7. Class reference¶
7.1. Support Vector Machines¶
Support Vector Machine algorithms.
| svm.SVC([C, kernel, degree, gamma, coef0, ...]) | C-Support Vector Classification. |
| svm.LinearSVC([penalty, loss, dual, tol, C, ...]) | Linear Support Vector Classification. |
| svm.NuSVC([nu, kernel, degree, gamma, ...]) | Nu-Support Vector Classification. |
| svm.SVR([kernel, degree, gamma, coef0, ...]) | epsilon-Support Vector Regression. |
| svm.NuSVR([nu, C, kernel, degree, gamma, ...]) | Nu Support Vector Regression. |
| svm.OneClassSVM([kernel, degree, gamma, ...]) | Unsupervised Outliers Detection. |
| svm.l1_min_c(X, y[, loss, fit_intercept, ...]) | Return the maximum value for C that yields a model with coefficients |
7.1.1. For sparse data¶
Support Vector Machine algorithms for sparse matrices.
This module should have the same API as scikits.learn.svm, except that matrices are expected to be in some sparse format supported by scipy.sparse.
Note
Some fields, like dual_coef_ are not sparse matrices strictly speaking. However, they are converted to a sparse matrix for consistency and efficiency when multiplying to other sparse matrices.
| svm.sparse.SVC([C, kernel, degree, gamma, ...]) | SVC for sparse matrices (csr). |
| svm.sparse.NuSVC([nu, kernel, degree, ...]) | NuSVC for sparse matrices (csr). |
| svm.sparse.SVR([kernel, degree, gamma, ...]) | SVR for sparse matrices (csr) |
| svm.sparse.NuSVR([nu, C, kernel, degree, ...]) | NuSVR for sparse matrices (csr) |
| svm.sparse.OneClassSVM([kernel, degree, ...]) | NuSVR for sparse matrices (csr) |
| svm.sparse.LinearSVC([penalty, loss, dual, ...]) | Linear Support Vector Classification, Sparse Version |
7.1.2. Low-level methods¶
| svm.libsvm.fit | Train the model using libsvm (low-level method) |
| svm.libsvm.decision_function | Predict margin (libsvm name for this is predict_values) |
| svm.libsvm.predict | Predict target values of X given a model (low-level method) |
| svm.libsvm.predict_proba | Predict probabilities svm_model stores all parameters needed to predict a given value. |
| svm.libsvm.cross_validation | Binding of the cross-validation routine (low-level routine) |
7.2. Generalized Linear Models¶
scikits.learn.linear_model is a module to fit genelarized linear models. It includes Ridge regression, Bayesian Regression, Lasso and Elastic Net estimators computed with Least Angle Regression and coordinate descent.
It also implements Stochastic Gradient Descent related algorithms.
| linear_model.LinearRegression([fit_intercept]) | Ordinary least squares Linear Regression. |
| linear_model.Ridge([alpha, fit_intercept]) | Ridge regression. |
| linear_model.RidgeCV([alphas, ...]) | Ridge regression with built-in cross-validation. |
| linear_model.Lasso([alpha, fit_intercept]) | Linear Model trained with L1 prior as regularizer (aka the Lasso) |
| linear_model.LassoCV([eps, n_alphas, ...]) | Lasso linear model with iterative fitting along a regularization path |
| linear_model.ElasticNet([alpha, rho, ...]) | Linear Model trained with L1 and L2 prior as regularizer |
| linear_model.ElasticNetCV([rho, eps, ...]) | Elastic Net model with iterative fitting along a regularization path |
| linear_model.LARS([fit_intercept, verbose]) | Least Angle Regression model a.k.a. LAR |
| linear_model.LassoLARS([alpha, ...]) | Lasso model fit with Least Angle Regression a.k.a. LARS |
| linear_model.LogisticRegression([penalty, ...]) | Logistic Regression. |
| linear_model.SGDClassifier([loss, penalty, ...]) | Linear model fitted by minimizing a regularized empirical loss with SGD. |
| linear_model.SGDRegressor([loss, penalty, ...]) | Linear model fitted by minimizing a regularized empirical loss with SGD |
| linear_model.BayesianRidge([n_iter, eps, ...]) | Bayesian ridge regression |
| linear_model.ARDRegression([n_iter, eps, ...]) | Bayesian ARD regression. |
| linear_model.lasso_path(X, y, **fit_params) | Compute Lasso path with coordinate descent |
| linear_model.lars_path(X, y[, Xy, Gram, ...]) | Compute Least Angle Regression and LASSO path |
7.2.1. For sparse data¶
scikits.learn.linear_model.sparse is the sparse counterpart of scikits.learn.linear_model.
