Location:
Search - www.linear.
Search list
Description: This MEX performs 2d bilinear interpolation using an NVIDIA graphics chipset. To compile and run this software, one needs the NVIDIA CUDA Toolkit (http://www.nvidia.com/object/cuda_get.html) and, of course, an NVIDIA graphics card of reasonably modern vintage.
BUILDING INSTRUCTIONS: Change the 'MATLAB' (and if necessary, 'MEX') variables in the Makefile to appropriate values, then simply run 'make' at a prompt and an executable (mex/mexmac/mexmaci/dll?) file will be created.
This code uses your GPU's built-in bilinear texture interpolation capability, and is very fast. For reasonably sized operations (taking, say, a 50x50 matrix up to 1000x1000) CUDA-based code is 5-10x faster than linear interp2 (as tested on a MBP 2.4GHz C2D, GeForce 8600M GT).
With very (VERY) large matrices, however, it has the capability of completely crashing your computer or giving bizarre results. Be careful!
Platform: |
Size: 37881 |
Author: whitewalter |
Hits:
Description: ************************************************************************ * * * * * THIS IS THE H Y P L A S 2.0 README FILE * * ----------------- * * * * HYPLAS is a finite element program for implicit small and large * * strain analisys of hyperelastic and elasto-plastic two-dimensional * * and axisymmetric solids * * * * HYPLAS v2.0 is the companion software to the textbook: * * EA de Souza Neto, D Peric & DRJ Owen. Computational Methods for * * Plasticity: Theory and Applications. Wiley, Chichester, 2008. * * (www.wiley.com/go/desouzaneto) * * * * Copyright (c) 1998-2008 EA de Souza Neto, D Peric, D.R.J. Owen * *----------------------------------------------------------------------* * File last updated: 18 October 2008 * * * * This file belongs in the directory ../HYPLAS_v2.0 * ************************************************************************ * * * I M P O R T A N T * * * * READ SECTIONS 0 TO 3 OF THIS FILE CAREFULLY BEFORE ATTEMPTING * * TO COMPILE AND RUN THE PROGRAM HYPLAS ON YOUR COMPUTER !! * * * * THE AUTHORS DO NOT GUARANTEE THAT ANY SUGGESTIONS/INSTRUCTIONS * * GIVEN IN THIS README FILE WILL WORK ON ANY PARTICULAR OPERATING * * SYSTEM. IF YOU DECIDE TO FOLLOW ANY SUGGESTIONS/INSTRUCTIONS * * GIVEN HERE YOU MUST DO SO AT YOUR OWN RISK. * * * * * * BUG REPORTS: Please send bug reports to * * * * hyplas_v2.0@live.co.uk * * * * Messages sent to the authors' personal email addresses * * will NOT be answered. * ************************************************************************ This file contains the following sections: 0. Copyright statement and disclaimer 0.(a) Copyright statement 0.(b) Disclaimer 0.(c) Conditions of use 1. Introduction 1.(a) Note on portability 2. Compiling and running HYPLAS 2.(a) Memory requirements 2.(b) Testing a newly compiled executable 3. The HYPLAS directory tree 4. Cross-referencing between the source code and the textbook 5. HYPLAS error messaging 6. Further remarks on HYPLAS ************************************************************************ 0. COPYRIGHT STATEMENT AND DISCLAIMER ================================== 0.(a) Copyright statement ------------------- You may only use this program for your own private purposes. You are not allowed, in any circumstances, to distribute this program (including its source code, executable and any other files related to it, either in their original version or any modifications introduced by you, the authors or any other party) in whole or in part, either freely or otherwise, in any medium, without the prior written consent of the copyright holders. 0.