资源列表
kernel_matrix
- 超限学习机应用部分的内容,用于生成核函数的矩阵变量欧米伽。-Machine learning content overrun part of the matrix variable is used to generate kernels Omega.
cnn
- 卷积神经网络(cnn)的练习程序,包含多个exercise-convolutional nerual network
ANN
- 先用BP神经网络算法对国民经济生产总值做预测,然后用灰色预测模型GM(1,1)做预测,最后将两者结合建立组合预测模型。-Firstly,the BP artificial neural network algorithm to predict the GDP and then construct the GM(1,1) model,at last build out the combined model to predict the GDP int the future.
PsoANN
- 微粒群算法结合神经网络进行建筑能耗预测,有效的提高了预测精度,但是延长了运算时间-building energy consumption prediction using particle swarm optimization and artificial neural networks
simmg1
- 基于Matlab平台的排队论中M/G/1模型的实现 源代码-Matlab platform based on queuing theory in M/G/1 model realization source code
混沌时间序列关联积分求解
- 生成Lorenz时间序列,运用GP算法计算关联维数
Non_manuver_4targets_PHD
- 本文是实现PHD滤波的程序代码,内带有该文章。用于仿真多目标杂波环境下对目标的跟踪与估计-This is the realization of PHD filtering program code with the article within. Clutter Simulation multi-target environment for target tracking and estimation for
15bus-ELD
- 电力系统15机经济负荷分配;降维;遗传算法-power system 15 bus economic load dispatch genetic algorithm
clustering-master
- 稀疏子空间聚类的算法程序实现,用Matlab实现的SSC算法。-Sparse subspace clustering algorithm procedures to achieve, SSC algorithm implemented with Matlab.
DP
- 动态规划方法解决单个水库运行调度问题,求解最优控制目标。-Dynamic programming method to solve a single reservoir operation scheduling problem, solving the optimal control target.
self-taught-learning
- 自主学习把稀疏自编码器和分类器实现结合。先通过稀疏自编码对无标签的5-9的手写体进行训练得到最优参数,然后通过前向传播,得到训练集和测试集的特征,通过0-4有标签训练集训练出softmax模型,然后输入测试集到分类模型实现分类。-Independent Learning the encoder and the sparse classifiers achieve the combination. First through sparse coding since no label was ha
Collaborative-fuzzy-clustering
- 协同模糊聚类算法,这是聚类算法的一种。国内研究的人不多。-Collaborative fuzzy clustering,