Description: The following code implements a kernel Wiener Filter algorithm in MATLAB.
The algorithm dependes on the eigenvalue decomposition, thus only a few thousand of data samples for training dataset is applicable so far.
-The following code implements a kernel Wiener Filter algorithm in MATLAB.
The algorithm dependes on the eigenvalue decomposition, thus only a few thousand of data samples for training dataset is applicable so far.
Platform: |
Size: 2614272 |
Author:eetuna |
Hits:
Description: The following code implements a kernel wiener filter algorithm in MATLAB.The algorithm dependes on the eigenvalue decomposition, thus only a few thousand of data samples for training dataset is applicable so far. Platform: |
Size: 2614272 |
Author:eetuna |
Hits:
Description: ORL人脸图像库，共40人，每人10幅图像，其中每人的前5幅作为训练样本，后5幅作为测试分类样本，统计正确分类率。分类准则为最近邻规则。
真实的图像尺寸为112x92，列向量堆积对应人脸库矩阵的每一列。 -ORL face image database, a total of 40 per 10 images, each of which the first five as training samples, after the 5 categories as a test sample, correct classification rate statistics. Classification criteria for the nearest neighbor rule. The real image size is 112x92, the corresponding column vector face database matrix accumulation of each column. Platform: |
Size: 3500032 |
Author:limei |
Hits:
Description: 基于支持向量机的人脸检测训练集增强算法实现。根据支持向量机(support vector machine，简称SVM)~ ，对基于边界的分类算"~(geometric approach)~
言，类别边界附近的样本通常比其他样本包含有更多的分类信息．基于这一基本思路，以人脸检测问题为例．探讨了
对给定训练样本集进行边界增强的问题，并为此而提出了一种基于支持向量机和改进的非线性精简集算法
IRS(improved reduced set)的训练集边界样本增强算法，用以扩大-91l练集并改善其样本分布．其中，所谓IRS算法是指
在精简集(reduced se0算法的核函数中嵌入一种新的距离度量一一图像欧式距离一一来改善其迭代近似性能，IRS
可以有效地生成新的、位于类别边界附近的虚拟样本以增强给定训练集．为了验证算法的有效性，采用增强的样本
集训练基于AdaBoost的人脸检测器，并在MIT+CMU正面人脸测试库上进行了测试．实验结果表明通过这种方法
能够有效地提高最终分类器的人脸检测性能．-According to support vector machines(SVMs)，for those geometric approach based classification
methods，examples close to the class boundary usually are more informative than others．Taking face detection as an
example，this paper addresses the problem of enhancing given training set and presents a nonlinear method to tackle
the problem effectively．Based on SVM and improved reduced set algorithm (IRS)，the method generates new
examples lying close to the face／non—face class boundary to enlarge the original dataset and hence improve its
sample distribution．The new IRS algorithm has greatly improved the approximation performance of the original
reduced set(RS)method by embedding a new distance metric called image Euclidean distance(IMED)into the
keme1 function．To verify the generalization capability of the proposed method，the enhanced dataset is used to train
an AdaBoost．based face detector and test it on the MIT+CMU frontal face test set．The experimental results show
that the origina Platform: |
Size: 649216 |
Author:郭事业 |
Hits:
Description: PVM is a fast supvervised leanring algorithm who combine dimensioin reduction and neural network training.