(b) Disclaimer ---------- This program (including its source code, executable and any other files related to it) is provided "as is" without warranty of any kind, either expressed or implied, including, but not limited to, any implied warranties of fitness for purpose. In particular, THIS PROGRAM IS BY NO MEANS GUARANTEED TO BE FREE FROM ERRORS. This program (or any modification incorporated to it by you, the authors or any other party) will run entirely at your risk. The results produced by this program are in no way guaranteed to be fit for any purpose. Under no circumstances will the authors/copyright holders be liable to anyone for damages, including any general, special, incidental or consequential damages arising from the use or inability to use the program (including, but not limited to, loss or corruption of data, failure of the program to operate in any particular way as well as damages arising from the use of any results produced by the program for any purpose). 0.(c) Conditions of use ----------------- You may only use this program if you fully understand and agree with the terms of the above disclaimer. You must not use this program if you do not agree with or do not understand (fully or in part) these conditions of use. 1. INTRODUCTION ============ HYPLAS is a finite element code for small and large strain analysis of hyperelastic and elasto-plastic solids. Most procedures implemented in HYPLAS are described in detail in its companion textbook: EA de Souza Neto, D Peric & DRJ Owen. Computational Methods for Plasticity: Theory and Applications. Wiley, Chichester, 2008 (www.wiley.com/go/desouzaneto). 1.(a) Note on Portability ------------------- HYPLAS has been written in standard ANSI FORTRAN 77. Currently, the only known (and deliberate) exceptions to the FORTRAN 77 ANSI standard are the instructions: INCLUDE '' used in many routines to include the HYPLAS database files (common blocks and global variables), and; CALL GETENV('HYPLASHOME',HYPLASHOME) used in subroutine "ERRPRT" (file ../HYPLAS_v2.0/src/GENERAL/errprt.f). This instruction inquires the name of the system environment variable HYPLASHOME and writes it on the character string HYPLASHOME. This instruction is NOT part of the ANSI FORTRAN 77 standard, but seems to work in most currently available FORTRAN 77 compilers. 2. COMPILING AND RUNNING H Y P L A S ================================== The HYPLAS source code is stored in directory ../HYPLAS_v2.0/src/ (../HYPLAS_v2.0/ being the current directory) and all its subdirectories. To generate an executable file, you just need to compile the FORTRAN source files: ../HYPLAS_v2.0/src/hyplas.f and ../HYPLAS_v2.0/src/*/*.f together. We recommend that the executable HYPLAS be stored in the directory ../HYPLAS_v2.0/bin to which the environment variable HYPLASHOME should be set (see below how to set a system environmental variable). WINDOWS (R) systems ------------------- On Microsoft Windows(R) systems, HYPLAS has been successfully compiled using Intel Visual Fortran Compiler(R) integrated with Microsoft Visual Studio(R). Here you only need to create a project that contains all Fortran source files mentioned above as well as the include files ..\HYPLAS_v2.0\src\*.INC On a Windows XP system, the system environment variable HYPLASHOME can be set as follows: 1. Open a File Manager 2. Right-click on the "My Computer" icon 3. Select "Properties" on the drop-down menu 4. A new window named "System Properties" will pop-up. Here select the "Advanced" tab. 5. On the "Advanced" tab, click the "Environment Variables" button. 6. A new window titled "Environment Variables" will pop-up. Here click the button "New" in the "System Variables" section of the window. 7. A new window will pop-up titled "New System Variable". Here you should fill the fields "Variable name" and "Variable Value", respectively, with HYPLASHOME and the path name (in full) of the directory ..\HYPLAS_v2.0\bin. 8. Press "OK" on the relevant pop-up windows. 9. The next time the computer is REBOOTED, this variable will be set to the correct path and HYPLAS should be able to find the error messages file ERROR.RUN if required. UNIX/LINUX systems ------------------ In a UNIX/LINUX operating system using a C-shell, for instance, the HYPLASHOME environment variable should be set with the command: setenv HYPLASHOME where here denotes the full path to the directory ../HYPLAS_v2.0/bin. To compile HYPLAS (from directory ../HYPLAS_v2.0/src) with a FORTRAN 77 compiler such as g77, you can use the command: g77 -o ../bin/hyplas hyplas.f */*.f Note that the executable file "hyplas" will be stored in the directory ../HYPLAS_2.0/bin (i.e. the directory set in the HYPLASHOME environment variable). Alternatively, you may use the Makefile provided (with suitable modifications, if needed) to create the HYPLAS executable. IMPORTANT: Before generating a HYPLAS executable, read Sections 2.