Lorenz-chaos

TP1_loi_de_bernouli avec matlab

虚拟机放置matlab代码 随机搜索算法 效率较高 速度较快 求解结果准确 (Virtual machine placement Matlab code random search algorithm is more efficient, faster solution results accurate)

Estimates the power spectral densities (PSDs) using both the ICOMP method and Mallows CL for deciding the diagonal loading.

estimate the test accuracy,training accuray and validation accuracy of a neural network with 10-fold cross validation.

Adding 2 types of noise to an image and then removing the noise and evaluating the removal process

深度神经网络在函数近似中表现优越，然而需要从头开始训练。另一方面，贝叶斯方法，像高斯过程（GPs），可以利用利用先验知识在测试阶段进行快速推理。然而，高斯过程的计算量很大，也很难设计出合适的先验。本篇论文中我们提出了一种神经模型，条件神经过程（CNPs），可以结合这两者的优点。CNPs受灵活的随机过程的启发，比如GPs，但是结构是神经网络，并且通过梯度下降训练。CNPs通过很少的数据训练后就可以进行准确的预测，然后扩展到复杂函数和大数据集。我们证明了这个方法在一些典型的机器学习任务上面的的表现和功能，比如回归，分类和图像补全(Deep neural networks perform well in function approximation, but they need to be trained from scratch. On the other hand, Bayesian methods, such as Gauss Process (GPs), can make use of prior knowledge to conduct rapid reasoning in the testing stage. However, the calculation of Gauss process is very heavy, and it is difficult to design a suitable priori. In this paper, we propose a neural model, conditional neural processes (CNPs), which can combine the advantages of both. CNPs are inspired by flexible stochastic processes, such as GPs, but are structured as neural networks and trained by gradient descent. CNPs can predict accurately with very little data training, and then extend to complex functions and large data sets. We demonstrate the performance and functions of this method on some typical machine learning tasks, such as regression, classification and image completion.)

说明：  削减平均最值恒虚警检测器（TMGO and TMSO CFAR）(detectors based on TMGO and TMSO CFAR techniques)

分岔图程序，有关位移方面的，很实用，可下载参考。(The bifurcation diagram of procedure, the displacement of aspects, very practical, you can download the reference.)

阵列信号处理方面10个经典程序，包括波束形成，capon，MVCR等等，以及仿真图(10 classic program in array signal processing )

说明：  遗传算法车间调度,车间作业调度问题遗传算法 -------------------------------------------------------------------------- 输入参数列表 M 遗传进化迭代次数 N 种群规模(取偶数) Pm 变异概率 T m×n的矩阵，存储m个工件n个工序的加工时间 P 1×n的向量，n个工序中，每一个工序所具有的机床数目 输出参数列表 Zp 最优的Makespan值 Y1p 最优方案中，各工件各工序的开始时刻，可根据它绘出甘特图 Y2p 最优方案中，各工件各工序的结束时刻，可根据它绘出甘特图 Y3p 最优方案中，各工件各工序使用的机器编号 Xp 最优决策变量的值，决策变量是一个实数编码的m×n矩阵 LC1 收敛曲线1，各代最优个体适应值的记录 LC2 收敛曲线2，各代群体平均适应值的记录 最后，程序还将绘出三副图片：两条收敛曲线图和甘特图（各工件的调度时序图）(Genetic algorithm scheduling, job shop scheduling problems with genetic algorithms -------------------------------------------------------------------------- genetic evolution of the input parameter list, the number of iterations M N population size (taken even) Pm mutation probability T m × n matrix, stored m one piece n a process of processing time 1 × n vector, n a process in which each machine processes the number of Zp with the best value Y1p Optimal Makespan programs, the start time of each process the workpiece can be drawn based on its optimal solution Gantt Y2p, each time the workpiece end of the process, according to its draw Gantt Y3p optimal solution, each piece of the processes using machine code Xp optimal decision variable, decision variable is a real m × n matrix encoded LC1 convergence curve 1, the generation of the best individual record of LC2 fitness convergence curve 2, the average fitness value on behalf of groups record Finally, the program will draw three pict)