Location:
Search - Marquardt
Search list
Description: Levenberg-Marquardt优化单应性矩阵，也可经过修改用于相机标定参数的优化
Platform: |
Size: 1824 |
Author: 卫东 |
Hits:
Description: 本程序是BP算法的演示程序, 其中的Levenberg-Marquardt算法具有实用价值。随代码还提供帮助文件，非常方便。-BP algorithm Demonstration Program, which Levenberg-Marquardt algorithm has practical value. With the code also provides help files, a very convenient.
Platform: |
Size: 332800 |
Author: 林瑞伦 |
Hits:
Description: L-M算法。除了动量法(基于梯度下降的训练算法)外，学习率自适应调整策略是BP算法改进的另一种途径，它利用Levenberg-Marquardt优化方法，从而使得学习时间更短。其缺点是，对于复杂的问题，该方法需要很大的存储空间。 -L-M algorithm. In addition to momentum (based on the gradient descent algorithm for training), learning rate adaptive strategy is to improve the algorithm BP Another approach, which uses Levenberg-Marquardt optimization method, which makes learning time is even shorter. Its weakness is that the complex problems that the method requires a lot of storage space.
Platform: |
Size: 1024 |
Author: 辜小花 |
Hits:
Description: matlab使用BP网络的例子,使用Levenberg Marquardt算法提高训练速度,效果不错-use Matlab BP example, the use of Levenberg Marquardt algorithm to improve training speed, good results
Platform: |
Size: 4096 |
Author: 汤泽世 |
Hits:
Description: 使用Marquardt方法的非线性嘴小二乘拟合算法-use of small mouth nonlinear least square fitting algorithm
Platform: |
Size: 1024 |
Author: 游江 |
Hits:
Description: 神经网络进行手写数字识别：
本程序是BP算法的演示程序, 其中的Levenberg-Marquardt算法具有实用价值.
带有图形界面-neural network handwritten numeral recognition : this program is the BP algorithm Demonstration Program, The Levenberg-Marquardt algorithm is practical value. with a graphical interface
Platform: |
Size: 133120 |
Author: HK |
Hits:
Description: bp神经网络算法是解决最优化问题的先进算法之一，本论文讨论了神经网络中使用最为广泛的前馈神经网络。其网络权值学习算法中影响最大的就是误差反向传播算法（back-propagation简称BP算法）。BP算法存在局部极小点，收敛速度慢等缺点。基于优化理论的Levenberg-Marquardt算法忽略了二阶项。该文讨论当误差不为零或者不为线性函数即二阶项S(W)不能忽略时的Hesse矩阵的近似计算，进而训练网络。-bp neural network algorithm to solve optimization problems, one of the advanced algorithm, the paper discusses the neural network in the most widely used feed-forward neural network. Its network weights learning algorithm in the greatest impact is the error back-propagation algorithm (back-propagation algorithm referred to as BP). BP algorithm for the existence of local minimum points, such as the shortcomings of slow convergence. Optimization theory based on the Levenberg-Marquardt algorithm ignores the second-order item. In this paper, the discussion when the error is not zero or not that is second-order linear function of S (W) can not be ignored when the Hesse matrix of approximate calculation, and then training the network.
Platform: |
Size: 19456 |
Author: 刘慧 |
Hits:
Description: Levenberg-Marquardt优化单应性矩阵，也可经过修改用于相机标定参数的优化-Levenberg-Marquardt optimization homography can also be modified for optimization of camera calibration parameters
Platform: |
Size: 2048 |
Author: 卫东 |
Hits:
Description: Levenberg-Marquardt的c++优化代码。该代码适用各种非线性最小二乘问题-Levenberg-Marquardt
Platform: |
Size: 57344 |
Author: 卫东 |
Hits:
Description: Java实现的各种数学算法，如矩阵运算，
levenberg-marquardt优化算法等。-Java realize a variety of mathematical algorithms, such as matrix operations, levenberg-marquardt optimization algorithm.
Platform: |
Size: 51200 |
Author: 张如飞 |
Hits:
Description: 本程序是BP算法的演示程序, 其中的Levenberg-Marquardt算法具有实用价值.
一、网络训练
程序默认状态是样本训练状态,现将样本训练状态下的如何训练网络进行说明:
1.系统精度:
定义系统目标精度,根据需要定义网络训练误差精度.误差公式是对训练出网络的输出层节点和实际的网络输出结果求平方差的和.