(a) and 2.(b) below. 2.(a) Memory Requirements ------------------- HYPLAS memory requirements depend on the array dimensioning parameters set in files: ../HYPLAS_v2.0/src/ ELEMENTS.INC GLBDBASE.INC MATERIAL.INC MAXDIM.INC Files ELEMENTS.INC, GLBDBASE.INC and MATERIAL.INC contain parameters which are associated with the currently implemented finite elements and materials. DO NOT MODIFY THEM ! unless you are absolutely sure of what you are doing (only developers coding new elements or new material models/analysis types may need to modify them by changing the existing dimensioning parameters and/or including new parameters). The ONLY dimensioning file that can be safely modified by the average user is the file MAXDIM.INC This file contains the array dimensioning parameters related to the maximum permissible dimension of problems to be analysed by HYPLAS. These parameters include the maximum number of nodes, elements, element groups, etc. If necessary, CHANGE THESE PARAMETERS TO SUIT YOUR PROBLEM SIZE/MEMORY REQUIREMENTS before compiling HYPLAS. 2.(b) Testing a newly compiled executable ----------------------------------- After you have successfully compiled the HYPLAS source code and created an executable file, the next step is to run some tests to verify that HYPLAS is working well. To do this, proceed as follows: The directory ../HYPLAS_v2.0/book_examples/data_files contains a series of data files named .dat of benchmarked examples described in the companion textbook. The corresponding (benchmarked) result files are in the directory ../HYPLAS_v2.0/book_examples/result_files This directory contains a series of result files named .res generated with the current version of HYPLAS on a tested platform. All these files have been named such that their names start with the textbook section number where the corresponding example is described. For instance, files 14_9_2_tresca.dat and 14_9_2_tresca.res refer to a problem described in section 14.9.2 of the textbook, and so on. To check that HYPLAS is working well on your platform, after compiling HYPLAS, run the program HYPLAS for the examples of files .dat and compare the newly generated results .res with their benchmarked counterparts (of the same filename) in the result_files directory. To run an example, execute HYPLAS and use the keyboard to enter the name of the corresponding data file in full (including the extension .dat). To compare the benchmarked .res files against their newly generated you may proceed as follows: 1. On MICROSOFT WINDOWS systems - Here we have successfully used the software "ExamDiff" (the task was made particularly easy by selecting "View" and then the "Show Differences Only" option - this refers to version 1.8 of this software). 2. On UNIX/LINUX systems - Here we use the "diff" command from a shell window (and set the option to ignore blank spaces). A shell script may be used to perform this task automatically (including running HYPLAS and checking for result file differences) for all benchmarked examples provided. IMPORTANT: THE ONLY ACCEPTABLE DIFFERENCES BETWEEN A THE NEWLY GENERATED RESULT FILES AND THEIR BENCHMARKED COUNTERPARTS ARE THE DIMENSIONING PARAMETERS (FROM FILE MAXDIM.INC) USED TO COMPILE THE NEW EXECUTABLE (THESE PARAMETERS ARE PRINTED RIGHT AT THE BEGINNING OF THE RESULT FILES) AND NUMERICAL DIFFERENCES IN RESULTS DUE TO NUMERICAL "ROUNDING-OFF" (THESE ARE VERY SMALL DIFFERENCES THAT DEPEND ON THE PRECISION OF ARITHMETIC OPERATIONS IN THE PLATFORM USED). ALSO NOTE THAT THE EXAMPLES OF THE COMPANION TEXTBOOK DO NOT COVER ALL FEATURES OF HYPLAS. HENCE THIS TEST DOES NOT GUARANTEE THAT EVERYTHING IS WORKING PROPERLY. 