From f9926488297677bea8c4249a524c74bef5d60489 Mon Sep 17 00:00:00 2001 From: Jesse Krijthe Date: Mon, 27 Aug 2018 09:53:25 +0200 Subject: [PATCH] Fixes issue #8 and updates roxygen --- DESCRIPTION | 2 +- R/LearningCurve.R | 4 ++-- man/CrossValidationSSL.Rd | 15 ++++++++------- man/EMLeastSquaresClassifier.Rd | 1 + man/EMLinearDiscriminantClassifier.Rd | 1 + man/GRFClassifier.Rd | 6 ++++-- man/Generate2ClassGaussian.Rd | 4 +++- man/GenerateFourClusters.Rd | 1 + man/GenerateSlicedCookie.Rd | 1 + man/ICLeastSquaresClassifier.Rd | 8 +++++--- man/ICLinearDiscriminantClassifier.Rd | 1 + man/KernelICLeastSquaresClassifier.Rd | 7 ++++--- man/KernelLeastSquaresClassifier.Rd | 1 + man/LaplacianKernelLeastSquaresClassifier.Rd | 1 + man/LaplacianSVM.Rd | 1 + man/LearningCurveSSL.Rd | 18 ++++++++++-------- man/LeastSquaresClassifier.Rd | 4 +++- man/LinearDiscriminantClassifier.Rd | 1 + man/LinearSVM.Rd | 1 + man/LinearTSVM.Rd | 5 +++-- man/LogisticLossClassifier.Rd | 5 +++-- man/LogisticRegression.Rd | 1 + man/LogisticRegressionFast.Rd | 4 ++-- man/MCLinearDiscriminantClassifier.Rd | 1 + man/MCNearestMeanClassifier.Rd | 1 + man/MCPLDA.Rd | 1 + man/MajorityClassClassifier.Rd | 1 + man/NearestMeanClassifier.Rd | 4 +++- man/QuadraticDiscriminantClassifier.Rd | 1 + man/S4VM.Rd | 1 + man/SSLDataFrameToMatrices.Rd | 1 + man/SVM.Rd | 1 + man/SelfLearning.Rd | 5 +++-- man/TSVM.Rd | 5 +++-- man/USMLeastSquaresClassifier.Rd | 1 + man/WellSVM.Rd | 1 + man/add_missinglabels_mar.Rd | 1 + man/df_to_matrices.Rd | 1 + man/evaluation-measures.Rd | 17 +++++++++-------- man/generateABA.Rd | 1 + man/generateCrescentMoon.Rd | 1 + man/generateParallelPlanes.Rd | 1 + man/generateSpirals.Rd | 1 + man/generateTwoCircles.Rd | 1 + man/line_coefficients-methods.Rd | 1 - man/loss-methods.Rd | 14 ++++---------- man/missing_labels.Rd | 1 + man/rssl-predict.Rd | 15 ++++++++------- man/split_dataset_ssl.Rd | 1 + man/split_random.Rd | 1 + man/stat_classifier.Rd | 4 ++-- man/svmlin.Rd | 6 ++++-- man/true_labels.Rd | 1 + 53 files changed, 116 insertions(+), 69 deletions(-) diff --git a/DESCRIPTION b/DESCRIPTION index 054b972..dfa22f2 100644 --- a/DESCRIPTION +++ b/DESCRIPTION @@ -79,4 +79,4 @@ Collate: 'svmd.R' 'svmlin.R' 'testdata-data.R' -RoxygenNote: 6.0.1 +RoxygenNote: 6.1.0 diff --git a/R/LearningCurve.R b/R/LearningCurve.R index 88f7891..d822876 100644 --- a/R/LearningCurve.R +++ b/R/LearningCurve.R @@ -21,9 +21,9 @@ print.LearningCurve <- function(x,...) { #' #' \code{measures} is a named list of performance measures. These are functions that accept seven arguments: a trained classifier, a numeric design matrix of the labeled objects, a factor of labels, a numeric design matrix of unlabeled objects and a factor of labels for the unlabeled objects, a numeric design matrix of the test objects and a factor of labels of the test objects. See \code{\link{measure_accuracy}} for an example. #' -#' This function allows for two different types of learning curves to be generated. If \code{type="unlabeled"}, the number of labeled objects remains fixed at the value of \code{n_l}, where \code{sizes} controls the number of unlabeled objects. \code{n_test} controls the number of objects used for the test set, while all remaining objects are used if \code{with_replacement=FALSE} if which case objects are drawn without replacement from the input dataset. We make sure each class is represented by at least \code{n_min} labeled objects of each class. For \code{n_l}, additional options include: "enough" which takes the max of the number of features and 20, max(ncol(X)+5,20), "d" which takes the number of features or "2d" which takes 2 times the number of features. +#' This function allows for two different types of learning curves to be generated. If \code{type="unlabeled"}, the number of labeled objects remains fixed at the value of \code{n_l}, where \code{sizes} controls the number of unlabeled objects. \code{n_test} controls the number of objects used for the test set, while all remaining objects are used if \code{with_replacement=FALSE} in