Примеры

Release Highlights for scikit-learn 0.23 ¶

Демонстрация алгоритма спектральной сокластеризации ¶

Демонстрация алгоритма спектральной бикластеризации ¶

Biclustering documents with the Spectral Co-clustering algorithm ¶

Comparison of Calibration of Classifiers ¶

Probability Calibration curves ¶

Probability calibration of classifiers ¶

Probability Calibration for 3-class classification ¶

Recognizing hand-written digits ¶

Normal, Ledoit-Wolf and OAS Linear Discriminant Analysis for classification ¶

Plot classification probability ¶

Classifier comparison ¶

Linear and Quadratic Discriminant Analysis with covariance ellipsoid ¶

An example of K-Means++ initialization ¶

Plot Hierarchical Clustering Dendrogram ¶

Feature agglomeration ¶

A demo of the mean-shift clustering algorithm ¶

Demonstration of k-means assumptions ¶

Online learning of a dictionary of parts of faces ¶

Vector Quantization Example ¶

Demo of affinity propagation clustering algorithm ¶

Agglomerative clustering with and without structure ¶

Various Agglomerative Clustering on a 2D embedding of digits ¶

Segmenting the picture of greek coins in regions ¶

K-means Clustering ¶

Spectral clustering for image segmentation ¶

A demo of structured Ward hierarchical clustering on an image of coins ¶

Demo of DBSCAN clustering algorithm ¶

Color Quantization using K-Means ¶

Hierarchical clustering: structured vs unstructured ward ¶

Agglomerative clustering with different metrics ¶

Inductive Clustering ¶

Demo of OPTICS clustering algorithm ¶

Compare BIRCH and MiniBatchKMeans ¶

Empirical evaluation of the impact of k-means initialization ¶

Adjustment for chance in clustering performance evaluation ¶

Comparison of the K-Means and MiniBatchKMeans clustering algorithms ¶

Feature agglomeration vs. univariate selection ¶

Comparing different hierarchical linkage methods on toy datasets ¶

A demo of K-Means clustering on the handwritten digits data ¶

Selecting the number of clusters with silhouette analysis on KMeans clustering ¶

Comparing different clustering algorithms on toy datasets ¶

Ledoit-Wolf vs OAS estimation ¶

Sparse inverse covariance estimation ¶

Shrinkage covariance estimation: LedoitWolf vs OAS and max-likelihood ¶

Robust vs Empirical covariance estimate ¶

Robust covariance estimation and Mahalanobis distances relevance ¶

Principal Component Regression vs Partial Least Squares Regression ¶

Compare cross decomposition methods ¶

The Digit Dataset ¶

The Iris Dataset ¶

Plot randomly generated classification dataset ¶

Plot randomly generated multilabel dataset ¶

Decision Tree Regression ¶

Multi-output Decision Tree Regression ¶

Plot the decision surface of a decision tree on the iris dataset ¶

Post pruning decision trees with cost complexity pruning ¶

Understanding the decision tree structure ¶

Beta-divergence loss functions ¶

PCA example with Iris Data-set ¶

Incremental PCA ¶

Comparison of LDA and PCA 2D projection of Iris dataset ¶

Factor Analysis (with rotation) to visualize patterns ¶

Blind source separation using FastICA ¶

Principal components analysis (PCA) ¶

FastICA on 2D point clouds ¶

Kernel PCA ¶

Model selection with Probabilistic PCA and Factor Analysis (FA) ¶

Sparse coding with a precomputed dictionary ¶

Image denoising using dictionary learning ¶

Faces dataset decompositions ¶

Pixel importances with a parallel forest of trees ¶

Feature importances with forests of trees ¶

Plot individual and voting regression predictions ¶

Decision Tree Regression with AdaBoost ¶

IsolationForest example ¶

Plot the decision boundaries of a VotingClassifier ¶

Comparing random forests and the multi-output meta estimator ¶

Prediction Intervals for Gradient Boosting Regression ¶

Gradient Boosting regularization ¶

Plot class probabilities calculated by the VotingClassifier ¶

Gradient Boosting regression ¶

OOB Errors for Random Forests ¶

Two-class AdaBoost ¶

Hashing feature transformation using Totally Random Trees ¶

Multi-class AdaBoosted Decision Trees ¶

Discrete versus Real AdaBoost ¶

Early stopping of Gradient Boosting ¶

Feature transformations with ensembles of trees ¶

Gradient Boosting Out-of-Bag estimates ¶

Single estimator versus bagging: bias-variance decomposition ¶

Categorical Feature Support in Gradient Boosting ¶

Plot the decision surfaces of ensembles of trees on the iris dataset ¶

Combine predictors using stacking ¶

Plot individual and voting regression predictions ¶

Monotonic Constraints ¶

Outlier detection on a real data set ¶

Out-of-core classification of text documents ¶

Outlier detection on a real data set ¶

Compressive sensing: tomography reconstruction with L1 prior (Lasso ¶

Topic extraction with Non-negative Matrix Factorization and Latent Dirichlet Allocation ¶