3. THE H Y P L A S DIRECTORY TREE ================================ 3.(a) Summary ------- ../ HYPLAS_v2.0/ bin/ book_examples/ data_files/ result_files/ man/ html/ src/ CRYSTAL/ DAMAGE/ DAMAGED_ELASTIC/ DRUCKER_PRAGER/ ELASTIC/ ELEMENTS/ GENERAL/ MATERIALS/ MATHS/ MOHR_COULOMB/ OGDEN/ TRESCA/ VON_MISES/ VON_MISES_MIXED/ 3.(b) Description ----------- The HYPLAS program directory tree is organised as follows: ../HYPLAS_v2.0/ (this directory) This is the HYPLAS root directory, where the HYPLAS directory tree starts. ../HYPLAS_v2.0/bin/ This directory contains the file ERROR.RUN where most HYPLAS error/warning messages are. IMPORTANT: the environment variable HYPLASHOME should be set to this directory. Otherwise, HYPLAS will not find its error/warning messages when required. We also recommend that the EXECUTABLE of HYPLAS be stored in this directory. ../HYPLAS_v2.0/book_examples/ This directory has the following subdirectories: ../HYPLAS_v2.0/book_examples/data_files ../HYPLAS_v2.0/book_examples/result_files Refer to Section 2.(b) above for further details. ../HYPLAS_v2.0/man/ This is the HYPLAS documentation/manuals directory. It contains the following files: input_man.txt - A concise input data manual for HYPLAS in ASCII format; hyplas_calltree.txt - Contains a flowgraph (shows the call tree) of HYPLAS in ASCII-format. Note: calls to function subprograms are not included in this flowgraph; and the subdirectory: ../HYPLAS_v2.0/man/html This directory contains the hypertext (HTML) format Fortran source code and of manual pages of the entire HYPLAS program. Manual pages with descriptions of each function/subprogram including their argument list are linked to their corresponding HTML-format source code. This allows the user the navigate through the HYPLAS source code using a web browser. To start at the main program, use your web browser to open the file hyplas.html. This facility should be helpful to those trying to understand the flow of program HYPLAS. ../HYPLAS_v2.0/src/ This directory (and its subdirectories) contains the Fortran source code of HYPLAS. The files containing the sources are named following the standard practice: .f where is the name of the FORTRAN procedure (subroutine, function subprogram, etc.) whose source code is in file .f. The source code of the HYPLAS main program is in file hyplas.f and the HYPLAS database (COMMON blocks, array dimensioning parameters and other global parameters) is coded in the "include files" ELEMENTS.INC GLDBASE.INC MATERIAL.INC MAXDIM.INC in this directory. In addition, this directory contains a file named "Makefile" (UNIX-LINUX Release only) which may be used for compiling and linking HYPLAS in UNIX/LINUX systems. The subdirectories of ../HYPLAS_v2.0/src are as follows: ../HYPLAS_v2.0/src/CRYSTAL Contains the source code of all procedures related to the finite strain single crystal plasticity model implemented in HYPLAS. ../HYPLAS_v2.0/src/DAMAGE Source files of the procedures related to the Lemaitre ductile damage model implementation. ../HYPLAS_v2.0/src/DAMAGED_ELASTIC Source files of the procedures related to the damaged elasticity model with crack closure effect. ../HYPLAS_v2.0/src/DRUCKER_PRAGER Source files of the procedures related to the implemented Drucker-Prager plasticity model. ../HYPLAS_v2.0/src/ELASTIC Source files of the procedures related to the linear elasticity model (Hencky model under large strains) implemented. ../HYPLAS_v2.0/src/ELEMENTS Source files of the element interfaces and element-related procedures. ../HYPLAS_v2.0/src/GENERAL Source files of general procedures. ../HYPLAS_v2.0/src/MATERIALS Source files of the material interfaces. ../HYPLAS_v2.0/src/MATHS Source files of the mathematical procedures. ../HYPLAS_v2.0/src/MOHR_COULOMB Source files of the procedures related to the implemented Mohr-Coulomb plasticity model. ../HYPLAS_v2.0/src/OGDEN Source files of the procedures related to the implemented Ogden hyperelasticity model. ../HYPLAS_v2.0/src/TRESCA Source files of the procedures related to the implemented Tresca plasticity model. ../HYPLAS_v2.0/src/VON_MISES Source files of the procedures related to the implemented von Mises plasticity model with isotropic hardening. ../HYPLAS_v2.0/src/VON_MISES_MIXED Source files of the procedures related to the implemented von Mises plasticity model with mixed isotropic/kinematic hardening. 