搜索资源列表
BP
- BP神经网络程序,包含一个C语言的原始程序,看到BP的算法流程,还有两个改进的BP程序,包括自适应动量下降法和梯度下降法-BP neural network program, the original program that contains a C language, see BP algorithm process, there are two improved BP procedures, including adaptive momentum descent and gradient
linear-regressionpy
- python实现简单的单变量线性回归,使用MSE以及梯度下降两种方法-python simple univariate linear regression, using two methods MSE and gradient descent
linearregmatlab
- matlab实现简单的单变量线性回归,并详细实现了梯度下降过程-matlab simple univariate linear regression, and the detailed implementation of the gradient descent
steepest-descent--and-Newton
- function[x,val,k] dampnm(fun,gfun,hesse,x0) 功能:用阻尼牛顿法求解无约束问题:min f(x) 输入:x0是初始点,fun,gfun,hesse分别是目标函数值,梯度,hesse矩阵的函数 输出:x,val分别是近似最优点和最优值,k是迭代次数-The function/x, val, k dampnm (fun, gfun, Hesse, x0) function: with damped Newton method f
fullersteep
- Application of Gradient Descent methodology to Fuller s Problem of Optimization
cnn
- 卷积神经网络的实现代码,包括训练,测试,SGD随机梯度下降法等功能。-Convolution neural network implementation code, including training, testing, SGD stochastic gradient descent and other functions.
Attitude-Algorithm
- 两种基于互补滤波的姿态算法,对陀螺仪和加速度计进行数据融合,解算姿态。一种使用了梯度下降算法,另一种使用反馈修正和李雅普诺夫分析。-Two algorithms for attitude estimate based on complementary filtering. Data form Gyroscope and Accelerometer are fused. One is based on gradient descent algorithm while the other is ba
tiduxiajiang
- 梯度下降法的图形表示,里面有详细的中文注释,运行即可看到结果。-Gradient descent graphical representation, which the Chinese have detailed notes, you can see the results running.
MIT_lec5_steep_desce
- Lecture about Gradient Descent Algorithm
bpback
- 神经网络比较基础的算法,实现梯度下降和反向传播,以及L2规范化、交叉熵代价函数的引入,卷积神经网络 该算法用于mnist数据测试,有详细中文注释-Neural network based on the comparison algorithm, gradient descent and back-propagation, and L2 standardization introduced cross entropy cost function, convolution neural netw
windows
- 最速梯度下降和广义逆最小二乘求解感知机,可以调的参数有迭代下降的因子还有广义逆求解时的向量b。-Steepest gradient descent and generalized least squares for solving inverse Perceptron, you can adjust the parameters as well as iterative decrease factor when solving generalized inverse vector b.
ex4
- 手写识别的实现,由bp算法和梯度下降算法实现的神经网络-Handwriting recognition achieved by the bp algorithm and gradient descent algorithm neural network
GUI_TDtest
- 梯度下降法进行系统辨识及GUI操作实现,注:主要使用第二类梯度下降法实现,谢谢大家!-Gradient descent method for system identification Matlab procedures and GUI interface operation
Linear-Regression
- 使用梯度下降法和解析法实现的线性回归算法。-Gradient descent algorithm using the linear regression method and analytical method to achieve.
Logistic-Regression
- 使用梯度下降法、随机梯度下降法和牛顿法实现的逻辑回归算法。-Using a gradient descent method, stochastic gradient descent algorithm logistic regression method and Newton' s method to achieve.
gradientDescent
- 机器学习中用于求解代价函数最小值的梯度下降算法-Machine Learning for solving the minimum cost function gradient descent algorithm
OVM
- A Dynamical Tikhonov Regularization for Solving Ill-posed Linear Algebraic Systems.pdf,以上论文提出的求解病态线性方程组的一种较新梯度下降法-A Dynamical Tikhonov Regularization for Solving Ill-posed Linear Algebraic Systems.pdf, more than one paper proposes solving ill-conditi
rgd
- Rosen梯度下降法的matlab实现,线搜索采用的是0.618法,其中的目标函数可以自己修改。-Rosen gradient descent method of Matlab implementation, line search using the 0.618 method, which can modify the objective function.
grad_descent_
- gradient descent algorithm
Conjugate-Gradient-Method
- 共轭梯度法(Conjugate Gradient)是介于最速下降法与牛顿法之间的一个方法,它仅需利用一阶导数信息,但克服了最速下降法收敛慢的缺点,又避免了牛顿法需要存储和计算Hesse矩阵并求逆的缺点,共轭梯度法不仅是解决大型线性方程组最有用的方法之一,也是解大型非线性最优化最有效的算法之一。 在各种优化算法中,共轭梯度法是非常重要的一种。其优点是所需存储量小,具有步收敛性,稳定性高,而且不需要任何外来参数。-Conjugate gradient method (Conjugate Gradie