I have prepared the code (including six algorithms KPVM, EL M/SVD, BP/SVD and BPVM, and one dataset "Face") and put them in one zip file "pvm_code.zip", you can unzip it and run "Face_mean.m" function in Matlab environment . Before carried out experiments, please include “DimReduce” and “IncPACK” package (in the pvm_code folder)in the Matlab path setting. You will see Avarage Training time, Avarage Testing Time, Avarage Training Accuracy and Avarage Testing Accuracy of 50 trials. Because some dataset is large even they are zipped, so we just upload one dataset "Face". Other dataset can be downloaded from UCI website. According the parameter settings listed in Table 2 and Table 9, you can get the experimental results. But because the dataset is randomly split in each trial, the result may be slightly different. -I have prepared the code (including six algorithms KPVM, ELM, ELM/SVD, BP/SVD and BPVM, and one dataset "Face") and put them in one zip file "pvm_code.zip", you can unzip it and run "Face_mean.m" function in Matlab environment . Before carried out experiments, please include “DimReduce” and “IncPACK” package (in the pvm_code folder)in the Matlab path setting. You will see Avarage Training time, Avarage Testing Time, Avarage Training Accuracy and Avarage Testing Accuracy of 50 trials. Because some dataset is large even they are zipped, so we just upload one dataset "Face". Other dataset can be downloaded from UCI website. According the parameter settings listed in Table 2 and Table 9, you can get the experimental results. But because the dataset is randomly split in each trial, the result may be slightly different. Platform: |
Size: 6611968 |
Author:dengwanyu |
Hits:
Description: Implement the K nearest neighbor algorithm by your own instead of using available software.
2. Use K-fold cross validation to generate training and testing datasets. You should try different K values (3~8) to see how they affect your result.
3. Train the classifier using your training dataset, and test the classifier using your testing dataset.
4. Repeat the experiment (Step 2 and Step 3) 30 times. For each time, you need to record the training data accuracy and testing data accuracy. Finally, you can obtain the average training data accuracy and average testing data accuracy.
Platform: |
Size: 161792 |
Author:Chang |
Hits:
Description: [model] = SVMTRAIN(X, Y, C, kernelFunction, tol, max_passes) trains an
SVM classifier and returns trained model. X is the matrix of training
examples. Each row is a training example, and the jth column holds the
jth feature. Y is a column matrix containing 1 for positive examples
and 0 for negative examples. C is the standard SVM regularization
parameter. tol is a tolerance value used for determining equality of
floating point numbers. max_passes controls the number of iterations
over the dataset (without changes to alpha) before the algorithm quits.- [model] = SVMTRAIN(X, Y, C, kernelFunction, tol, max_passes) trains an
SVM classifier and returns trained model. X is the matrix of training
examples. Each row is a training example, and the jth column holds the
jth feature. Y is a column matrix containing 1 for positive examples
and 0 for negative examples. C is the standard SVM regularization
parameter. tol is a tolerance value used for determining equality of
floating point numbers. max_passes controls the number of iterations
over the dataset (without changes to alpha) before the algorithm quits. Platform: |
Size: 2048 |
Author:yaoyt |
Hits:
Description: It provides deep learning tools of deep belief networks (DBNs).-Run testDNN to try!
Each function includes description. Please check it!
It provides deep learning tools of deep belief networks (DBNs) of stacked restricted Boltzmann machines (RBMs). It includes the Bernoulli-Bernoulli RBM, the Gaussian-Bernoulli RBM, the contrastive divergence learning for unsupervised pre-training, the sparse constraint, the back projection for supervised training, and the dropout technique.
The sample codes with the MNIST dataset are included in the mnist folder. Please, see readme.txt in the mnist folder.
Platform: |
Size: 4789248 |
Author:Kevin |
Hits:
Description: ace Detection Program for MATLAB 2013a
using Gabor Feature Extraction and Neural Networks
----------------------------------------------------------------
1- copy all files and directories to the MATLAB s work folder
* In order to run the program you must have Image Processing and Neural Networks Toolboxes
2- (Important) Navigate to the root folder which contains "main.m".
3- Type "main" or "run main" in the command window
4. Only fort the first time, the program creates
Gabor filters and stores them in ./data/gabor.mat
Training set dataset and stores it in ./data/imgdb.mat
Neural Network and stores it in ./data/net.mat
5- imgdb is short for "image data base".
6- When the program menu appears click on "Train Network" and wait until the program is
done with the training-ace Detection Program for MATLAB 2013a
using Gabor Feature Extraction and Neural Networks
----------------------------------------------------------------
1- copy all files and directories to the MATLAB s work folder
* In order to run the program you must have Image Processing and Neural Networks Toolboxes
2- (Important) Navigate to the root folder which contains "main.m".
