The code performs a number (ITERS) of iterations of the
Bailey s 6-step FFT algorithm (following the ideas in the
CMU Task parallel suite).
1.- Generates an input signal vector (dgen) with size
n=n1xn2 stored in row major order
In this code the size of the input signal
is NN=NxN (n=NN, n1=n2=N)
2.- Transpose (tpose) A to have it stored in column
major order
3.- Perform independent FFTs on the rows (cffts)
4.- Scale each element of the resulting array by a
factor of w[n]**(p*q)
5.- Transpose (tpose) to prepair it for the next step
6.- Perform independent FFTs on the rows (cffts)
7.- Transpose the resulting matrix
The code requires nested Parallelism.
For the incomplete methods, we kept the representation of the queens by a table and the method of calculation to determine if two queens are in conflict, which is much faster for this kind of problems than the representation by a matrix.
heuristics: descent.
Tests: 100 queens in less than 1 second and 67 iterations. 500 queens in 1 second and 257 iterations. 1000 queens in 11 seconds and 492 iterations.
heuristics: Simulated annealing.
Tests: 100 queens in less than 1 second and 47 iterations. 500 queens in 5 seconds and 243 iterations. 1000 queens in 13 seconds and 497 iterations.
heuristics: based on Simulated Annealing.
Tests: 100 queens in less than 1 second and 60 iterations. 500 queens in 1 second and 224 iterations. 1000 queens in 5 seconds and 459 iterations. 10 000 queens in 20 minutes 30 seconds and 4885 iterations.
This forced me to write about more interesting and comprehensive sorting methods, the result of which is this one. Through this writing I have tried to give in-depth coverage of the entire sort algorithm I hope Peter wouldn t mind reading it. This article assumes that you really don t know about the iterations, looping, and so forth hence, it explains these in detail first.
Using Jacobi method and Gauss-Seidel iterative methods to solve the following system
The required precision is =0.00001, and the maximum iteration number N=25. Compare the number of iterations and the convergence of these two methods
ET++ is a portable and homogenous object-oriented class library integrating user
interface building blocks, basic data structures, and high level application framework
components. ET++ eases the building of highly interactive applications with consistent
user interfaces following the direct manipulation principle. The ET++ class library is
implemented in C++ and can be used on several operating systems and window system
platforms. Since its initial conception the class library has been continuously
redesigned and improved. It started with an architecture which was close to MacApp.
During several iterations a new and unique architecture evolved. A byproduct of the
ET++ project is a set of tools, which were designed to support the exploration of ET++
applications at run-time.
設計模式一書引用的主要參考例程，一個跨平臺的應用框架，基于C++實現，是學習面向對象的經典源碼.
Sequential Monte Carlo without Likelihoods
粒子濾波不用似然函數的情況下
本文摘要:Recent new methods in Bayesian simulation have provided ways of evaluating posterior distributions
in the presence of analytically or computationally intractable likelihood functions.
Despite representing a substantial methodological advance, existing methods based on rejection
sampling or Markov chain Monte Carlo can be highly inefficient, and accordingly
require far more iterations than may be practical to implement. Here we propose a sequential
Monte Carlo sampler that convincingly overcomes these inefficiencies. We demonstrate
its implementation through an epidemiological study of the transmission rate of tuberculosis.
％ 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
％
observable distribution grid are investigated. A distribution
grid is observable if the state of the grid can be fully determined.
For the simulations, the modified 34-bus IEEE test feeder is used.
The measurements needed for the state estimation are generated
by the ladder iterative technique. Two methods for the state
estimation are analyzed: Weighted Least Squares and Extended
Kalman Filter. Both estimators try to find the most probable
state based on the available measurements. The result is that
the Kalman filter mostly needs less iterations and calculation
time. The disadvantage of the Kalman filter is that it needs some
foreknowlegde about the state.