4. CROSS-REFERENCING BETWEEN THE SOURCE CODE AND THE TEXTBOOK ========================================================== Many references are made in the textbook to various subprograms of HYPLAS. These are usually made when a particular procedure described in the text is implemented in the program. The reader should refer to the textbook index. Also, a substantial number of comment lines have been added to the source code of HYPLAS with reference to sections, figures, boxes, etc of the textbook related to the part of the code in question. Such references are usually displayed after the word "REFERENCE:" (in capitals) on commented lines. Searching for this word will take you to the line of code where the particular routine has a reference to the textbook. NOTE: Occasional references to other textbooks/journal papers are also made following the word "REFERENCE:" on commented lines. 5. HYPLAS ERROR MESSAGING ====================== Most error/warning messages issued by HYPLAS are in the ASCII-format file ERROR.RUN (kept in the HYPLASHOME directory - ../HYPLAS_v2.0/bin). All such error/warning messages have an identification code (e.g. ED0015) which is printed both to the standard output (this is usually the computer screen) and to the relevant results file. If you wish to find where in the source code a particular message is being issued, then perform a search for the corresponding message identification code in the entire source code of HYPLAS. 6. FURTHER REMARKS ON HYPLAS ========================= 6.(a) Program efficiency THIS SECTION IS OF INTEREST ONLY TO THOSE WANTING TO MAKE HYPLAS RUN FASTER. It is particularly stressed in the textbook that this program has not been designed having efficiency in mind (refer to Section 5.1.2 of the textbook). Its structure has been designed mainly to illustrate in a relatively clear manner the computer implementation of the techniques and algorithms described in the text, with a particular view to the implementation of solid constitutive models and finite elements. For those who are especially interested in the speed of the code, there are a few tips that could help in this direction. Unfortunately, these involve modifications to the source code which is probably most appropriate to readers with a good level of experience in finite element programming. To those with this particular interest, we can suggest the following: (i) The use of faster linear solvers This is probably the change that would result in a greater gain in efficiency. The Frontal Method adopted in subroutine FRONT (file ../HYPLAS_v2.0/src/GENERAL/front.f) has been designed originally to save memory (back in the days when computer memory was severely limited). There are currently a vast number of methodologies which focus on speeding up the linear solution, in addition to reducing memory storage requirements (which is a particularly important issue in the solution of large scale problems). Some of these are extensions/refinements of the original Frontal solver. We remark that a number of such procedures (with their respective source codes) are available (conditions may apply) from the LAPACK (Linear Algebra PACKage - http://www.netlib.org/lapack) repository or from the HSL Library (http://www.cse.cse.scitech.ac.uk/nag/hsl). For the reader interested in gaining speed, we would recommend the replacement of the existing solver of FRONT by a faster one. We remark though that this is a substantial programming task. Another aspect here is the fact that computing times in FRONT are directly linked to the frontwidth of the system which, in the present version of HYPLAS is fixed and depends, for a given mesh, on how the degrees of freedom are numbered (node numbering). The incorporation of a frontwidth optimiser (which re-numbers the degrees of freedom in order to minimise the frontwidth) in FRONT could produce some good savings in computing times. Such savings become particularly noticeable in larger problems where the original node numbering produces an excessively large frontwidth. (ii) Material-specific computations The issues pointed out here affect only the computing times for specific material models and are expected to have a much lower impact in overall speed than the linear solver issue discussed above. Some of the material model-specific computations carried out in HYPLAS could be made a bit faster. For example, for isotropic models whose stress update is carried out in the principal stress space (such as the Tresca and Mohr-Coulomb models - see routines SUTR and SUMC, files ../HYPLAS_v2.0/src/TRESCA/sutr.f and ../HYPLAS_v2.0/MOHR_COULOMB/sumc.f, respectively) the spectral decomposition of the stress in carried out in the state update update routine and then repeated in the corresponding routine for computation of the consistent tangent operator (refer to files ../HYPLAS_v2.0/src/TRESCA/cttr.f and ../HYPLAS_v2.0/src/MOHR_COULOMB/ctmc.f, respectively, for the Tresca and Mohr-Coulomb plasticity models). Some savings in computing time can be achieved here by storing the stress eigenprojection tensors (these can be stored as state variables) during the execution of the state updating and then retrieving them later for use in the computation of the consistent tangent operator. This change can be incorporated to the code relatively easily. The computation of the exponential map and is derivative for the single crystal plasticity model (routines EXPMAP, file ../HYPLAS_v2.0/src/CRYSTAL/expmap.f and DEXPMP, file ../HYPLAS_v2.0/src/CRYSTAL/dexpmp.f) is carried out in three dimensions (these routines have been adapted from an earlier three-dimensional code). To improve efficiency, these can be adapted to work only in two-dimensional problems by removing the unnecessary operations related to the third dimension. 6.(b) Output of nodal averaged values The reader should be aware that the way in which nodal averaged values of stresses and other variables are calculated in HYPLAS is very basic (and rudimentary). This feature of the program is made available only to help those interested in producing contour plots, etc from results presented in HYPLAS result files and should be useful in many circumstances of interest. This facility has in fact been used in producing many of the figures presented in the textbook. But note, for example, that the values of incremental plastic multipliers for plasticity models may take (inadmissible) negative values when extrapolated from Gauss-point to nodes and averaged. We remark that more sophisticated and refined techniques of transferring Gauss point values of variables to nodal points and obtaining the corresponding smoothed field are available in the current literature. These fall outside the scope of the companion textbook of HYPLAS.
Platform: |
Size: 11008084 |
Author: gtcewli3 |
Hits:
Description: 一个不错的线性规划问题求解的程序-a good linear programming problem solving process
Platform: |
Size: 3072 |
Author: 宋芳芳 |
Hits:
Description: 利用闭散列方法中的线形探查建立散列表。用c实现的。-closed hash method using the linear exploration establish hashtable. C used to achieve.
Platform: |
Size: 31744 |
Author: 曾德胜 |
Hits:
Description: Gauss列主元消去法解线性方程组.mylu函数中，U0、L0、P0分别为Gauss列主元分解中得U，L，P.-out PCA Gauss elimination method for solving linear equations. Mylu function, U0, L0, P0 Gauss out of the main yuan decomposition in the U, L, P.
Platform: |
Size: 1024 |
Author: jiangyi |
Hits:
Description: 这是一个线性表的问题但是我不知道是不是对的,请大家看看,-This is a linear form of the problem but I do not know is not right, please take a look at everyone,
Platform: |
Size: 1024 |
Author: 潘妍妍 |
Hits:
Description: 这个程序是关于线性网络化求解问题，算法实现上比较有特点。-This process is on the linear network to solve the problem, the algorithm has the characteristics to achieve more.
Platform: |
Size: 2048 |
Author: stone |
Hits:
Description: 美国标准化组织和马里兰大学共同开发.实现java matrix包.