3- Type "main" or "run main" in the command window
4. Only fort the first time, the program creates
Gabor filters and stores them in ./data/gabor.mat
Training set dataset and stores it in ./data/imgdb.mat
Neural Network and stores it in ./data/net.mat
5- imgdb is short for "image data base".
6- When the program menu appears click on "Train Network" and wait until the program is
done with the training Platform: |
Size: 180224 |
Author:manu |
Hits:
Description: 文本分类：朴素贝叶斯分类器例子，采用Multi-Variate Bernoulli Event Model。一个文件为训练，一个文件为测试，采用20newsgroups数据集。-Text classification: Naive Bayes classifier example, the use of Multi-Variate Bernoulli Event Model. A file for training, a file for testing, using 20newsgroups dataset. Platform: |
Size: 6167552 |
Author:nick |
Hits:
Description: 属于机器学习的范畴，通过输入训练样本，通过分类或线性回归得到标签的假设性函数-The Gesture Recognition Toolkit (GRT) is a cross-platform, open-source, C++ machine learning library that has been specifically designed for real-time gesture recognition.
In addition to a comprehensive C++ API, the GRT now also includes an easy-to-use graphical user interface (GUI) which enables user s to stream real-time data into the GUI via the Open Sound Control network protocol. Using the GUI you can:
(1) Setup and configure a gesture recognition pipeline that can be used for classification, regression, or timeseries analysis.
(2) Stream real-time data into the GUI via Open Sound Control (OSC) another application (such as Processing, Max, Pure Data, Openframeworks, etc.).
(3) Record, label, save and load your training data.
(4) Train a model for classification or regression.
(5) Test the generalization abilities of the model (using another test dataset or cross validation).
(6) Perform real-time prediction on new data streamed into the GUI via OSC.
(7) Stre Platform: |
Size: 21445632 |
Author:王哲 |
Hits:
Description: 交叉验证(Cross-validation)主要用于建模应用中，例如PCR 、PLS 回归建模中。在给定的建模样本中，拿出大部分样本进行建模型，留小部分样本用刚建立的模型进行预报，并求这小部分样本的预报误差，记录它们的平方加和。这个过程一直进行，直到所有的样本都被预报了一次而且仅被预报一次。把每个样本的预报误差平方加和，称为PRESS(predicted Error Sum of Squares)-Cross-validation, sometimes called rotation estimation,[1][2][3] is a model validation technique for assessing how the results of a statistical analysis will generalize to an independent data set. It is mainly used in settings where the goal is prediction, and one wants to estimate how accurately a predictive model will perform in practice. In a prediction problem, a model is usually given a dataset of known data on which training is run (training dataset), and a dataset of unknown data (or first seen data) against which the model is tested (testing dataset).[4] The goal of cross validation is to define a dataset to test the model in the training phase (i.e., the validation dataset), in order to limit problems like overfitting, give an insight on how the model will generalize to an independent dataset (i.e., an unknown dataset, for instance a real problem), etc. Platform: |
Size: 4096 |
Author:liufengfeng |
Hits:
Description: In qsar studies, to make sure that the structures of the training and test sets
represent those of the whole dataset, diversity analysis can be done on
the dataset Platform: |
Size: 10240 |
Author:afy |
Hits:
Description: 简易处理空间异质性问题。需要做好分层准备。(Simple training network can deal with spatial heterogeneity easily. We need to be well prepared.) Platform: |
Size: 43008 |
Author:dandan24k |
Hits:
Description: 用决策树-回归分析模型来分析鸢尾花数据，训练后最终可以得到模型的准确率(Using decision tree-regression analysis model to analyze iris data, the accuracy of the model can be obtained after training.) Platform: |
Size: 7168 |
Author:LLLLLL。.... |
Hits:
Search - dataset training