which case objects are drawn without replacement from the input dataset. We make sure each class is represented by at least \code{n_min} labeled objects of each class. For \code{n_l}, additional options include: "enough" which takes the max of the number of features and 20, max(ncol(X)+5,20), "d" which takes the number of features or "2d" which takes 2 times the number of features. #' -#' If \code{type="unlabeled"} the total number of objects remains fixed, while the fraction of labeled objects is changed. \code{frac} sets the fractions of labeled objects that should be considered, while \code{test_fraction} determines the amount of objects left out to serve as the test set. +#' If \code{type="fraction"} the total number of objects remains fixed, while the fraction of labeled objects is changed. \code{frac} sets the fractions of labeled objects that should be considered, while \code{test_fraction} determines the fraction of the total number of objects left out to serve as the test set. #' #' @family RSSL utilities #' diff --git a/man/CrossValidationSSL.Rd b/man/CrossValidationSSL.Rd index bc80c00..702936d 100644 --- a/man/CrossValidationSSL.Rd +++ b/man/CrossValidationSSL.Rd @@ -8,13 +8,14 @@ \usage{ CrossValidationSSL(X, y, ...) -\method{CrossValidationSSL}{list}(X, y, ..., verbose = FALSE, mc.cores = 1) - -\method{CrossValidationSSL}{matrix}(X, y, classifiers, measures = list(Error = - measure_error), k = 10, repeats = 1, verbose = FALSE, - leaveout = "test", n_labeled = 10, prop_unlabeled = 0.5, time = TRUE, - pre_scale = FALSE, pre_pca = FALSE, n_min = 1, low_level_cores = 1, - ...) +\method{CrossValidationSSL}{list}(X, y, ..., verbose = FALSE, + mc.cores = 1) + +\method{CrossValidationSSL}{matrix}(X, y, classifiers, + measures = list(Error = measure_error), k = 10, repeats = 1, + verbose = FALSE, leaveout = "test", n_labeled = 10, + prop_unlabeled = 0.5, time = TRUE, pre_scale = FALSE, + pre_pca = FALSE, n_min = 1, low_level_cores = 1, ...) } \arguments{ \item{X}{design matrix of the labeled objects} diff --git a/man/EMLeastSquaresClassifier.Rd b/man/EMLeastSquaresClassifier.Rd index b5a4f1b..125cbed 100644 --- a/man/EMLeastSquaresClassifier.Rd +++ b/man/EMLeastSquaresClassifier.Rd @@ -105,3 +105,4 @@ Other RSSL classifiers: \code{\link{EMLinearDiscriminantClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/EMLinearDiscriminantClassifier.Rd b/man/EMLinearDiscriminantClassifier.Rd index 746135f..e49a7cd 100644 --- a/man/EMLinearDiscriminantClassifier.Rd +++ b/man/EMLinearDiscriminantClassifier.Rd @@ -57,3 +57,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/GRFClassifier.Rd b/man/GRFClassifier.Rd index fd7ff0d..0fc1579 100644 --- a/man/GRFClassifier.Rd +++ b/man/GRFClassifier.Rd @@ -4,8 +4,9 @@ \alias{GRFClassifier} \title{Label propagation using Gaussian Random Fields and Harmonic functions} \usage{ -GRFClassifier(X, y, X_u, adjacency = "nn", adjacency_distance = "euclidean", - adjacency_k = 6, adjacency_sigma = 0.1, class_mass_normalization = TRUE, +GRFClassifier(X, y, X_u, adjacency = "nn", + adjacency_distance = "euclidean", adjacency_k = 6, + adjacency_sigma = 0.1, class_mass_normalization = TRUE, scale = FALSE, x_center = FALSE) } \arguments{ @@ -108,3 +109,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/Generate2ClassGaussian.Rd b/man/Generate2ClassGaussian.Rd index 8b95b0a..7b8663a 100644 --- a/man/Generate2ClassGaussian.Rd +++ b/man/Generate2ClassGaussian.Rd @@ -4,7 +4,8 @@ \alias{generate2ClassGaussian} \title{Generate data from 2 Gaussian distributed classes} \usage{ -generate2ClassGaussian(n = 10000, d = 100, var = 1, expected = TRUE) +generate2ClassGaussian(n = 10000, d = 100, var = 1, + expected = TRUE) } \arguments{ \item{n}{integer; Number of examples to generate} @@ -31,3 +32,4 @@ Other RSSL datasets: \code{\link{generateABA}}, \code{\link{generateSpirals}}, \code{\link{generateTwoCircles}} } +\concept{RSSL datasets} diff --git a/man/GenerateFourClusters.Rd b/man/GenerateFourClusters.Rd index 211be0b..90bc4e2 100644 --- a/man/GenerateFourClusters.Rd +++ b/man/GenerateFourClusters.Rd @@ -29,3 +29,4 @@ Other RSSL datasets: \code{\link{generate2ClassGaussian}}, \code{\link{generateSpirals}}, \code{\link{generateTwoCircles}} } +\concept{RSSL datasets} diff --git a/man/GenerateSlicedCookie.Rd b/man/GenerateSlicedCookie.Rd index e1db89b..fc00d84 100644 --- a/man/GenerateSlicedCookie.Rd +++ b/man/GenerateSlicedCookie.Rd @@ -32,3 +32,4 @@ Other RSSL datasets: \code{\link{generate2ClassGaussian}}, \code{\link{generateSpirals}}, \code{\link{generateTwoCircles}} } +\concept{RSSL datasets} diff --git a/man/ICLeastSquaresClassifier.Rd b/man/ICLeastSquaresClassifier.Rd index 6e6ab37..85d1895 100644 --- a/man/ICLeastSquaresClassifier.Rd +++ b/man/ICLeastSquaresClassifier.Rd @@ -5,9 +5,10 @@ \title{Implicitly Constrained Least Squares Classifier} \usage{ ICLeastSquaresClassifier(X, y, X_u = NULL, lambda1 = 0, lambda2 = 0, - intercept = TRUE, x_center = FALSE, scale = FALSE, method = "LBFGS", - projection = "supervised", lambda_prior = 0, trueprob = NULL, - eps = 1e-09, y_scale = FALSE, use_Xu_for_scaling = TRUE) + intercept = TRUE, x_center = FALSE, scale = FALSE, + method = "LBFGS", projection = "supervised", lambda_prior = 0, + trueprob = NULL, eps = 1e-09, y_scale = FALSE, + use_Xu_for_scaling = TRUE) } \arguments{ \item{X}{Design matrix, intercept term is added within the function} @@ -105,3 +106,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/ICLinearDiscriminantClassifier.Rd b/man/ICLinearDiscriminantClassifier.Rd index d21723b..6667256 100644 --- a/man/ICLinearDiscriminantClassifier.Rd +++ b/man/ICLinearDiscriminantClassifier.Rd @@ -56,3 +56,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/KernelICLeastSquaresClassifier.Rd b/man/KernelICLeastSquaresClassifier.Rd index ca930c2..b0f5272 100644 --- a/man/KernelICLeastSquaresClassifier.Rd +++ b/man/KernelICLeastSquaresClassifier.Rd @@ -4,9 +4,10 @@ \alias{KernelICLeastSquaresClassifier} \title{Kernelized Implicitly Constrained Least Squares Classification} \usage{ -KernelICLeastSquaresClassifier(X, y, X_u, lambda = 0, kernel = vanilladot(), - x_center = TRUE, scale = TRUE, y_scale = TRUE, lambda_prior = 0, - classprior = 0, method = "LBFGS", projection = "semisupervised") +KernelICLeastSquaresClassifier(X, y, X_u, lambda = 0, + kernel = vanilladot(), x_center = TRUE, scale = TRUE, + y_scale = TRUE, lambda_prior = 0, classprior = 0, + method = "LBFGS", projection = "semisupervised") } \arguments{ \item{X}{matrix; Design matrix for labeled data} diff --git a/man/KernelLeastSquaresClassifier.Rd b/man/KernelLeastSquaresClassifier.Rd index 546e579..43e4156 100644 --- a/man/KernelLeastSquaresClassifier.Rd +++ b/man/KernelLeastSquaresClassifier.Rd @@ -127,3 +127,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/LaplacianKernelLeastSquaresClassifier.Rd b/man/LaplacianKernelLeastSquaresClassifier.Rd index bda728e..137278c 100644 --- a/man/LaplacianKernelLeastSquaresClassifier.Rd +++ b/man/LaplacianKernelLeastSquaresClassifier.Rd @@ -150,3 +150,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/LaplacianSVM.Rd b/man/LaplacianSVM.Rd index e663ca6..671a0c0 100644 --- a/man/LaplacianSVM.Rd +++ b/man/LaplacianSVM.Rd @@ -160,3 +160,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/LearningCurveSSL.Rd b/man/LearningCurveSSL.Rd index d254110..7edeb15 100644 --- a/man/LearningCurveSSL.Rd +++ b/man/LearningCurveSSL.Rd @@ -7,12 +7,13 @@ \usage{ LearningCurveSSL(X, y, ...) -\method{LearningCurveSSL}{matrix}(X, y, classifiers, measures = list(Accuracy - = measure_accuracy), type = "unlabeled", n_l = NULL, - with_replacement = FALSE, sizes = 2^(1:8), n_test = 1000, - repeats = 100, verbose = FALSE, n_min = 1, dataset_name = NULL, - test_fraction = NULL, fracs = seq(0.1, 0.9, 0.1), time = TRUE, - pre_scale = FALSE, pre_pca = FALSE, low_level_cores = 1, ...) +\method{LearningCurveSSL}{matrix}(X, y, classifiers, + measures = list(Accuracy = measure_accuracy), type = "unlabeled", + n_l = NULL, with_replacement = FALSE, sizes = 2^(1:8), + n_test = 