Description: 在机器学习中利用欧氏距离设计一个KNN分类器，实现五折交叉验证，并用PCA进行降维-Develop a k-NN classifier with Euclidean distance and simple voting.Perform 5-fold cross validation, find out which k performs the best (in terms of accuracy)。Use PCA to reduce the dimensionality to 6, then perform 2) again. Platform: |
Size: 6144 |
Author:yuanxin |
Hits:
Description: 根据用户的电影收视率来预测用户的性别和年龄并且使用10折交叉验证。-predict the gender and age of users based on their ratings on movies.perform 10-fold cross validation. In each fold, users.datx(x=0,…,9) is used for test, and all other users. datxare used for training. You can use movies.dat and ratings.dat in both training and test and in whatever way Platform: |
Size: 1981440 |
Author:yuanxin |
Hits:
Description: 基于BP和GRNN神经网络的粮食产量预测研究，通过训练样本和测试样本的交叉验证，实现粮食产量预测效果的最佳化-Prediction of Grain Yield BP and GRNN based training through cross-validation and testing samples, to achieve the best effect of the Grain Production Forecast Platform: |
Size: 93184 |
Author:antry |
Hits:
Description: 神经网络实例文件说明：
1. chapter26_lvq.m为主程序，将该文件夹设置为MATLAB当前工作路径，运行即可。
2. crossvalidation_lvq.mat为增加了交叉验证功能(确定最佳的隐含层神经元个数)的LVQ程序。
3. chapter26_bp.m为对比的BP程序。
4. data.mat为数据文件。
5. 该程序在MATLAB2009a版本下测试通过，个别函数在低版本中不存在或者调用格式有所不同，参照对应版本中的帮助文档修改即可。
-Examples of neural network file specifications:
1. chapter26_lvq.m main program, the folder is set to MATLAB current working directory, run it.
2. crossvalidation_lvq.mat to increase cross-validation (to determine the optimal number of neurons in the hidden layer) of LVQ program.
3. chapter26_bp.m contrast to BP for the program.
4. data.mat data file.
5. The program in MATLAB2009a version tested, the individual does not exist or function call format is different, with reference to the corresponding version of the revised document to help low version. Platform: |
Size: 90112 |
Author:LIULIU |
Hits:
Description: 一个用python编写的带交叉验证的SVM分类程序，前提是必须正确安装里面所用到的Python库-Written in python with a cross-validation of SVM classification procedures, must be properly installed inside the premise is used by the Python library Platform: |
Size: 3072 |
Author:卓泽赢 |
Hits:
Description: 使用交叉验证对KNN算法和SVM算法进行参数优选，其中KNN使用matlab自带的knnclassify，svm使用LIBsvm-Using cross validation for the choosing of hyperparameter for KNN and SVM Platform: |
Size: 1024 |
Author:天钦 |
Hits:
Description: BP神经网络，采用交叉验证来测试数据的精度-The BP neural network, using cross validation to the accuracy of test data Platform: |
Size: 1024 |
Author:小丸子 |
Hits:
Description: How to install the libsvm for MATLAB on Unix machine
Linear-kernel SVM for binary classification
kernel SVM for binary classification
cross validation for C and Gamma
multi-class SVM: one-vs-rest (OVR)
More ready-to-use matlab example
Available matlab codes to download Platform: |
Size: 1572864 |
Author:hoangthom |
Hits:
Description: 本案例是采用GRNN（径向基）神经网络预测水资源量，GRNN神经网络适合于小样本、高精度预测，本实例附有数据（不全），下载者可以直接将数据更改为自己的数据便可以使用。程序中使用了交叉验证方法，使预测精度大大提高-This case is the use of GRNN (Radial Basis Function) neural network to predict the amount of water, GRNN neural networks suitable for small samples, high-precision prediction, the present example with data (incomplete), can directly download data to their own data will change can use. Procedures used in the cross-validation method, the prediction accuracy is improved greatly Platform: |
Size: 14336 |
Author:张进 |
Hits:
Description: 算法流程：选定训练集和测试集-数据预处理-交叉验证选择最佳参数-分类准确率-预测-利用最佳参数训练SVM-Algorithm flow: selected training set and test set- data preprocessing- cross-validation selection of the best parameters- classification accuracy- prediction- training SVM using the best parameters Platform: |
Size: 289792 |
Author:王华 |
Hits:
Description: Iris数据的最近邻分类与k近邻分类程序，以及5路交叉验证，适合于新手学习，附有数据集-And nearest neighbor classification k-nearest neighbor classification procedure Iris data, as well as 5-way cross-validation, suitable for novices to learn, with data collection Platform: |
Size: 5120 |
Author:taotao |
Hits:
Description: 采用SVM设计男女生分类器。采用的特征包含身高、体重、是否喜欢数学、是否喜欢文学、是否喜欢运动共五个特征。要求：采用平台提供的软件包进行分类器的设计以及测试，尝试不同的核函数设计分类器，采用交叉验证的方式实现对于性能指标的评判（包含SE,SP,ACC和AUC，AUC的计算基于平台的软件包）。-Using SVM classifier is designed for boys and girls. Characterized by the use of include height, weight, whether you like math, whether like literature, whether you like sports a total of five feature. Requirements: The software package provided by the platform classifier design and testing, try different design kernel classifiers, using cross validation manner for performance uation (including SE, SP, ACC and AUC, AUC calculated based on platform software package). Platform: |
Size: 21504 |
Author:xiapan |
Hits:
Description: 采用决策树设计男女生分类器。采用的特征包含身高、体重、是否喜欢数学、是否喜欢文学、是否喜欢运动共五个特征。要求：采用平台提供的软件包进行分类器的设计以及测试，采用交叉验证的方式实现对于性能指标的评判（包含SE,SP,ACC和AUC，AUC的计算基于平台的软件包）。-Decision tree classifier is designed for boys and girls. Characterized by the use of include height, weight, whether you like math, whether like literature, whether you like sports a total of five feature. Requirements: The software package provided by the platform classifier design and testing, the cross-validation manner for performance uation (including SE, SP, and AUC, AUC ACC computing platform based software package). Platform: |
Size: 10240 |
Author:xiapan |
Hits:
Description: 1. 使用matlab自带的人脸识别工具（Viola-Jones算法）找出人脸的位置，并裁剪出人脸区域。
2. 使用Gabor滤波器识别出人脸的局部特征及纹理。
3. 训练一个SVM进行表情分类。
4. 交叉验证得到表情分类正确率为83.3 。
操作说明和系统描述请见ReadMe.-1. Using matlab with face detection tool (Viola-Jones algorithm) to find the location of a human face, and cut out the face region.
2. Use Gabor filter to identify the face local feature and texture.
3. Adopt SVM to train a facial expression classifier.
4. Use cross-validation to exam the classification accuracy: 83.3 . Platform: |
Size: 5593088 |
Author:冯溢 |
Hits:
Description: svm 的参数优化，利用交叉验证法选择最优参数c g，最终提高训练集的分类准确率,更好的提高分类器性能-Svm parameter optimization, the use of cross-validation method to the optimal parameter c g, and ultimately improve the training set classification accuracy，better improve the classifier performance Platform: |
Size: 1024 |
Author:赵珂 |
Hits:
Description: cheshi.m为使用交叉验证的GRNN神经网络预测程序
xiaoguobijiao.m为BP和GRNN效果比较程序(Cheshi.m is a cross validation GRNN neural network prediction program
Xiaoguobijiao.m is the BP and GRNN effect comparison program) Platform: |
Size: 7168 |
Author:海阔天空1998
|
Hits:
Search - cross validation