▍1. Autoencoder_Code
神经网络中深度学习理论的求解算法,发掘数据中的结构信息(deep learning to discover the structure hidden in the data)
神经网络中深度学习理论的求解算法,发掘数据中的结构信息(deep learning to discover the structure hidden in the data)
说明: 反时限曲线整定-用MATLAB画出电流与线路长度的关系(Inverse time curve setting - relationship between current and line length)
基于罚函数的粒子群算法的函数寻优,求解结果收敛较快且不易陷入局部最优(The function optimization of the particle swarm algorithm based on the penalty function is faster and difficult to get into the local optimal)
PSO+BP预测风速首先对风速数据进行预处理,接着进行PSOBP仿真预测。。。(PSO+BP forecast wind velocity data preprocessing first, followed by PSOBP simulation. . .)
说明: 目标跟踪算法HCF的改进版,将HCF中的VGG-19特征换成DenseNet特征(The improved version of target tracking algorithm HCF, the vgg-19 feature in HCF is replaced by densenet feature)
说明: 用matlab实现kmeans算法,成功运行,亲测好用(Use kmlab to implement kmeans algorithm, run successfully, pro-test easy to use)
利用大量图像数据对卷积神经网络算法进行训练,通过卷积、池化、下采样以及全连接层训练后的卷积神经网络在图像识别精度越来越高。(By using a large number of image data to train the convolutional neural network algorithm, the accuracy of the image recognition is higher and higher by convolution, pooling, down sampling and full connection layer training.)
卷积神经网络是近年来广泛应用于模式识别、图像处理等领域的一种高效识别算法,它具有结构简单、训练参数少和适应性强等特点。(Convolutional Neural Network Library)
matlab 实现鸢尾花数据集的KNN算法(Matlab implementation of KNN algorithm)
模糊C均值脑电分类并使用了支持向量机对比,其中支持向量机使用了三种方法参数寻优。(fuzzy C means clustering for EEG classification,and use the SVM for campare. The SVM applied three methods to find the optimal apartments.)
利用遗传算法对支持向量机(SVM)进行分类优化(optimizate the support vector machine (SVM) classification by using the genetic algorithm)
利用matlab实现深度学习的bp算法,旨在实现深度神经网络的架构(Using MATLAB to realize the bp algorithm of deep learning, aiming at realizing the architecture of deep neural network)
说明: 利用matlab实现深度学习的bp算法,旨在实现深度神经网络的架构(Using MATLAB to realize the bp algorithm of deep learning, aiming at realizing the architecture of deep neural network)
是MATLAB的BP预测程序,有实际参考价值,需要的就下吧(MATLAB BP is forecasting process, the practical reference value, we need it on the next)
说明: matbal编的bp算法,可以用来预测,有详细算法(matbal series bp algorithm can be used to predict a detailed algorithm)
说明: 文件内容包括buck电路的开环传递函数、加PI后的闭环传递推导;m文件;仿真mdl文件。实现功能——通过闭环BODE图整定出PID参数。(The contents of the file include the open-loop transfer function of the buck circuit, the closed-loop transfer derivation after adding PI, the m file, and the simulation MDL file. Realization function: PID parameters are adjusted by closed-loop Bode diagram.)
生成四种典型波形,利用神经网络完成分类,可直接使用(Four typical waveform were generated and classification by the neural network is completed, can be used directly)
说明: 通过人工神经网络与自回归滑动平均模型的结合来进行预测,将历史时间序列作为输入,下一时刻的预测值作为输出(By combining the artificial neural network with the autoregressive moving average model, the historical time series is taken as the input and the prediction value of the next time is taken as the output)
通过DE的方法对BP神经网络进行优化,用PM2.5数据进行验证。(On the BP neural network is optimized by using the method of DE, with PM2.5 data for validation.)