▍1. tagging_rnn_tf
循环神经网络工程实例,包括下载用于训练和测试的标签数据、网络模型和标签分类。(Examples of RNN engineering include downloading label data, network model and label classification for training and testing.)
循环神经网络工程实例,包括下载用于训练和测试的标签数据、网络模型和标签分类。(Examples of RNN engineering include downloading label data, network model and label classification for training and testing.)
说明: pytorch-resnet实现分类的功能,好用。(Python RESNET realizes the function of classification, which is easy to use)
说明: 1,載入iris 數劇集並設置提取60%的數據做為訓練集,40%數據做為測試集執行KNN 分類算法,並執行五次的交叉驗證(CV)和顯示準確率和混淆矩陣,並找出最佳K值。 2,載入iris 數劇集並設置提取60%的數據做為訓練集,40%數據做為測試集,執行支持向量機(SVM) 算法,並執行五次的交叉驗證(CV)和顯示準確率和混淆矩陣。 3,載入iris 數據集並執行KMEANS 聚類算法家數具分成五個群體,顯示準確率 4,載入iris 數據集執行線性迴歸算法,並利用特徵萃取(Feature Extraction)中的Recursive feature elimination (RFE)對iris 數據集中的特徵欄位進行重要性排序(1. Load iris series and set 60% data as training set and 40% data as test set to implement KNN classification algorithm, and perform five times of cross validation (CV) and display accuracy and confusion matrix, and find out the best K value. 2. Load iris series and set 60% data as training set and 40% data as test set, execute support vector machine (SVM) algorithm, and perform five times of cross validation (CV) and display accuracy and confusion matrix. 3. Loading iris data set and executing kmeans clustering algorithm, the number of users is divided into five groups, showing the accuracy rate 4. Load iris data set, execute linear regression algorithm, and use recursive feature elimination (RFE) in feature extraction to sort the importance of feature fields in iris dataset)
说明: python爬虫案例,以hao123网站爬虫为例,文件包含.py文件等源码(the example of python spider)
SVM: 一种分类器,采用最大化分类间隔进行优化参数。 关于这个分类器两点比较重要: 1)SMO优化算法需要掌握, 可以具体参看两篇文章,John Platt的文章 以及“Improvements to Platt s SMO algorithm for SVM Classifier Design” 2)核函数的使用,如何将核函数使用到SVM中,核函数就是空间转换的函数, 说白了就是距离计算函数,如何将同类之间的距离计算的比较近,如何将低维空间转换到易于分类的高维空间。 我自己看书写的程序,采用Python实现,注释比较多(SVM: a classification, a classification intervals to maximize the use of optimization parameters. About this classifier is more important points: 1) SMO optimization algorithms need to know, you can see the specific two articles, John Platt article And "Improvements to Platt s SMO algorithm for SVM Classifier Design" 2) the use of nuclear function, how to use the SVM kernel function, the kernel function is a function space conversion, That white is the distance calculation function, how relatively close distance between similar calculations, how to convert low-dimensional space to a high-dimensional space is easy classification. I see myself writing program, using Python, notes more)
说明: 利用jupyternotebook对LSTM网络仿真练习,实现一个LSTM+DNN结构的预测网络。主要使用的keras建模。tensorflow2.0(A prediction network of LSTM + DNN structure is realized by using jupyter notebook to simulate LSTM network. The main use of keras modeling. tensorflow2.0)
说明: 对轴承的时频图进行特征提取与融合,构建轴承健康指示量(Feature extraction and fusion of bearing time-frequency map are carried out to construct bearing health indicator)
说明: 利用CNN实现boston房价的预测,内含代码的详细讲解。(Use CNN to realize the prediction of Boston house price, including the detailed explanation of the code.)
说明: 对于想要学习机器学习的人来说。个人感觉是比较好的教材,业内人称西瓜书(The best textbook for machine learning)
说明: k-means实战,包括一个具体的例子。非常适合初级学习k-means聚类算法的人群(K-means actual combat, including a specific example)
本代码是使用lstm进行时间序列预测,能够很清晰的说明如何使用lstm(Time series prediction using LSTM)
说明: 用CNN对电能质量扰动进行分类,可以直接使用,数据都在里面,仅供参考(Use CNN to classify power quality disturbances, which can be used directly, and the data are in it for reference only)
说明: 机器视觉,行人识别,hog特征提取的python语言实现(Machine vision, pedestrian recognition, hog feature extraction)
说明: 机器学习实战中chapter03中所有案例的代码,带详细讲解说明(Machine learning actual combat chapter 03 in all cases of code, with detailed explanation)
用卷积神经网络(CNN)实现人脸识别,效果还可以,一个是training的程序,可以训练网络。一个是use程序,可以识别人脸(Using convolution neural network (CNN) to achieve face recognition, the effect is also available, one is the training program, can train the network. One is a use program that recognizes a face)
说明: 一些聚类常用的数据集,自己做实验也在用的,很好用也很齐全。(Some of the commonly used clustering data sets, their own experiments are also used, very easy to use and very complete.)
中科大高级计算机网络第二次试验: 1)使用Mininet 、MinEdit模拟数据中心网络; 2)使用Controller,实现由链路故障和没有故障时的两种情况; 3)不使用Controller的模拟,手动配置Fattree的路由转发表; (USTC Advanced Computer Network Second test: 1) the use of Mininet, MinEdit analog data center network 2) Controller, implemented by the two cases and no link failure fault 3) without analog Controller, manual configuration Fattree the routing tables )
中科大高级计算机网络第三次作业:使用Pyretic完成一个基于mac学习的第二层防火墙(USTC Advanced Computer Network third job: Using Pyretic complete a second layer firewall mac-based learning)
利用遗传算法来解决经典的旅行商TSP问题,python 编码(Using genetic algorithm to solve the classic TSP traveling salesman problem, the programming language is Python)