JAMA由六个java类组成:Matrix, CholeskyDecomposition , LUDecomposition QRDecomposition SingularValueDecomposition EigenvalueDecomposition .
它提供了广泛的构造函数,丰富的get,set访问子matrix和元素matrix.
基本的矩阵操作有:矩阵的加,乘,矩阵范数.和提供了方便的打印矩阵的方法.
提供了五个基本的矩阵分离类.可以通过Matrix类调用它们来进行线性计算,翻转和其它矩阵方法.
-The United States Organization for Standardization and the joint development of MD University. Realize java matrix package. JAMA composed by six java category: Matrix, CholeskyDecomposition, LUDecomposition QRDecomposition SingularValueDecomposition EigenvalueDecomposition. It provides a wide range of structural function, the rich get, set to visit sub-matrix and element matrix. Have basic matrix operations: matrix increase, multiplication, matrix norm. and provides a convenient matrix print method. provides a separation of five basic types of matrix. can call Matrix category for their linear terms, matrix inversion and other methods.
Platform: |
Size: 104448 |
Author: 孙娜 |
Hits:
Description: Linear Neighborhood Propagation 系数求解-Linear Neighborhood Propagation coefficients
Platform: |
Size: 1024 |
Author: 唐金辉 |
Hits:
Description: 返回
第一章 关于寿星万年历
第二章 公历与儒略日
第三章 回历计算
第四章 天文坐标
第五章 近现代农历的天文学性质及其规则
第六章 离散序列的直线拟合算法
第七章 关于deltat T的计算
使用方法
显示某日的信息摘要后，如果按住ctrl键不放，再将鼠标移出日历区，即可保持该日信息摘要不变
-Shouxing to return to the first chapter on the chapter calendar and Julian calendar date back to reckoning in Chapter III Chapter IV Chapter V of the astronomical coordinates of modern Chinese astronomy and the nature of the rules of Chapter VI of the linear sequence of discrete Chapter VII of the fitting algorithm on deltat T method of calculation shows that the use of the information summary of a certain day, if holding down the ctrl key, and then the mouse out of the district calendar, you can keep the same message digest of the day
Platform: |
Size: 723968 |
Author: 王国行 |
Hits:
Description: Linear and Switching Voltage
Regulator Fundamentals
Platform: |
Size: 141312 |
Author: ham |
Hits:
Description: 用单纯形法求解线性规划最优解，能输出求解过程中产生的单纯形表，并输出最优解-Simplex method for solving linear programming optimal solution, can solve the output generated in the process of simplex tableau, and output the optimal solution
Platform: |
Size: 1024 |
Author: 颜 |
Hits:
Description: 求单应矩阵的算法。应用LM算法。输入2幅图片上的4对对应点，应用该算法可以找到单应矩阵。-http://www.ics.forth.gr/~lourakis/homest/
C/C++ Library for Robust, Non-linear Homography Estimation
Platform: |
Size: 56320 |
Author: cao |
Hits:
Description: hurst指数
This the source code associated with the web page
http://www.bearcave.com/misl _tech/wavelets/hurst/index.html
There are the following sub-directories in this directory:
hurst
This is the "main" directory, which contains the C++ source code for
the Hurst exponent estimation functions and related tests. The
other directories all support the C++ code in this directory.
stat
Statistics functions (linear regression, autocorrelation, standard
deviation, and probability density function calculation)
data
Source code to read files downloaded from finance.yahoo.com, plus
sample market data files
freq
Wavelet spectrum calculation C++ source code
histo
Source code for calculating histograms
include
Wavelet functions-
This is the source code associated with the web page
http://www.bearcave.com/misl/misl_tech/wavelets/hurst/index.html
There are the following sub-directories in this directory:
hurst
This is the "main" directory, which contains the C++ source code for
the Hurst exponent estimation functions and related tests. The
other directories all support the C++ code in this directory.
