％ EM algorithm for k multidimensional Gaussian mixture estimation
％
％ Inputs:
％ X(n,d) - input data, n=number of observations, d=dimension of variable
％ k - maximum number of Gaussian components allowed
％ ltol - percentage of the log likelihood difference between 2 iterations ([] for none)
％ maxiter - maximum number of iteration allowed ([] for none)
％ pflag - 1 for plotting GM for 1D or 2D cases only, 0 otherwise ([] for none)
％ Init - structure of initial W, M, V: Init.W, Init.M, Init.V ([] for none)
％
％ Ouputs:
％ W(1,k) - estimated weights of GM
％ M(d,k) - estimated mean vectors of GM
％ V(d,d,k) - estimated covariance matrices of GM
％ L - log likelihood of estimates
％
this directory
contains the following:
* The acdc algorithm for finding the
approximate general (non-orthogonal)
joint diagonalizer (in the direct Least Squares sense) of a set of Hermitian matrices.
[acdc.m]
* The acdc algorithm for finding the
same for a set of Symmetric matrices.
[acdc_sym.m](note that for real-valued matrices the Hermitian and Symmetric cases are similar however, in such cases the Hermitian version
[acdc.m], rather than the Symmetric version[acdc_sym] is preferable.
* A function that finds an initial guess
for acdc by applying hard-whitening
followed by Cardoso s orthogonal joint
diagonalizer. Note that acdc may also
be called without an initial guess,
in which case the initial guess is set by default to the identity matrix.
The m-file includes the joint_diag
function (by Cardoso) for performing
the orthogonal part.
[init4acdc.m]
Because of the poor observability of Inertial Navigation System on stationary base, the estimation
error of the azimuth will converge very slowly in initial alignment by means of Kalmari filtering, and making the
time initial alignment is longer. In this paper, a fast estimation method of the azimuth error is creatively proposed
for the initial alignment of INS on stationary base. On the basis of the the fast convergence of the leveling error,
the azimuth error can be directly calculated. By means of this fast initial alignment method, the time of initial
alignment is reduced greatly. The computer simulation results illustrate the efficiency of the method.
The "1818" is my first search word in this website, So I upload Lcd drive(Ssd1818a) C language code as my support.
96*64 dot matrix Graphic
LCD controller: SSD1818A
Parallel interface
initial routine code
Command write and Data write
Base functions and application code
Ascii font library
Product Photo
Mcu: Pic16F887...etc.
Keywords: Ssd1818a, dot matrix graphic, Parallel interface, initial routine, base font8x8 librarys
December 19, 2006 - Ant 1.7.0 Available
Apache Ant 1.7.0 is now available for download.
Ant 1.7 introduces a resource framework. Some of the core ant tasks such as <copy/> are now able to process not only file system resources but also zip entries, tar entries, paths, ... Resource collections group resources, and can be further combined with operators such as union and intersection. This can be extended by custom resources and custom tasks using resources.
Ant 1.7 starts outsourcing of optional tasks to Antlibs. The .NET antlib in preparation will replace the .NET optional tasks which ship in Ant. Support for the version control system Subversion will be only provided as an antlib to be released shortly.
Ant 1.7 fixes also a large number of bugs.
Ant 1.7 has some initial support for Java6 features.
function [U,V,num_it]=fcm(U0,X)
％ MATLAB (Version 4.1) Source Code (Routine fcm was written by Richard J.
％ Hathaway on June 21, 1994.) The fuzzification constant
％ m = 2, and the stopping criterion for successive partitions is epsilon =??????.
％*******Modified 9/15/04 to have epsilon = 0.00001 and fix univariate bug********
％ Purpose:The function fcm attempts to find a useful clustering of the
％ objects represented by the object data in X using the initial partition in U0.
Batch version of the back-propagation algorithm.
％ Given a set of corresponding input-output pairs and an initial network
％ [W1,W2,critvec,iter]=batbp(NetDef,W1,W2,PHI,Y,trparms) trains the
％ network with backpropagation.
％
％ The activation functions must be either linear or tanh. The network
％ architecture is defined by the matrix NetDef consisting of two
％ rows. The first row specifies the hidden layer while the second
％ specifies the output layer.
％