Faces recognition example using eigenfaces and SVMs ¶

Model Complexity Influence ¶

Visualizing the stock market structure ¶

Wikipedia principal eigenvector ¶

Species distribution modeling ¶

Libsvm GUI ¶

Prediction Latency ¶

Univariate Feature Selection ¶

Test with permutations the significance of a classification score ¶

Model-based and sequential feature selection ¶

Recursive feature elimination with cross-validation ¶

Pipeline Anova SVM ¶

Comparison of F-test and mutual information ¶

Recursive feature elimination ¶

Concentration Prior Type Analysis of Variation Bayesian Gaussian Mixture ¶

Gaussian Mixture Model Sine Curve ¶

GMM covariances ¶

Gaussian Mixture Model Selection ¶

Gaussian Mixture Model Ellipsoids ¶

Density Estimation for a Gaussian mixture ¶

Illustration of Gaussian process classification (GPC) on the XOR dataset ¶

Gaussian process classification (GPC) on iris dataset ¶

Comparison of kernel ridge and Gaussian process regression ¶

Illustration of prior and posterior Gaussian process for different kernels ¶

Iso-probability lines for Gaussian Processes classification (GPC) ¶

Probabilistic predictions with Gaussian process classification (GPC) ¶

Gaussian process regression (GPR) with noise-level estimation ¶

Gaussian Processes regression: basic introductory example ¶

Gaussian process regression (GPR) on Mauna Loa CO2 data. ¶

Gaussian processes on discrete data structures ¶

Lasso path using LARS ¶

Plot Ridge coefficients as a function of the regularization ¶

SGD: Maximum margin separating hyperplane ¶

SGD: convex loss functions ¶

Ordinary Least Squares and Ridge Regression Variance ¶

Plot Ridge coefficients as a function of the L2 regularization ¶

SGD: Penalties ¶

Logistic function ¶

Polynomial interpolation ¶

Regularization path of L1- Logistic Regression ¶

Logistic Regression 3-class Classifier ¶

SGD: Weighted samples ¶

Non-negative least squares ¶

Linear Regression Example ¶

Robust linear model estimation using RANSAC ¶

Sparsity Example: Fitting only features 1 and 2 ¶

HuberRegressor vs Ridge on dataset with strong outliers ¶

Lasso on dense and sparse data ¶

Comparing various online solvers ¶

Joint feature selection with multi-task Lasso ¶

MNIST classification using multinomial logistic + L1 ¶

Plot multi-class SGD on the iris dataset ¶

Orthogonal Matching Pursuit ¶

Lasso and Elastic Net for Sparse Signals ¶

Curve Fitting with Bayesian Ridge Regression ¶

Theil-Sen Regression ¶

Plot multinomial and One-vs-Rest Logistic Regression ¶

Robust linear estimator fitting ¶

L1 Penalty and Sparsity in Logistic Regression ¶

Lasso and Elastic Net ¶

Automatic Relevance Determination Regression (ARD) ¶

Bayesian Ridge Regression ¶

Lasso model selection: Cross-Validation / AIC / BIC ¶

Multiclass sparse logistic regression on 20newgroups ¶

Early stopping of Stochastic Gradient Descent ¶

Poisson regression and non-normal loss ¶

Tweedie regression on insurance claims ¶

Common pitfalls in interpretation of coefficients of linear models ¶

Partial Dependence and Individual Conditional Expectation Plots ¶

Permutation Importance vs Random Forest Feature Importance (MDI) ¶

Permutation Importance with Multicollinear or Correlated Features ¶

Scalable learning with polynomial kernel aproximation ¶

Manifold learning on handwritten digits: Locally Linear Embedding, Isomap… ¶

Swiss Roll reduction with LLE ¶

Comparison of Manifold Learning methods ¶

Multi-dimensional scaling ¶

t-SNE: The effect of various perplexity values on the shape ¶

Manifold Learning methods on a severed sphere ¶

Compact estimator representations ¶

ROC Curve with Visualization API ¶

Visualizations with Display Objects ¶

Isotonic Regression ¶

Advanced Plotting With Partial Dependence ¶

Face completion with a multi-output estimators ¶

Multilabel classification ¶