1000, repeats = 100, verbose = FALSE, n_min = 1, + dataset_name = NULL, test_fraction = NULL, fracs = seq(0.1, 0.9, + 0.1), time = TRUE, pre_scale = FALSE, pre_pca = FALSE, + low_level_cores = 1, ...) } \arguments{ \item{X}{design matrix} @@ -66,9 +67,9 @@ Evaluate semi-supervised classifiers for different amounts of unlabeled training \code{measures} is a named list of performance measures. These are functions that accept seven arguments: a trained classifier, a numeric design matrix of the labeled objects, a factor of labels, a numeric design matrix of unlabeled objects and a factor of labels for the unlabeled objects, a numeric design matrix of the test objects and a factor of labels of the test objects. See \code{\link{measure_accuracy}} for an example. -This function allows for two different types of learning curves to be generated. If \code{type="unlabeled"}, the number of labeled objects remains fixed at the value of \code{n_l}, where \code{sizes} controls the number of unlabeled objects. \code{n_test} controls the number of objects used for the test set, while all remaining objects are used if \code{with_replacement=FALSE} if which case objects are drawn without replacement from the input dataset. We make sure each class is represented by at least \code{n_min} labeled objects of each class. For \code{n_l}, additional options include: "enough" which takes the max of the number of features and 20, max(ncol(X)+5,20), "d" which takes the number of features or "2d" which takes 2 times the number of features. +This function allows for two different types of learning curves to be generated. If \code{type="unlabeled"}, the number of labeled objects remains fixed at the value of \code{n_l}, where \code{sizes} controls the number of unlabeled objects. \code{n_test} controls the number of objects used for the test set, while all remaining objects are used if \code{with_replacement=FALSE} in which case objects are drawn without replacement from the input dataset. We make sure each class is represented by at least \code{n_min} labeled objects of each class. For \code{n_l}, additional options include: "enough" which takes the max of the number of features and 20, max(ncol(X)+5,20), "d" which takes the number of features or "2d" which takes 2 times the number of features. -If \code{type="unlabeled"} the total number of objects remains fixed, while the fraction of labeled objects is changed. \code{frac} sets the fractions of labeled objects that should be considered, while \code{test_fraction} determines the amount of objects left out to serve as the test set. +If \code{type="fraction"} the total number of objects remains fixed, while the fraction of labeled objects is changed. \code{frac} sets the fractions of labeled objects that should be considered, while \code{test_fraction} determines the fraction of the total number of objects left out to serve as the test set. } \examples{ set.seed(1) @@ -118,3 +119,4 @@ Other RSSL utilities: \code{\link{SSLDataFrameToMatrices}}, \code{\link{split_dataset_ssl}}, \code{\link{split_random}}, \code{\link{true_labels}} } +\concept{RSSL utilities} diff --git a/man/LeastSquaresClassifier.Rd b/man/LeastSquaresClassifier.Rd index e0a3e87..0b3513e 100644 --- a/man/LeastSquaresClassifier.Rd +++ b/man/LeastSquaresClassifier.Rd @@ -5,7 +5,8 @@ \title{Least Squares Classifier} \usage{ LeastSquaresClassifier(X, y, lambda = 0, intercept = TRUE, - x_center = FALSE, scale = FALSE, method = "inverse", y_scale = FALSE) + x_center = FALSE, scale = FALSE, method = "inverse", + y_scale = FALSE) } \arguments{ \item{X}{matrix; Design matrix for labeled data} @@ -58,3 +59,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/LinearDiscriminantClassifier.Rd b/man/LinearDiscriminantClassifier.Rd index 987eae4..dd283de 100644 --- a/man/LinearDiscriminantClassifier.Rd +++ b/man/LinearDiscriminantClassifier.Rd @@ -56,3 +56,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/LinearSVM.Rd b/man/LinearSVM.Rd index ff1417d..5ed84a4 100644 --- a/man/LinearSVM.Rd +++ b/man/LinearSVM.Rd @@ -55,3 +55,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/LinearTSVM.Rd b/man/LinearTSVM.Rd index bebede7..23baf9c 100644 --- a/man/LinearTSVM.Rd +++ b/man/LinearTSVM.Rd @@ -4,8 +4,8 @@ \alias{LinearTSVM} \title{Linear CCCP Transductive SVM classifier} \usage{ -LinearTSVM(X, y, X_u, C, Cstar, s = 0, x_center = FALSE, scale = FALSE, - eps = 1e-06, verbose = FALSE, init = NULL) +LinearTSVM(X, y, X_u, C, Cstar, s = 0, x_center = FALSE, + scale = FALSE, eps = 1e-06, verbose = FALSE, init = NULL) } \arguments{ \item{X}{matrix; Design matrix, intercept term is added within the function} @@ -61,3 +61,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/LogisticLossClassifier.Rd b/man/LogisticLossClassifier.Rd index 1afa917..84f03f4 100644 --- a/man/LogisticLossClassifier.Rd +++ b/man/LogisticLossClassifier.Rd @@ -4,8 +4,8 @@ \alias{LogisticLossClassifier} \title{Logistic Loss Classifier} \usage{ -LogisticLossClassifier(X, y, lambda = 0, intercept = TRUE, scale = FALSE, - init = NA, x_center = FALSE, ...) +LogisticLossClassifier(X, y, lambda = 0, intercept = TRUE, + scale = FALSE, init = NA, x_center = FALSE, ...) } \arguments{ \item{X}{Design matrix, intercept term is added within the function} @@ -56,3 +56,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/LogisticRegression.Rd b/man/LogisticRegression.Rd index bb91eef..9ceca6a 100644 --- a/man/LogisticRegression.Rd +++ b/man/LogisticRegression.Rd @@ -49,3 +49,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/LogisticRegressionFast.Rd b/man/LogisticRegressionFast.Rd index 18ab650..886a2fd 100644 --- a/man/LogisticRegressionFast.Rd +++ b/man/LogisticRegressionFast.Rd @@ -4,8 +4,8 @@ \alias{LogisticRegressionFast} \title{Logistic Regression implementation that uses R's glm} \usage{ -LogisticRegressionFast(X, y, lambda = 0, intercept = TRUE, scale = FALSE, - init = NA, x_center = FALSE) +LogisticRegressionFast(X, y, lambda = 0, intercept = TRUE, + scale = FALSE, init = NA, x_center = FALSE) } \arguments{ \item{X}{matrix; Design matrix for labeled data} diff --git a/man/MCLinearDiscriminantClassifier.Rd b/man/MCLinearDiscriminantClassifier.Rd index 7b83617..d6ed945 100644 --- a/man/MCLinearDiscriminantClassifier.Rd +++ b/man/MCLinearDiscriminantClassifier.Rd @@ -57,3 +57,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/MCNearestMeanClassifier.Rd b/man/MCNearestMeanClassifier.Rd index 05f4820..bd1a20c 100644 --- a/man/MCNearestMeanClassifier.Rd +++ b/man/MCNearestMeanClassifier.Rd @@ -58,3 +58,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/MCPLDA.Rd b/man/MCPLDA.Rd index 7407170..3a1614e 100644 --- a/man/MCPLDA.Rd +++ b/man/MCPLDA.Rd @@ -49,3 +49,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/MajorityClassClassifier.Rd b/man/MajorityClassClassifier.Rd index b892629..4d105c6 100644 --- a/man/MajorityClassClassifier.Rd +++ b/man/MajorityClassClassifier.Rd @@ -40,3 +40,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/NearestMeanClassifier.Rd b/man/NearestMeanClassifier.Rd index a11f695..a192480 100644 --- a/man/NearestMeanClassifier.Rd +++ b/man/NearestMeanClassifier.Rd @@ -4,7 +4,8 @@ \alias{NearestMeanClassifier} \title{Nearest Mean Classifier} \usage{ -NearestMeanClassifier(X, y, prior = NULL, x_center = FALSE, scale = FALSE) +NearestMeanClassifier(X, y, prior = NULL, x_center = FALSE, + scale = FALSE) } \arguments{ \item{X}{matrix; Design matrix for labeled data} @@ -53,3 +54,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/QuadraticDiscriminantClassifier.Rd b/man/QuadraticDiscriminantClassifier.Rd index 6f20307..3b85809 100644 --- a/man/QuadraticDiscriminantClassifier.Rd +++ b/man/QuadraticDiscriminantClassifier.Rd @@ -53,3 +53,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/S4VM.Rd b/man/S4VM.Rd index 6fa9e41..f3728c5 100644 --- a/man/S4VM.Rd +++ b/man/S4VM.Rd @@ -104,3 +104,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/SSLDataFrameToMatrices.Rd b/man/SSLDataFrameToMatrices.Rd index 6e99c74..bc8a651 100644 --- a/man/SSLDataFrameToMatrices.Rd +++ b/man/SSLDataFrameToMatrices.Rd @@ -30,3 +30,4 @@ Other RSSL utilities: \code{\link{LearningCurveSSL}}, \code{\link{split_dataset_ssl}}, \code{\link{split_random}}, \code{\link{true_labels}} } +\concept{RSSL utilities} diff --git a/man/SVM.Rd b/man/SVM.Rd index cc4a657..4457662 100644 --- a/man/SVM.Rd +++ b/man/SVM.Rd @@ -58,3 +58,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/SelfLearning.Rd b/man/SelfLearning.Rd index 63334c8..af79073 100644 --- a/man/SelfLearning.Rd +++ b/man/SelfLearning.Rd @@ -4,8 +4,8 @@ \alias{SelfLearning} \title{Self-Learning approach to Semi-supervised Learning} \usage{ -SelfLearning(X, y, X_u = NULL, method, prob = FALSE, cautious = FALSE, - max_iter = 100, ...) +SelfLearning(X, y, X_u = NULL, method, prob = FALSE, + cautious = FALSE, max_iter = 100, ...) } \arguments{ \item{X}{matrix; Design matrix for labeled data} @@ -66,3 +66,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/TSVM.Rd b/man/TSVM.Rd index ea31bb4..3693bae 100644 --- a/man/TSVM.Rd +++ b/man/TSVM.Rd @@ -5,8 +5,8 @@ \title{Transductive SVM classifier using the convex concave procedure} \usage{ TSVM(X, y, X_u, C, Cstar, kernel = kernlab::vanilladot(), - balancing_constraint = TRUE, s = 0, x_center = TRUE, scale = FALSE, - eps = 1e-09, max_iter = 20, verbose = FALSE) + balancing_constraint = TRUE, s = 0, x_center = TRUE, + scale = FALSE, eps = 1e-09, max_iter = 20, verbose = FALSE) } \arguments{ \item{X}{matrix; Design matrix for labeled data} @@ -152,3 +152,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/USMLeastSquaresClassifier.Rd b/man/USMLeastSquaresClassifier.Rd index 4b16c6b..4e6d93e 100644 --- a/man/USMLeastSquaresClassifier.Rd +++ b/man/USMLeastSquaresClassifier.Rd @@ -59,3 +59,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{SelfLearning}}, \code{\link{TSVM}}, \code{\link{WellSVM}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/WellSVM.Rd b/man/WellSVM.Rd index 5d4fd1e..bac2267 100644 --- a/man/WellSVM.Rd +++ b/man/WellSVM.Rd @@ -82,3 +82,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{svmlin}} } +\concept{RSSL classifiers} diff --git a/man/add_missinglabels_mar.Rd b/man/add_missinglabels_mar.Rd index ea95022..997f3cd 100644 --- a/man/add_missinglabels_mar.Rd +++ b/man/add_missinglabels_mar.Rd @@ -25,3 +25,4 @@ Other RSSL utilities: \code{\link{LearningCurveSSL}}, \code{\link{split_dataset_ssl}}, \code{\link{split_random}}, \code{\link{true_labels}} } +\concept{RSSL utilities} diff --git a/man/df_to_matrices.Rd b/man/df_to_matrices.Rd index 98f821e..d7bfcf2 100644 --- a/man/df_to_matrices.Rd +++ b/man/df_to_matrices.Rd @@ -23,3 +23,4 @@ Other RSSL utilities: \code{\link{LearningCurveSSL}}, \code{\link{split_dataset_ssl}}, \code{\link{split_random}}, \code{\link{true_labels}} } +\concept{RSSL utilities} diff --git a/man/evaluation-measures.Rd b/man/evaluation-measures.Rd index 5db4e20..276c70c 100644 --- a/man/evaluation-measures.Rd +++ b/man/evaluation-measures.Rd @@ -8,20 +8,20 @@ \alias{measure_losstrain} \title{Performance measures used in classifier evaluation} \usage{ -measure_accuracy(trained_classifier, X_l = NULL, y_l = NULL, X_u = NULL, - y_u = NULL, X_test = NULL, y_test = NULL) +measure_accuracy(trained_classifier, X_l = NULL, y_l = NULL, + X_u = NULL, y_u = NULL, X_test = NULL, y_test = NULL) measure_error(trained_classifier, X_l = NULL, y_l = NULL, X_u = NULL, y_u = NULL, X_test = NULL, y_test = NULL) -measure_losstest(trained_classifier, X_l = NULL, y_l = NULL, X_u = NULL, - y_u = NULL, X_test = NULL, y_test = NULL) +measure_losstest(trained_classifier, X_l = NULL, y_l = NULL, + X_u = NULL, y_u = NULL, X_test = NULL, y_test = NULL) -measure_losslab(trained_classifier, X_l = NULL, y_l = NULL, X_u = NULL, - y_u = NULL, X_test = NULL, y_test = NULL) +measure_losslab(trained_classifier, X_l = NULL, y_l = NULL, + X_u = NULL, y_u = NULL, X_test = NULL, y_test = NULL) -measure_losstrain(trained_classifier, X_l = NULL, y_l = NULL, X_u = NULL, - y_u = NULL, X_test = NULL, y_test = NULL) +measure_losstrain(trained_classifier, X_l = NULL, y_l = NULL, + X_u = NULL, y_u = NULL, X_test = NULL, y_test = NULL) } \arguments{ \item{trained_classifier}{the trained classifier object} @@ -61,3 +61,4 @@ Other RSSL utilities: \code{\link{LearningCurveSSL}}, \code{\link{split_dataset_ssl}}, \code{\link{split_random}}, \code{\link{true_labels}} } +\concept{RSSL utilities} diff --git a/man/generateABA.Rd b/man/generateABA.Rd index 3687af2..748419d 100644 --- a/man/generateABA.Rd +++ b/man/generateABA.Rd @@ -29,3 +29,4 @@ Other RSSL datasets: \code{\link{generate2ClassGaussian}}, \code{\link{generateSpirals}}, \code{\link{generateTwoCircles}} } +\concept{RSSL datasets} diff --git a/man/generateCrescentMoon.Rd b/man/generateCrescentMoon.Rd index 8674be4..65a1d31 100644 --- a/man/generateCrescentMoon.Rd +++ b/man/generateCrescentMoon.Rd @@ -29,3 +29,4 @@ Other RSSL datasets: \code{\link{generate2ClassGaussian}}, \code{\link{generateSpirals}}, \code{\link{generateTwoCircles}} } +\concept{RSSL datasets} diff --git a/man/generateParallelPlanes.Rd b/man/generateParallelPlanes.Rd index 187d59d..896ee08 100644 --- a/man/generateParallelPlanes.Rd +++ b/man/generateParallelPlanes.Rd @@ -32,3 +32,4 @@ Other RSSL datasets: \code{\link{generate2ClassGaussian}}, \code{\link{generateSpirals}}, \code{\link{generateTwoCircles}} } +\concept{RSSL datasets} diff --git a/man/generateSpirals.Rd b/man/generateSpirals.Rd index 6592e2e..6e79893 100644 --- a/man/generateSpirals.Rd +++ b/man/generateSpirals.Rd @@ -27,3 +27,4 @@ Other RSSL datasets: \code{\link{generate2ClassGaussian}}, \code{\link{generateSlicedCookie}}, \code{\link{generateTwoCircles}} } +\concept{RSSL datasets} diff --git a/man/generateTwoCircles.Rd b/man/generateTwoCircles.Rd index b4b6f0d..736105e 100644 --- a/man/generateTwoCircles.Rd +++ b/man/generateTwoCircles.Rd @@ -23,3 +23,4 @@ Other RSSL datasets: \code{\link{generate2ClassGaussian}}, \code{\link{generateSlicedCookie}}, \code{\link{generateSpirals}} } +\concept{RSSL datasets} diff --git a/man/line_coefficients-methods.Rd b/man/line_coefficients-methods.Rd index c7eb628..81053d9 100644 --- a/man/line_coefficients-methods.Rd +++ b/man/line_coefficients-methods.Rd @@ -8,7 +8,6 @@ \alias{line_coefficients} \alias{line_coefficients,LeastSquaresClassifier-method} \alias{line_coefficients,NormalBasedClassifier-method} -\alias{line_coefficients,LeastSquaresClassifier-method} \alias{line_coefficients,LogisticRegression-method} \alias{line_coefficients,LinearSVM-method} \alias{line_coefficients,LogisticLossClassifier-method} diff --git a/man/loss-methods.Rd b/man/loss-methods.Rd index 7111499..5936e86 100644 --- a/man/loss-methods.Rd +++ b/man/loss-methods.Rd @@ -15,7 +15,6 @@ \alias{loss,LogisticLossClassifier-method} \alias{loss,MajorityClassClassifier-method} \alias{loss,SVM-method} -\alias{loss,LinearSVM-method} \alias{loss,SelfLearning-method} \alias{loss,USMLeastSquaresClassifier-method} \alias{loss,svmlinClassifier-method} @@ -29,7 +28,8 @@ loss(object, ...) \S4method{loss}{LogisticRegression}(object, newdata, y = NULL) -\S4method{loss}{KernelLeastSquaresClassifier}(object, newdata, y = NULL, ...) +\S4method{loss}{KernelLeastSquaresClassifier}(object, newdata, y = NULL, + ...) \S4method{loss}{LinearSVM}(object, newdata, y = NULL) @@ -41,7 +41,8 @@ loss(object, ...) \S4method{loss}{SelfLearning}(object, newdata, y = NULL, ...) -\S4method{loss}{USMLeastSquaresClassifier}(object, newdata, y = NULL, ...) +\S4method{loss}{USMLeastSquaresClassifier}(object, newdata, y = NULL, + ...) \S4method{loss}{svmlinClassifier}(object, newdata, y = NULL) } @@ -58,14 +59,7 @@ loss(object, ...) numeric; the total loss on the test data } \description{ -Loss of a classifier or regression function - -Return the negative log likelihood on the given dataset - Hinge loss on new objects of a trained LinearSVM Hinge loss on new objects of a trained SVM - -Loss method for SelfLearning Classifier -The loss method delegates prediction to the specific model object } diff --git a/man/missing_labels.Rd b/man/missing_labels.Rd index 9f525a2..aa3b86c 