最大训练次数:
默认为10000次,根据需要调整,如果到达最大训练次数网络还未能达到目标精度,程序退出.
3.步长:
默认为0.01,由于采用变步长算法,一般不需人工设置.
4.输入层数目:
人工神经网络的输入层神经元的节点数目.
5.隐含层数目:
人工神经网络的隐含层神经元的节点数目.
6.输出层数目:
人工神经网络的输出层神经元的节点数目.
7.训练算法:
强烈建议选取Levenberg-Marquardt算法,该算法经过测试比较稳定.
8.激活函数:
不同的网络激活函数表现的性能不同,可根据实际情况选择.
9.样本数据的处理:
由于程序没有实现归一化功能, 因此用来训练的样本数据首先要归一化后才能进行训练.
Platform: |
Size: 344064 |
Author: starboy_2nd |
Hits:
Description: 这是一个马奎特最优化算法，通过测试，比一般的最优化算法运算速度要快很多，而且精度也挺高。希望对大家有用-This is an optimization algorithm Marquette, passed the test than the general optimization algorithm to much faster computation speed and accuracy挺高also. Hope useful for everyone
Platform: |
Size: 184320 |
Author: 马见青 |
Hits:
Description: 最新的Levenberg Marquardt 算法。用于非线性最小二乘问题的参数估计和优化！-Latest Levenberg Marquardt algorithm. Nonlinear least squares problem for parameter estimation and optimization!
Platform: |
Size: 67584 |
Author: zhangxsh |
Hits:
Description:
％ Train a two layer neural network with the Levenberg-Marquardt
％ method.
％
％ If desired, it is possible to use regularization by
％ weight decay. Also pruned (ie. not fully connected) networks can
％ be trained.
％
％ Given a set of corresponding input-output pairs and an initial
％ network,
％ [W1,W2,critvec,iteration,lambda]=marq(NetDef,W1,W2,PHI,Y,trparms)
％ trains the network with the Levenberg-Marquardt method.
％
％ The activation functions can be either linear or tanh. The
％ network architecture is defined by the matrix NetDef which
％ has two rows. The first row specifies the hidden layer and the
％ second row specifies the output layer.- Train a two layer neural network with the Levenberg-Marquardt method. If desired, it is possible to use regularization by weight decay. Also pruned (ie. not fully connected) networks can be trained. Given a set of corresponding input-output pairs and an initial network, [W1, W2, critvec, iteration, lambda] = marq (NetDef, W1, W2, PHI, Y, trparms) trains the network with the Levenberg-Marquardt method . The activation functions can be either linear or tanh. The network architecture is defined by the matrix NetDef which has two rows. The first row specifies the hidden layer and the second row specifies the output layer.
Platform: |
Size: 3072 |
Author: 张镇 |
Hits:
Description: Levenberg-Marquardt最小二乘拟合-Levenberg-Marquardt LeastSquares Fitting
Platform: |
Size: 23552 |
Author: mywk |
Hits:
Description: a copylefted C/C++ implementation of the Levenberg-Marquardt non-linear
least squares algorithm
Platform: |
Size: 88064 |
Author: newbin |
Hits:
Description: 用gsl实现Levenberg-Marquardt算法-use gsl to implement Levenberg-Marquardt algo
Platform: |
Size: 3178496 |
Author: cpper |
Hits:
Description: Levenberg-Marquardt 信赖域方法求解非线性方程组的Matlab程序-Levenberg-Marquardt trust region method for solving nonlinear equations of the Matlab program
Platform: |
Size: 1024 |
Author: 蔚无 |
Hits:
Description: 非线性最小二乘法，Levenberg-Marquardt算法，很详细的理论推导，讲解。对研究最小二乘法和Levenberg-Marquardt算法的同学有很大的帮助！-Nonlinear least squares method, Levenberg-Marquardt algorithm, a very detailed theoretical analysis to explain. The study of least squares and Levenberg-Marquardt algorithm is very helpful to the students!
Platform: |
Size: 396288 |
Author: 陈世界 |
Hits:
Description: 这里有个Levenberg-Marquardt算法的程序和ppt介绍-Here' s a Levenberg-Marquardt algorithm described procedures and ppt
Platform: |
Size: 2695168 |
Author: 王毅 |
Hits: