搜索资源列表
Conjugateexamples
- 实现共轭梯度法的实例,该算法是一种优化算法,其具体优越性相信用者自知!-achieve conjugate gradient method example, the algorithm is an optimization algorithm, the specific advantages of knowing who to believe!
gongetidu
- 优化算法,共厄梯度法 fortran 90编译-optimization algorithm, a total of Ecuador gradient method FORTRAN 90 compiler
rbf.two
- 基于梯度法的RBF 网设计算法,适用于大专本科毕业设计,以及专业技术人员参考
Untitled6
- 基于梯度法的RBF 网设计算法,已验证,可用
Seven-RBF_NN--code
- 七个RBF神经网络的源代码:基于梯度法、OLS 、聚类、k均值聚类、函数逼近的RBF 网设计算法,及预测模型 -Seven RBF neural network source code: gradient-based method, OLS, clustering, k-means clustering, function approximation of the RBF network design algorithms, and predictive models
RBF
- 基于梯度法编写的RBF神经网络程序,实现对输入数据的逼近-Gradient method based on the preparation process of the RBF neural network to achieve the approximation of the input data
Steepest
- 计算梯度下降法计算极值,只能找到局部最小点。可以通过调整步长实现全局最小-Calculation of gradient descent method to calculate extreme value, can only find local minimum point. By adjusting the step size can achieve the global minimum
zuisutidu
- 最速梯度下降法-Steepest gradient descent method
BpTRAINING
- 自适应步长BP神经网络训练算法,采用最小误差和梯度下降法更新权值- BP neural network training anaysis, realized by using error feed back, gradient descent applied updating of synaptic weights
improveBPNet
- 改进的BP算法实现程序,以共轭梯度法实现BP神经网络。测试数据以txt格式给出。-Improved BP algorithm procedures in order to conjugate gradient method to achieve BP neural network. Test data given in txt format.
conjugate
- 共轭梯度法求极值相关源代码,本例为二元函数-Conjugate gradient extremum associated source code and binary function in this case
conjg
- 《神经网络与机器学习》书中的,根据共轭梯度法进行双月型数据的分类-" Neural Networks and Machine Learning" book, according to the conjugate gradient method for data classification based bimonthly
Improved-BP-neural-network-algorithm
- 改进型的BP神经网络算法,基于量化共轭梯度法搜索-Improved BP neural network algorithm
cg
- 最优化方法中的共轭梯度法,使我们老师写的matlab代码,绝对没有问题-Optimization Method of Conjugate Gradient Method
BP网络
- BP(Back Propagation)网络是1986年由Rumelhart和McCelland为首的科学家小组提出,是一种按误差逆传播算法训练的多层前馈网络,是目前应用最广泛的神经网络模型之一。BP网络能学习和存贮大量的输入输出模式映射关系,而无需事前揭示描述这种映射关系的数学方程。它的学习规则是使用最速下降法(梯度法),通过反向传播来不断调整网络的权值和阈值,使网络的误差平方和最小。BP神经网络模型拓扑结构包括输入层(input layer)、隐层(hide layer)和输出层(outpu
rbf
- 自己编写RBF神经网络程序,RBF神经网络隐层采用标准Gaussian径向基函数,输出层采用线性激活函数,其中数据中心、扩展常数和输出权值均用梯度法求解,它们的学习率均为0.001。其中隐节点数选为10,初始输出权值取[-0.1,0.1]内的随机值,初始数据中心取[-1,1]内的随机值,初始扩展常数取[0.1,0.3]内的随机值,输入采用[0 1]的随机阶跃输入(Write your own RBF neural network, RBF neural network hidden layer
Handwritten_digit_classification
- 分别使用梯度法和牛顿法训练数据,从而得到3和5两个数字的训练模型,对测试集进行判决,得到训练错误率(The training data were trained by the gradient method and Newton method, and the training models of 3 and 5 numbers were obtained. The test set was judged and the training error rate was obtained.)
RBF自适应
- 基于梯度下降法RBF自适应神经网络控制(RBF adaptive neural network control based on gradient descent method)
梯度下降法 回溯直线搜索 python代码
- 梯度下降法 回溯直线搜索 python代码 包含回溯直线搜索,以及初始值相同时不同alpha,beta值对下降速度的影响测试 用jupyter notebook打开
gradient_descent
- 梯度下降法python编程实例 附带相关数据文件在data.csv中 这个是油管up主Siraj Raval的课程代码(A demo of gradient descent algorithm. This is the code for "Intro - The Math of Intelligence" by Siraj Raval on Youtube)