![](https://img-blog.csdnimg.cn/20201014180756918.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
机器学习
文章平均质量分 92
Santorinisu
记录只为分享,分享益于成长!
展开
-
机器学习-DBSCAN算法
Section I: Brief Introduction on DBSCANDensity-based Spatial Clustering of Applications with Noise (DBSCAN), which does not make assumptions about spherical clusters like k-means, nor does it partiti...原创 2020-02-24 13:30:09 · 515 阅读 · 0 评论 -
机器学习-Agglomerative聚类算法
Section I: Code Bundle and Result Analyses代码import pandas as pdimport numpy as npimport warningswarnings.filterwarnings("ignore")np.random.seed(123)#Section 1: Generate random datavariables=...原创 2020-02-23 20:10:01 · 1357 阅读 · 0 评论 -
机器学习-层级聚类算法(Hierarchy Cluster)
Section I: Brief Introduction on Hierarchy ClusterThe two standard algorithms for agglomerative hierarchical clustering are single linkage and complete linkage. Using single linkage, the distances be...原创 2020-02-23 19:59:05 · 1274 阅读 · 0 评论 -
机器学习-KMeans聚类(肘系数Elbow和轮廓系数Silhouette)
Section I: Brief Introduction on KMeans ClusterThe K-Means algorithm belongs to the category of prototype-based clustering. Prototype-based clustering means that each cluster is represented by a prot...原创 2020-02-23 17:10:06 · 11015 阅读 · 1 评论 -
机器学习-随机森林回归(Random Forest Regression)
Section I: Brief Introduction on Random Forest RegressionThe random forest algorithm is an ensemble technique that combines multiple decision trees. A random forest usually has a better generalizatio...原创 2020-02-23 13:34:17 · 11590 阅读 · 0 评论 -
机器学习-决策树回归(Decision Tree Regression)
Section I: Brief Introduction on Decision Tree RegressionAn advantage of the decision tree algorithm is that it does not require any transformation of the features if we are dealing with nonlinear da...原创 2020-02-23 13:07:34 · 894 阅读 · 0 评论 -
机器学习-Random Sample Consensus Regression(RANSAC)回归
Section I: Brief Introduction on RANSACLinear regression models can be heavily impacted by the presence of outliers. In certain situations, a very small subset of our data can have a big effect on th...原创 2020-02-23 12:29:18 · 946 阅读 · 0 评论 -
机器学习-非线性关系拟合(Linear, Quadratic和Cubic)
Section I: Code Bundle and Result AnalysesThe relationship between house prices and LSTAT(percent lower status of the population) will be fitted via the second degree (quadratic) and the third degree...原创 2020-02-22 22:56:16 · 4713 阅读 · 0 评论 -
机器学习-三种回归方法(Ridge、LASSO和ElasticNet回归)
Section I: Brief Introduction on Three Regression ModelsRegulation is one approach to tackle the problem of overfitting by adding additional information, and thereby shrinking the parameter values of...原创 2020-02-22 20:44:22 · 2472 阅读 · 0 评论 -
机器学习-线性回归(Linear Regression)
Section I: Code Bundle and Result Analyses代码from sklearn import datasetsfrom sklearn.model_selection import train_test_splitfrom sklearn.linear_model import LinearRegressionimport matplotlib.pypl...原创 2020-02-22 20:07:59 · 466 阅读 · 0 评论 -
机器学习-AdaBoost法
Section I: Brief Introduction on AdaBoostIn boosting, the ensemble consists of very simple base classifiers, also often referred to as weal learners, which often only have a slight performance advant...原创 2020-02-22 17:14:37 · 253 阅读 · 0 评论 -
机器学习-Bagging法
Section I: Code Bundle and Result AnalysesPersonal Views:Bagging In A Nutshell Lies in:Boostrap samples with replacementDraw features with replacement代码:from sklearn.preprocessing import Label...原创 2020-02-22 16:53:41 · 308 阅读 · 0 评论 -
机器学习-多数投票方式-MajorityVotingClassifier
Section I: Code Bundle and Result Analyses第一部分:三种分类算法(Pipeline)的性能比较代码:from sklearn import datasetsfrom sklearn.model_selection import train_test_splitfrom sklearn.preprocessing import StandardSc...原创 2020-02-22 15:26:04 · 4917 阅读 · 0 评论 -
机器学习-单一模型/Ensemble模型-误差率解析
Section I: Code Bundle and Result Analyses第一部分:代码from scipy.special import combimport mathimport matplotlib.pyplot as pltimport warningswarnings.filterwarnings("ignore")plt.rcParams['figure.dp...原创 2020-02-22 13:11:36 · 777 阅读 · 0 评论 -
机器学习-类别不平衡-上下采样(Upsampling and Downsampling)
Section I: Brief Introduction on Upsampling/DownsamplingClass imbalance is a quite common problem when working with real-world data-samples from one class or multiple classes are over-represented in ...原创 2020-02-22 12:05:19 · 2066 阅读 · 0 评论 -
机器学习-Receiver Operating Charateristic(ROC)
Section I: Brief Introduction on ROC CurveReceiver Operating Charateristic(ROC) graphs are usefult tools to select models forclassification based on their performance with respect to th FPR and TPR, ...原创 2020-02-21 21:02:48 · 400 阅读 · 0 评论 -
机器学习-深嵌交叉验证(Nested Cross-Validation)
Section I: Code and Analyses第一部分:代码from sklearn import datasetsfrom sklearn.model_selection import train_test_splitfrom sklearn.preprocessing import StandardScalerfrom sklearn.svm import SVCfrom...原创 2020-02-21 18:28:10 · 2362 阅读 · 0 评论 -
机器学习-超参调整-网格搜索(Grid Search)
Section I: Code and Analyses第一部分:代码from sklearn import datasetsfrom sklearn.model_selection import train_test_splitfrom sklearn.preprocessing import StandardScalerfrom sklearn.svm import SVCfrom...原创 2020-02-21 17:25:09 · 822 阅读 · 0 评论 -
机器学习-验证曲线(过拟合与欠拟合的解决)
Section I: Brief Introduction on ValidatingCurvesValidation curves are a useful tool for improving the performance of a model by addressing issues such as overfitting or underfitting. Validation curv...原创 2020-02-21 16:44:59 · 1088 阅读 · 0 评论 -
机器学习-学习曲线(过拟合与欠拟合的判断)
Section I: Brief Introduction on LearningCurvesIf a model is too complex for a given training dataset-there are too many degrees of freedom or paramters in this model-the model tends to overfit the t...原创 2020-02-21 16:07:37 · 1454 阅读 · 0 评论 -
机器学习-KFold交叉验证
Section I: Brief Introduction on StratifiedKFoldA slight improvement over the standard k-fold cross-validation approach is stratified k-fold cross-validattion, which can yeild better bias and varianc...原创 2020-02-21 15:08:00 · 1278 阅读 · 0 评论 -
机器学习-PipeLine初识
学习:A Scikit_Learn pipeline can be regarded as a meta-estimator or several tansformers and estimator can also be wrappered around together. If we call the fit method of Pipeline, the data will be pass...原创 2020-02-21 14:03:14 · 300 阅读 · 0 评论 -
机器学习-特征抽取-核主成分法-Kernel Principal Component(KPCA)
Section I: Brief Introduction on KPCAPerforming a nonlinear mapping via Kernel PCA that transforms the data onto a higher-dimensional space. Then, a standard PCA in this higher-dimensional space to p...原创 2020-02-21 12:34:57 · 638 阅读 · 0 评论 -
机器学习-特征抽取-LDA(Linear Discriminant Analysis)
Section I: Brief Introduction on LDALinear Discriminat Analysis (LDA) can be used as a technique for feature extraction to increase the computational efficiency and reduce the degree of overfitting d...原创 2020-02-21 11:45:56 · 889 阅读 · 0 评论 -