stat
Statistics functions (linear regression, autocorrelation, standard
deviation, and probability density function calculation)
data
Source code to read files downloaded from finance.yahoo.com, plus
sample market data files
freq
Wavelet spectrum calculation C++ source code
histo
Source code for calculating histograms
include
Wavelet functions
Platform: |
Size: 490496 |
Author: hyj |
Hits:
Description: 下载说明
您下载的资源来自TI德州中文网(http://www.tichinese.com)
使用前请您先阅读以下条款：
1) 本站不保证所提供资源的完整性和安全性；
2) 请在使用前查毒 (这也是您使用其它网络资源所必须注意的) ；
3) 由本站提供的程序对您的网站或计算机造成严重后果的本站概不负责；
4) 转载本站提供的资源请勿删除本说明注释/文档；
5) 本站提供的一部分资源为网上搜集而来，如果该程序涉及或侵害到您的
版权请立即写信通知我们。
-Motion Tracking
===============
This tarball contains all code required to run the tracking algorithm
on a sequence of images.
Run the file run_tracker.m in Matlab and follow the instructions. You
will need to have a directory of sequentially numbered images
available. After entering the path and file types the tracker will
begin processing.
The code uses a solver for the linear assignment problem in lap.cpp
that must be mex-compiled before the tracker can function. My code is
set up to try to automatically compile the file src/lap/lap.cpp using
mex should it detect that it has not yet been compiled. This will
usually result in an informative message when you run the tracker for
the first time on your system, followed by Matlab exiting (this is
normal). Should automatic compilation of the file src/lap/lap.cpp fail
(i.e. upon restarting Matlab you get the same informative message as
before), then your system is probably not set up correctly and you
should invest
Platform: |
Size: 1135616 |
Author: 李岩 |
Hits:
Description: In putting together this course pack we decided not to simply include copies of the slides
for the course presentation, but to attempt to put together a small booklet of information
that could stand by itself. The course slides and other useful information, including a new
Java-based Kalman Filter Learning Tool are available at
http://www.cs.unc.edu/~tracker/ref/s2001/kalman/
In addition, we maintain a popular web site dedicated to the Kalman fi lter. This site
contains links to related work, papers, books, and even some software.
http://www.cs.unc.edu/~welch/kalman/
We expect that you (the reader) have a basic mathematical background, suffi cient to
understand explanations involving basic linear algebra, statistics, and random signals.
-In putting together this course pack we decided not to simply include copies of the slides
for the course presentation, but to attempt to put together a small booklet of information
that could stand by itself. The course slides and other useful information, including a new
Java-based Kalman Filter Learning Tool are available at
http://www.cs.unc.edu/~tracker/ref/s2001/kalman/
In addition, we maintain a popular web site dedicated to the Kalman fi lter. This site
contains links to related work, papers, books, and even some software.
http://www.cs.unc.edu/~welch/kalman/
We expect that you (the reader) have a basic mathematical background, suffi cient to
understand explanations involving basic linear algebra, statistics, and random signals.
Platform: |
Size: 537600 |
Author: ele rock |
Hits:
Description: 最常用的优化软件，能够优化很多实际问题，常包含线性及非线性-The most commonly used optimization software to optimize many practical problems, often including linear and nonlinear
Platform: |
Size: 12686336 |
Author: wenya |
Hits:
Description: This function solves linear least squares problem with nonnegative variables using the active set algorithm described in:
http://www.mathworks.com/matlabcentral/fileexchange/10908-active-set-algorithm
Platform: |
Size: 1024 |
Author: z |
Hits:
Description: Histograms of Oriented Optical Flow (HOOF) and Binet-Cauchy Kernels for Non-Linear Dynamical Systems (NLDS).
From http://www.vision.jhu.edu/code/
Platform: |
Size: 11264 |
Author: anhunt |
Hits:
Description: 实现线性预测 matlab线性预测 线性预测wav文件-Achieve linear prediction matlab linear prediction linear prediction wav file
Platform: |
Size: 1024 |
Author: 丁欣 |
Hits: