搜索资源列表
DecisionTree
- 该资源是决策树算法。从“样本特征说明”读取属性,从“样本数据”读取各属性值。 依次取20 —80 (每次递增10 )的样本作为训练集,剩余样本作为测试集测试。-This resource is about decision tree algorithm. The attributes and values can be read by the code from the two txt files. Each time choose a portion of all the data (20
KNN_2011211651
- 应用KNN算法解决0到9的手写数字识别问题,效果在90 以上,内部有两个数据集,一个为训练集(7291个数据),一个为测试数据(2791个数据),程序采用MATLAB编写,另附有文档,程序简洁易懂-Application KNN algorithm to solve 0-9 handwritten digit recognition problem, the effect is more than 90 , the inside has two data sets, one for the tr
bp_demo
- BP神经网络软件(BPGUI):支持手动输入数据和从外部导入数据两种方式;用户可以自定义训练集和测试集占全部数据的百分比,设定完成后,软件随机产生训练集和测试集;支持归一化和不归一化两种数据预处理方式;用户可以自定义网络的结构参数和训练参数;具有绘图功能,可以对比测试集的真实值与预测值;支持网络及数据的保存;具有菜单选项,用户可以通过菜单执行相应的操作;具有右键功能,用户可以通过右键快速执行相应的操作;-The BP neural network software (BPGUI) : suppo
dssTree
- 决策树java工程源码,access数据训练集-Java source tree works, access data training set
PJudgetopic
- 机器学习的方法短信情感分类,喜怒哀惧,43123条短信训练集-SMS emotion machine learning classification methods, joy, anger, sadness and fear, 43123 SMS training set
HMMSeg
- java ,隐马尔科夫的分词算法实现。包含10w条训练集,字典。也可以自己重新添加训练集。-java, hidden Markov segmentation algorithm. 10w of the training set contains dictionary. You can also add your own re-training set.
TestofFaceRecognition
- 基于BP神经网络的人脸识别系统的训练集源代码。-A face recognition system based on BP neural network the training set of source code.
Naive_Bayesian_classify_version
- 朴素贝耶稣算法进行文本分类,删除“无用词”,对训练集训练之后完成对测试集的测试,并输出测试集文档属于哪个分类-Tony simple algorithm for text classification Jesus, delete " without words" , after training set for the completion of the test set of tests and test sets the output document belongs Cat
CNN
- 用 卷积神经网络进行手写字符 识别,内含mnist训练集-Handwritten character recognition, containing mnist convolution neural network training set
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
plot_isotonic_regression
- 保序回归是寻找使训练集均方差最小的近似函数,它的优点是目标函数不要线性的。-The isotonic regression finds a non-decreasing approximation of a function while minimizing the mean squared error on the training data. The benefit of such a model is that it does not assume any form for the tar
BP_Classifier
- 用MATLAB实现的简单分类器,算法为BP神经网络,为监督学习,需要训练集(文件中附有训练集,供测试用),分类效果较好。-This program creates a Classifier to identify the gender by height and weight based on BP network.
SVM-class
- 这是关于svm的java源代码,带训练集,和测试集-This is about svm java source code, with training set and test set
fisher_classify
- MATLAB版本的LDA线性分类器,具体包括计算类内离散度矩阵,类间离散度矩阵,以及训练集各类在新坐标轴上的投影。代码原来用于肌电特征的分类,亦可用于其他机器学习案例-the LDA classifier wrote in MATLAB
BPtrain
- BP神经网络实现测试数据预测(将训练集与测试集数据进行归一化 建立BP神经网络,并训练;利用训练好的BP神经网络对测试集中的23个样本的抗压强度进行预测;输出结果并绘图)-BP neural network to predict the test data (the training set and test data set is normalized the BP neural network and training use of the trained BP neural netwo
GRNN_PNN
- 将训练集与测试集数据进行归一化; 建立GRNN或PNN神经网络; 利用建立好的神经网络对测试集中的26个乳腺组织样本的类型进行预测; 计算预测正确率(不必计算每类的正确率,只需计算正常或者病变两类的正确率,即只要预测结果与真实值属于同一大类,则认为是正确,否则认为预测错误)-The training set and test data set is normalized Establish GRNN or PNN neural network The use of wel
elmtrain
- 将整个数据集中的103个样本随机划分为训练集与测试集,其中训练集包含80个样本,测 试集包含23个样本; 建立极限学习机模型,并训练; 利用训练好的极限学习机模型对测试集中的23个样本进行预测; 输出结果并绘图(真实值与预测值对比图); -The 103 random samples of the entire data set is divided into training set and test set, wherein the training s
demoadaboost
- Adaboost是一种迭代算法,其核心思想是针对同一个训练集训练不同的分类器(弱分类器),然后把这些弱分类器集合起来,构成一个更强的最终分类器(强分类器)。-Adaboost is an iterative algorithm, the core idea is the same for a training set different classifiers (weak classifiers), and then set up these weak classifiers to form a
email
- 机器学习算法的数据集,包含训练集和测试集。主要用于邮件分类-Machine learning algorithms of data sets, including training set and testing set.Mainly used for E-mail classification
444
- 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 parameter