100644 --- a/man/missing_labels.Rd +++ b/man/missing_labels.Rd @@ -21,3 +21,4 @@ Other RSSL utilities: \code{\link{LearningCurveSSL}}, \code{\link{split_dataset_ssl}}, \code{\link{split_random}}, \code{\link{true_labels}} } +\concept{RSSL utilities} diff --git a/man/rssl-predict.Rd b/man/rssl-predict.Rd index 208c5c9..ba92a2e 100644 --- a/man/rssl-predict.Rd +++ b/man/rssl-predict.Rd @@ -13,7 +13,6 @@ \alias{responsibilities,GRFClassifier-method} \alias{predict,GRFClassifier-method} \alias{predict,KernelLeastSquaresClassifier-method} -\alias{predict,LeastSquaresClassifier-method} \alias{predict,LinearSVM-method} \alias{predict,LogisticLossClassifier-method} \alias{predict,MajorityClassClassifier-method} @@ -25,8 +24,8 @@ \alias{predict,svmlinClassifier-method} \title{Predict using RSSL classifier} \usage{ -\S4method{predict}{LeastSquaresClassifier}(object, newdata, probs = FALSE, - ...) +\S4method{predict}{LeastSquaresClassifier}(object, newdata, + probs = FALSE, ...) \S4method{predict}{NormalBasedClassifier}(object, newdata, probs = FALSE) @@ -41,16 +40,18 @@ \S4method{predict}{LinearSVM}(object, newdata, probs = FALSE) -\S4method{predict}{LogisticLossClassifier}(object, newdata, probs = FALSE) +\S4method{predict}{LogisticLossClassifier}(object, newdata, + probs = FALSE) -\S4method{predict}{MajorityClassClassifier}(object, newdata, probs = FALSE) +\S4method{predict}{MajorityClassClassifier}(object, newdata, + probs = FALSE) \S4method{predict}{SVM}(object, newdata) \S4method{predict}{SelfLearning}(object, newdata, ...) -\S4method{predict}{USMLeastSquaresClassifier}(object, newdata, probs = FALSE, - ...) +\S4method{predict}{USMLeastSquaresClassifier}(object, newdata, + probs = FALSE, ...) \S4method{predict}{WellSVM}(object, newdata, probs = FALSE, ...) diff --git a/man/split_dataset_ssl.Rd b/man/split_dataset_ssl.Rd index 2313a7c..9dcd218 100644 --- a/man/split_dataset_ssl.Rd +++ b/man/split_dataset_ssl.Rd @@ -27,3 +27,4 @@ Other RSSL utilities: \code{\link{LearningCurveSSL}}, \code{\link{missing_labels}}, \code{\link{split_random}}, \code{\link{true_labels}} } +\concept{RSSL utilities} diff --git a/man/split_random.Rd b/man/split_random.Rd index 1ab1279..819df07 100644 --- a/man/split_random.Rd +++ b/man/split_random.Rd @@ -40,3 +40,4 @@ Other RSSL utilities: \code{\link{LearningCurveSSL}}, \code{\link{split_dataset_ssl}}, \code{\link{true_labels}} } +\concept{RSSL utilities} diff --git a/man/stat_classifier.Rd b/man/stat_classifier.Rd index e47cab3..96b96bb 100644 --- a/man/stat_classifier.Rd +++ b/man/stat_classifier.Rd @@ -5,8 +5,8 @@ \title{Plot RSSL classifier boundaries} \usage{ stat_classifier(mapping = NULL, data = NULL, show.legend = NA, - inherit.aes = TRUE, breaks = 0, precision = 50, brute_force = FALSE, - classifiers = classifiers, ...) + inherit.aes = TRUE, breaks = 0, precision = 50, + brute_force = FALSE, classifiers = classifiers, ...) } \arguments{ \item{mapping}{aes; aesthetic mapping} diff --git a/man/svmlin.Rd b/man/svmlin.Rd index 7944c55..dd5b6c9 100644 --- a/man/svmlin.Rd +++ b/man/svmlin.Rd @@ -5,8 +5,9 @@ \title{svmlin implementation by Sindhwani & Keerthi (2006)} \usage{ svmlin(X, y, X_u = NULL, algorithm = 1, lambda = 1, lambda_u = 1, - max_switch = 10000, pos_frac = 0.5, Cp = 1, Cn = 1, verbose = FALSE, - intercept = TRUE, scale = FALSE, x_center = FALSE) + max_switch = 10000, pos_frac = 0.5, Cp = 1, Cn = 1, + verbose = FALSE, intercept = TRUE, scale = FALSE, + x_center = FALSE) } \arguments{ \item{X}{Matrix or sparseMatrix containing the labeled feature vectors, without intercept} @@ -95,3 +96,4 @@ Other RSSL classifiers: \code{\link{EMLeastSquaresClassifier}}, \code{\link{USMLeastSquaresClassifier}}, \code{\link{WellSVM}} } +\concept{RSSL classifiers} diff --git a/man/true_labels.Rd b/man/true_labels.Rd index 44faaef..b356ce3 100644 --- a/man/true_labels.Rd +++ b/man/true_labels.Rd @@ -22,3 +22,4 @@ Other RSSL utilities: \code{\link{LearningCurveSSL}}, \code{\link{split_dataset_ssl}}, \code{\link{split_random}} } +\concept{RSSL utilities}