机器学习-特征抽取-主成分分析法(Principal Component Analysis)
Section I: Brief Introduction on PCAPCA helps us to identify patterns in data based on the correlation between features. In a nutshell, PCA aims to find the directions of maximum variance in high-dim...原创 2020-02-21 11:14:49 · 671 阅读 · 0 评论 -
机器学习-特征选择-随机森林
Section I: Code Bundle第一部分:Feature Importance Sorted via Random Forest代码:import matplotlib.pyplot as pltfrom sklearn import datasetsfrom sklearn.preprocessing import StandardScalerfrom sklearn...原创 2020-02-21 10:36:52 · 1075 阅读 · 0 评论 -
机器学习-特征选择-序列后向选择Sequential Backward Selection方法
Section I: Brief Introduction on Sequential Backward Selection方法The idea behind the SBS algorithm is quite simple: SBS sequentially removes features from the full feature subset until new feature sub...原创 2020-02-20 22:18:45 · 5577 阅读 · 4 评论 -
机器学习-标签数值映射
Section I: Map Ordinary Categorical Label第一部分:代码import pandas as pdimport numpy as npdf=pd.DataFrame([['green','M',10.1,'class1'], ['red','L',13.5,'class2'], ['b...原创 2020-02-20 20:49:03 · 2088 阅读 · 0 评论 -
机器学习-K近邻
Section I: Brief Introduction on K-Nearest NeighborsK-Nearest neighbors (KNN) is particularly interesting because it is fundamentallyndifferent from the other learning algorithms. KNN is a typical ex...原创 2020-02-20 19:14:18 · 217 阅读 · 0 评论 -
机器学习-随机森林(Random Forest)
Section I: Brief Introduction on Random ForestRandom forests have gained huge popularity om applications of machine learning during the last decade due to their good classification performance,scalab...原创 2020-02-20 17:04:59 · 346 阅读 · 0 评论 -
机器学习-决策树(Decision Tree)
Section I: Brief Introduction on Decision TreeDecision tree classifiers are attractive models if model interpretability is of our concern. As the name decision tree suggests, we can think of this mod...原创 2020-02-20 16:45:30 · 483 阅读 · 0 评论 -
机器学习-支持向量机(Support Vector Machine)
Section I: Brief Introduction on SVMAnother powerful and widely used learning algorithm is the Support Vector Machine (SVM), which can be considered an extension of the perceptron. Using the perceptr...原创 2020-02-20 15:04:49 · 558 阅读 · 0 评论 -
机器学习-逻辑回归(Logistic Regression)
Section I: Brief Glimpse Into Logistic RegressionLogistic regression is a classification model that is very easy to implement but performs very well on linearly separable classes. It is one of the mo...原创 2020-02-20 14:00:38 · 522 阅读 · 0 评论 -
机器学习-感知机(Perceptron)-Scikit-Learn
Section I: Load package#Section 1: Load packagefrom sklearn import datasetsimport numpy as npimport matplotlib.pyplot as pltfrom sklearn.model_selection import train_test_splitfrom sklearn.prepr...原创 2020-02-20 10:55:16 · 675 阅读 · 0 评论 -
机器学习-Adative Linear Neuron(AdaLine)
Section I: Brief Introduction on AdaLineThe key difference between the AdaLine rule and Rosenblatt’s perceptron is that the weights are updated based on a linear activation function rather than a uni...原创 2020-02-19 21:03:02 · 574 阅读 · 0 评论 -
机器学习-感知机Perceptron
Section I: Brief Introduction on PerceptronThe whole idea behind the MCP neuron and Rosenblatt’s thresholded percepton model is to use a reductionist approach to mimic how a single neuron in the brai...原创 2020-02-19 16:35:41 · 425 阅读 · 0 评论