| linear_model.sparse.Lasso([alpha, fit_intercept]) | Linear Model trained with L1 prior as regularizer |
| linear_model.sparse.ElasticNet([alpha, rho, ...]) | Linear Model trained with L1 and L2 prior as regularizer |
| linear_model.sparse.SGDClassifier([loss, ...]) | Linear model fitted by minimizing a regularized empirical loss with SGD |
| linear_model.sparse.SGDRegressor([loss, ...]) | Linear model fitted by minimizing a regularized empirical loss with SGD |
7.4. Nearest Neighbors¶
Nearest Neighbor related algorithms
| neighbors.NeighborsClassifier([n_neighbors, ...]) | Classifier implementing k-Nearest Neighbor Algorithm. |
| neighbors.NeighborsRegressor([n_neighbors, ...]) | Regression based on k-Nearest Neighbor Algorithm |
| ball_tree.BallTree |
| neighbors.kneighbors_graph(X, n_neighbors[, ...]) | Computes the (weighted) graph of k-Neighbors for points in X |
7.5. Gaussian Mixture Models¶
Gaussian Mixture Models
| mixture.GMM([n_states, cvtype]) | Gaussian Mixture Model |
7.6. Hidden Markov Models¶
| hmm.GaussianHMM([n_states, cvtype, ...]) | Hidden Markov Model with Gaussian emissions |
| hmm.MultinomialHMM([n_states, startprob, ...]) | Hidden Markov Model with multinomial (discrete) emissions |
| hmm.GMMHMM([n_states, n_mix, startprob, ...]) | Hidden Markov Model with Gaussin mixture emissions |
7.7. Clustering¶
Clustering algorithms
| cluster.KMeans([k, init, n_init, max_iter, ...]) | K-Means clustering |
| cluster.MeanShift([bandwidth]) | MeanShift clustering |
| cluster.SpectralClustering([k, mode, ...]) | Apply k-means to a projection to the normalized laplacian |
| cluster.AffinityPropagation([damping, ...]) | Perform Affinity Propagation Clustering of data |
| cluster.Ward([n_clusters, memory, ...]) | Ward hierarchical clustering: constructs a tree and cuts it. |
7.8. Metrics¶
Metrics module with score functions, performance metrics and pairwise metrics or distances computation
| metrics.euclidean_distances(X, Y[, ...]) | Considering the rows of X (and Y=X) as vectors, compute the |
| metrics.confusion_matrix(y_true, y_pred[, ...]) | Compute confusion matrix to evaluate the accuracy of a classification |
| metrics.roc_curve(y_true, y_score) | compute Receiver operating characteristic (ROC) |
| metrics.auc(x, y) | Compute Area Under the Curve (AUC) using the trapezoidal rule |
| metrics.precision_score(y_true, y_pred[, ...]) | Compute the precision |
| metrics.recall_score(y_true, y_pred[, pos_label]) | Compute the recall |
| metrics.fbeta_score(y_true, y_pred, beta[, ...]) | Compute fbeta score |
| metrics.f1_score(y_true, y_pred[, pos_label]) | Compute f1 score |
| metrics.precision_recall_fscore_support(...) | Compute precisions, recalls, f-measures and support for each class |
| metrics.classification_report(y_true, y_pred) | Build a text report showing the main classification metrics |
| metrics.precision_recall_curve(y_true, ...) | Compute precision-recall pairs for different probability thresholds |
| metrics.r2_score(y_true, y_pred) | R^2 (coefficient of determination) regression score function |
| metrics.zero_one_score(y_true, y_pred) | Zero-One classification score |
| metrics.zero_one(y_true, y_pred) | Zero-One classification loss |
| metrics.mean_square_error(y_true, y_pred) | Mean square error regression loss |
7.8.1. Pairwise metrics¶
Utilities to evaluate pairwise distances or metrics between 2 sets of points.
| metrics.pairwise.euclidean_distances(X, Y[, ...]) | Considering the rows of X (and Y=X) as vectors, compute the |
| metrics.pairwise.linear_kernel(X, Y) | Compute the linear kernel between X and Y. |
| metrics.pairwise.polynomial_kernel(X, Y[, ...]) | Compute the polynomial kernel between X and Y. |
| metrics.pairwise.rbf_kernel(X, Y[, sigma]) | Compute the rbf (gaussian) kernel between X and Y. |
7.9. Covariance Estimators¶
7.9.1. Covariance estimators¶
scikits.learn.covariance is a module to fit to estimate robustly the covariance of features given a set of points. The precision matrix defined as the inverse of the covariance is also estimated. Covariance estimation is closely related to the theory of Gaussian Graphical Models.
| covariance.Covariance | |
| covariance.ShrunkCovariance([...]) | Covariance estimator with shrinkage |
| covariance.LedoitWolf([store_precision]) | LedoitWolf Estimator |
| covariance.ledoit_wolf(X[, assume_centered]) | Estimates the shrunk Ledoit-Wolf covariance matrix. |
| covariance.shrunk_covariance(emp_cov[, ...]) | Calculates a covariance matrix shrunk on the diagonal |
| covariance.oas(X[, assume_centered]) | Estimate covariance with the Oracle Approximating Shrinkage algorithm. |
7.10. Signal Decomposition¶
Matrix decomposition algorithms
| decomposition.PCA([n_components, copy, whiten]) | Principal component analysis (PCA) |
| decomposition.ProbabilisticPCA([...]) | Additional layer on top of PCA that adds a probabilistic evaluation |
| decomposition.RandomizedPCA(n_components[, ...]) | Principal component analysis (PCA) using randomized SVD |
| decomposition.KernelPCA([n_components, ...]) | Kernel Principal component analysis (KPCA) |
| decomposition.FastICA([n_components, ...]) | FastICA; a fast algorithm for Independent Component Analysis |
| decomposition.NMF([n_components, init, ...]) | Non-Negative matrix factorization by Projected Gradient (NMF) |
| decomposition.fastica(X[, n_components, ...]) | Perform Fast Independent Component Analysis. |
7.11. Linear Discriminant Analysis¶
| lda.LDA([n_components, priors]) | Linear Discriminant Analysis (LDA) |
7.12. Cross Validation¶
Utilities for cross validation and performance evaluation
| cross_val.LeaveOneOut(n[, indices]) | Leave-One-Out cross validation iterator |
| cross_val.LeavePOut(n, p[, indices]) | Leave-P-Out cross validation iterator |
| cross_val.KFold(n, k[, indices]) | K-Folds cross validation iterator |
| cross_val.StratifiedKFold(y, k[, indices]) | Stratified K-Folds cross validation iterator |
| cross_val.LeaveOneLabelOut(labels[, indices]) | Leave-One-Label_Out cross-validation iterator |
| cross_val.LeavePLabelOut(labels, p[, indices]) | Leave-P-Label_Out cross-validation iterator |
7.13. Grid Search¶
Tune the parameters of an estimator by cross-validation
| grid_search.GridSearchCV(estimator, param_grid) | Grid search on the parameters of a classifier |
7.14. Feature Selection¶
Feature slection module for python
| feature_selection.rfe.RFE([estimator, ...]) | Feature ranking with Recursive feature elimination |
| feature_selection.rfe.RFECV([estimator, ...]) | Feature ranking with Recursive feature elimination and cross validation |
7.15. Feature Extraction¶
Package for modules that deal with feature extraction from raw data
7.15.1. From images¶
Utilities to extract features from images.
| feature_extraction.image.img_to_graph(img[, ...]) | Graph of the pixel-to-pixel gradient connections |
| feature_extraction.image.grid_to_graph(n_x, n_y) | Graph of the pixel-to-pixel connections |
7.15.2. From text¶
Utilities to build dense feature vectors from text documents
| feature_extraction.text.RomanPreprocessor | Fast preprocessor suitable for roman languages .. |
| feature_extraction.text.WordNGramAnalyzer([...]) | Simple analyzer: transform a text document into a sequence of word tokens |
| feature_extraction.text.CharNGramAnalyzer([...]) | Compute character n-grams features of a text document |
| feature_extraction.text.CountVectorizer([...]) | Convert a collection of raw documents to a matrix of token counts |
| feature_extraction.text.TfidfTransformer([...]) | Transform a count matrix to a TF or TF-IDF representation |
| feature_extraction.text.Vectorizer([...]) | Convert a collection of raw documents to a matrix |
7.16. Pipeline¶
Pipeline: chain transforms and estimators to build a composite estimator.
| pipeline.Pipeline(steps) | Pipeline of transforms with a final estimator |
7.17. Partial Least Squares¶
Partial Least Square
| pls.PLSRegression([n_components, scale, ...]) | PLS regression (Also known PLS2 or PLS in case of one dimensional |
| pls.PLSCanonical([n_components, scale, ...]) | PLS canonical. PLSCanonical inherits from PLS with mode=”A” and |
| pls.CCA([n_components, scale, algorithm, ...]) | CCA Canonical Correlation Analysis. CCA inherits from PLS with |
| pls.PLSSVD([n_components, scale, copy]) | Partial Least Square SVD |