资源列表
Perceptron-numerical-experiment
- 1.设计一个单一感知器解决简单分类问题 2.设计多个感知器神经元解决分类问题，并绘出图像 3.利用标准化感知器学习规则消除奇异样本对训练的影响-Design perceptron to solve classification problems
LSPE_code
- 最小二乘参数估计的matlab程序代码，包括梯度校正参数估计法的matlab实现-Least squares parameter estimation matlab code, including gradient correction parameter estimation matlab implementation
EM_code
- 机器学习中EM算法的R语言实现，实测可用-EM Algorithm R language, found Available
ADP-code
- 自适应动态规划（ADP）actor-critic框架用于最优控制的matlab仿真程序代码，实测可用-Adaptive dynamic programming (ADP) actor-critic matlab simulation framework for optimal control of the program code, found available
BP-Adaboost
- BP-Adaboost模型即把BP神经网络作为弱分类器，反复训练BP神经网络预测样本输出，通过Adaboost算法得到多个BP神经网络弱分类器组成的强分类器。-The BP-Adaboost model uses BP neural network as weak classifier to repeatedly train BP neural network to predict the sample output. Adaboost algorithm is used to obtain a
MNIST(tensorflow)
- 基于tensorflow的手写识别，训练后可以识别手写数字-Based on tensorflow handwriting recognition, training can identify handwritten numbers
neural-network-and-genetic-algorithm
- 对于未知的非线性函数，仅通过函数的输入输出数据难以准确寻找函数极值，这类问题可以通过神经网络结合遗传算法求解，利用神经网络的非线性拟合能力和遗传算法的非线性寻优能力寻找函数极值。-For unknown nonlinear function, only through the function of input and output data is difficult to accurately find the function extreme value, this kind of prob
genetic-algorithm
- 根据遗传算法和BP神经网络理论，在matlab软件中编程实现基于遗传算法优化的BP神经网络非线性系统拟合算法。-According to the theory of genetic algorithm and BP neural network, in the matlab software programming to realize BP neural network nonlinear systems based on genetic algorithm optimization fitt
nonlinear-system-modeling
- 基于BP神经网络的非线性系统建模-非线性函数拟合，本案例拟合的非线性函数为y x1的平方+x2的平方。-The nonlinear system modeling based on BP neural network to nonlinear function fitting
breed-PSO
- 杂交粒子群算法是借鉴遗传算法中杂交的概念，在每次迭代中，根据杂交率选取指定数量的粒子放入杂交池内，池内的粒子随机地两两杂交。-The hybrid particle swarm algorithm is based on the concept of hybridization in genetic algorithm. In each iteration, the specified number of particles are put into the hybridization pool
linear-PSO
- 针对PSO算法容易早熟及后期容易在全局最优解附近震荡现象，可使用线性递减权重法PSO-For the PSO algorithm is easy to premature and late in the global optimal solution near the oscillation phenomenon, you can use the linear decrement weight PSO
darknet
- 神经网络引入后，检测框架变得更快更准确。然而，大多数检测方法受限于少量物体。检测和训练数据上联合训练物体检测器，用有标签的检测图像来学习精确定位，同时用分类图像来增加词汇和鲁棒性。原YOLO系统上生成YOLOv2检测器；在ImageNet中超过9000类的数据和COCO的检测数据上，合并数据集和联合训练YOLO9-After the neural network is introduced, it is becoming faster and more accurate detection fr