此外,通过迭代地将局部二次近似应用
资源推荐
资源详情
资源评论
收起资源包目录
最小二乘法简单求解(附matlab源代码和过程可视化,容易理解) (1404个子文件)
symfct.c 76KB
dpr1fact.c 36KB
spscale.c 35KB
ordmmd.c 34KB
triuaux.c 29KB
blkchol2.c 25KB
getada3.c 25KB
mexsvec.c 24KB
urotorder.c 20KB
blkchol.c 19KB
mexschur.c 16KB
symbfwblk.c 16KB
mexsmat.c 15KB
reflect.c 14KB
incorder.c 14KB
ddot.c 13KB
fwblkslv.c 12KB
bwblkslv.c 12KB
getada1.c 12KB
qrK.c 11KB
bwdpr1.c 11KB
mexmat.c 11KB
adendotd.c 11KB
symfctmex.c 10KB
iswnbr.c 10KB
mexProd2.c 10KB
findblks.c 10KB
finsymbden.c 9KB
psdinvjmul.c 9KB
getada2.c 9KB
mexskron.c 9KB
fwdpr1.c 8KB
sdmauxCmp.c 8KB
cholsplit.c 8KB
choltmpsiz.c 7KB
cvx_eliminate_mex.c 7KB
extractA.c 7KB
auxgivens.c 7KB
symbchol.c 7KB
adenscale.c 7KB
psdframeit.c 7KB
givensrot.c 7KB
vecsym.c 6KB
mexProd2nz.c 6KB
sdmauxCone.c 6KB
invcholfac.c 6KB
partitA.c 6KB
sdmauxTriu.c 6KB
bwblkslv2.c 6KB
ordmmdmex.c 5KB
mextriang.c 5KB
sortnnz.c 5KB
sqrtinv.c 5KB
mexinprod.c 5KB
mexMatvec.c 5KB
quadadd.c 5KB
cvx_bcompress_mex.c 5KB
blkmul.c 5KB
qblkmul.c 5KB
auxfwdpr1.c 4KB
mexqops.c 4KB
mexschurfun.c 4KB
blkaux.c 4KB
mextriangsp.c 4KB
mJdetd.c 3KB
sdmauxFill.c 3KB
sdmauxTriudot.c 3KB
sdmauxScalarmul.c 3KB
sdmauxRdot.c 2KB
mexexpand.c 2KB
mexnnz.c 2KB
COPYING 18KB
Copyright 1KB
cloud.css 16KB
basic.css 8KB
pygments.css 4KB
examples.css 3KB
qap9.dat-s 227KB
truss8.dat-s 204KB
theta3.dat-s 202KB
ss30.dat-s 160KB
arch8.dat-s 80KB
qpG11.dat-s 67KB
maxG11.dat-s 59KB
mcp250-1.dat-s 16KB
gurobi90.dll 30.54MB
mosek64_9_1.dll 19.55MB
cilkrts20.dll 306KB
grbgetkey.exe 14.67MB
grbprobe.exe 14.46MB
ajax-loader.gif 673B
plus.gif 89B
minus.gif 87B
bullet.gif 62B
GNU_General_Public_License_v2 18KB
blksdp.h 8KB
triuaux.h 3KB
reflect.h 2KB
givens.h 2KB
funcref.html 51KB
共 1404 条
- 1
- 2
- 3
- 4
- 5
- 6
- 15
资源评论
- 色空空色2023-07-28这个文件很简洁明了,我按照步骤一步步实践,成功解决了最小二乘法问题,非常满意。
- 申增浩2023-07-28这个文件的代码非常清晰易懂,通过作者的过程可视化,我对最小二乘法的求解方式有了更深刻的认识。
- 艾苛尔2023-07-28这个文件非常实用,我通过它轻松学会了最小二乘法的求解,非常感谢作者的分享。
- 是因为太久2023-07-28这个文件的价值远远超出了我的预期,我以前对最小二乘法一直有疑惑,但通过这个文件,我终于找到了一个简单而有效的解决方案。非常感谢作者的分享。
- 思想假2023-07-28这个文件的作者真是太厉害了,不仅提供了matlab源代码,还用简洁的语言解释了整个求解过程,让我能够轻松理解并应用到我的问题中。
医学信号图像处理匠人
- 粉丝: 5851
- 资源: 32
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功