Comparing anomaly detection algorithms for outlier detection on toy datasets ¶

The Johnson-Lindenstrauss bound for embedding with random projections ¶

Comparison of kernel ridge regression and SVR ¶

Explicit feature map approximation for RBF kernels ¶

Imputing missing values before building an estimator ¶

Imputing missing values with variants of IterativeImputer ¶

Plotting Cross-Validated Predictions ¶

Confusion matrix ¶

Plotting Validation Curves ¶

Detection error tradeoff (DET) curve ¶

Successive Halving Iterations ¶

Underfitting vs. Overfitting ¶

Parameter estimation using grid search with cross-validation ¶

Comparing randomized search and grid search for hyperparameter estimation ¶

Train error vs Test error ¶

Receiver Operating Characteristic (ROC) with cross validation ¶

Nested versus non-nested cross-validation ¶

Sample pipeline for text feature extraction and evaluation ¶

Demonstration of multi-metric evaluation on cross_val_score and GridSearchCV ¶

Balance model complexity and cross-validated score ¶

Comparison between grid search and successive halving ¶

Visualizing cross-validation behavior in scikit-learn ¶

Receiver Operating Characteristic (ROC) ¶

Precision-Recall ¶

Plotting Learning Curves ¶

Statistical comparison of models using grid search ¶

Classifier Chain ¶

Nearest Neighbors regression ¶

Approximate nearest neighbors in TSNE ¶

Simple 1D Kernel Density Estimation ¶

Kernel Density Estimate of Species Distributions ¶

Comparing Nearest Neighbors with and without Neighborhood Components Analysis ¶

Dimensionality Reduction with Neighborhood Components Analysis ¶

Novelty detection with Local Outlier Factor (LOF) ¶

Neighborhood Components Analysis Illustration ¶

Caching nearest neighbors ¶

Nearest Neighbors Classification ¶

Kernel Density Estimation ¶

Nearest Centroid Classification ¶

Outlier detection with Local Outlier Factor (LOF) ¶

Varying regularization in Multi-layer Perceptron ¶

Compare Stochastic learning strategies for MLPClassifier ¶

Restricted Boltzmann Machine features for digit classification ¶

Visualization of MLP weights on MNIST ¶

User Guide ¶

Concatenating multiple feature extraction methods ¶

Pipelining: chaining a PCA and a logistic regression ¶

Selecting dimensionality reduction with Pipeline and GridSearchCV ¶

Column Transformer with Mixed Types ¶

Column Transformer with Heterogeneous Data Sources ¶

Using KBinsDiscretizer to discretize continuous features ¶

Demonstrating the different strategies of KBinsDiscretizer ¶

Importance of Feature Scaling ¶

Map data to a normal distribution ¶

Feature discretization ¶

Compare the effect of different scalers on data with outliers ¶

Label Propagation learning a complex structure ¶

Label Propagation digits: Demonstrating performance ¶

Decision boundary of semi-supervised classifiers versus SVM on the Iris dataset ¶

Effect of varying threshold for self-training ¶

Semi-supervised Classification on a Text Dataset ¶

Label Propagation digits active learning ¶

Non-linear SVM ¶

SVM: Maximum margin separating hyperplane ¶

SVM with custom kernel ¶

SVM Tie Breaking Example ¶

SVM: Weighted samples ¶

Plot the support vectors in LinearSVC ¶

SVM: Separating hyperplane for unbalanced classes ¶

SVM-Kernels ¶

SVM-Anova: SVM with univariate feature selection ¶

Support Vector Regression (SVR) using linear and non-linear kernels ¶

SVM Margins Example ¶

One-class SVM with non-linear kernel (RBF) ¶

Plot different SVM classifiers in the iris dataset ¶

Scaling the regularization parameter for SVCs ¶

RBF SVM parameters ¶

Digits Classification Exercise ¶

Cross-validation on Digits Dataset Exercise ¶

SVM Exercise ¶

Cross-validation on diabetes Dataset Exercise ¶

FeatureHasher and DictVectorizer Comparison ¶

Clustering text documents using k-means ¶

Classification of text documents using sparse features ¶