2
0
Fork 0

- update: doc,

- wip: cleanup,
- update: datasets,
- remove: old code
This commit is contained in:
Daniel Kapla 2019-12-16 17:34:35 +01:00
parent 300fc11f3f
commit 9edefe994d
84 changed files with 812 additions and 4077 deletions

View File

@ -9,15 +9,10 @@ export(cve)
export(cve.call)
export(dataset)
export(directions)
export(elem.pairs)
export(estimate.bandwidth)
export(null)
export(predict_dim)
export(projTangentStiefel)
export(rStiefel)
export(retractStiefel)
export(skew)
export(sym)
import(stats)
importFrom(graphics,boxplot)
importFrom(graphics,lines)

View File

@ -20,7 +20,7 @@
#' zero random variable with finite \eqn{Var(\epsilon) = E(\epsilon^2)}, \eqn{g}
#' is an unknown, continuous non-constant function,
#' and \eqn{B = (b_1, ..., b_k)} is
#' a real \eqn{p \times k}{p x k} of rank \eqn{k <= p}{k \leq p}.
#' a real \eqn{p \times k}{p x k} of rank \eqn{k \leq p}{k <= p}.
#' Without loss of generality \eqn{B} is assumed to be orthonormal.
#'
#' @author Daniel Kapla, Lukas Fertl, Bura Efstathia
@ -36,26 +36,46 @@
#' @inherit CVE-package description
#'
#' @param formula an object of class \code{"formula"} which is a symbolic
#' description of the model to be fitted.
#' description of the model to be fitted like \eqn{Y\sim X}{Y ~ X} where
#' \eqn{Y} is a \eqn{n}-dimensional vector of the response variable and
#' \eqn{X} is a \eqn{n\times p}{n x p} matrix of the predictors.
#' @param data an optional data frame, containing the data for the formula if
#' supplied.
#' @param method specifies the CVE method variation as one of
#' supplied like \code{data <- data.frame(Y, X)} with dimension
#' \eqn{n \times (p + 1)}{n x (p + 1)}. By default the variables are taken from
#' the environment from which \code{cve} is called.
#' @param method This character string specifies the method of fitting. The
#' options are
#' \itemize{
#' \item "simple" exact implementation as described in the paper listed
#' below.
#' \item "weighted" variation with addaptive weighting of slices.
#' \item "simple" implementation as described in the paper.
#' \item "weighted" variation with adaptive weighting of slices.
#' }
#' see paper.
#' @param max.dim upper bounds for \code{k}, (ignored if \code{k} is supplied).
#' @param ... Parameters passed on to \code{cve.call}.
#' @param ... optional parameters passed on to \code{cve.call}.
#'
#' @return an S3 object of class \code{cve} with components:
#' \describe{
#' \item{X}{Original training data,}
#' \item{Y}{Responce of original training data,}
#' \item{X}{design matrix of predictor vector used for calculating
#' cve-estimate,}
#' \item{Y}{\eqn{n}-dimensional vector of responses used for calculating
#' cve-estimate,}
#' \item{method}{Name of used method,}
#' \item{call}{the matched call,}
#' \item{res}{list of components \code{V, L, B, loss, h} and \code{k} for
#' each \eqn{k=min.dim,...,max.dim} (dimension).}
#' \item{res}{list of components \code{V, L, B, loss, h} for
#' each \code{k = min.dim, ..., max.dim}. If \code{k} was supplied in the
#' call \code{min.dim = max.dim = k}.
#' \itemize{
#' \item \code{B} is the cve-estimate with dimension
#' \eqn{p\times k}{p x k}.
#' \item \code{V} is the orthogonal complement of \eqn{B}.
#' \item \code{L} is the loss for each sample seperatels such that
#' it's mean is \code{loss}.
#' \item \code{loss} is the value of the target function that is
#' minimized, evaluated at \eqn{V}.
#' \item \code{h} bandwidth parameter used to calculate
#' \code{B, V, loss, L}.
#' }
#' }
#' }
#'
#' @examples
@ -66,7 +86,7 @@
#' b1 <- rep(1 / sqrt(p), p)
#' b2 <- (-1)^seq(1, p) / sqrt(p)
#' B <- cbind(b1, b2)
#' # samplsize
#' # sample size
#' n <- 200
#' set.seed(21)
#' # creat predictor data x ~ N(0, I_p)
@ -139,10 +159,12 @@ cve <- function(formula, data, method = "simple", max.dim = 10L, ...) {
#' @inherit cve title
#' @inherit cve description
#'
#' @param X Design matrix with dimension \eqn{n\times p}{n x p}.
#' @param Y numeric array of length \eqn{n} of Responses.
#' @param h bandwidth or function to estimate bandwidth, defaults to internaly
#' estimated bandwidth.
#' @param nObs parameter for choosing bandwidth \code{h} using
#' \code{\link{estimate.bandwidth}} (ignored if \code{h} is supplied).
#' @param X data matrix with samples in its rows.
#' @param Y Responses (1 dimensional).
#' @param method specifies the CVE method variation as one of
#' \itemize{
#' \item "simple" exact implementation as described in the paper listed
@ -156,19 +178,23 @@ cve <- function(formula, data, method = "simple", max.dim = 10L, ...) {
#' @param tau Initial step-size.
#' @param tol Tolerance for break condition.
#' @param max.iter maximum number of optimization steps.
#' @param attempts number of arbitrary different starting points.
#' @param logger a logger function (only for advanced user, significantly slows
#' down the computation).
#' @param h bandwidth or function to estimate bandwidth, defaults to internaly
#' estimated bandwidth.
#' @param momentum number of [0, 1) giving the ration of momentum for eucledian
#' gradient update with a momentum term.
#' @param attempts If \code{V.init} not supplied, the optimization is carried
#' out \code{attempts} times with starting values drawn from the invariant
#' measure on the Stiefel manifold (see \code{\link{rStiefel}}).
#' @param momentum number of \eqn{[0, 1)} giving the ration of momentum for
#' eucledian gradient update with a momentum term. \code{momentum = 0}
#' corresponds to normal gradient descend.
#' @param slack Positive scaling to allow small increases of the loss while
#' optimizing.
#' @param gamma step-size reduction multiple.
#' optimizing, i.e. \code{slack = 0.1} allows the target function to
#' increase up to \eqn{10 \%} in one optimization step.
#' @param gamma step-size reduction multiple. If gradient step with step size
#' \code{tau} is not accepted \code{gamma * tau} is set to the next step
#' size.
#' @param V.init Semi-orthogonal matrix of dimensions `(ncol(X), ncol(X) - k)
#' as optimization starting value. (If supplied, \code{attempts} is
#' set to 1 and \code{k} to match dimension)
#' used as starting value in the optimization. (If supplied,
#' \code{attempts} is set to 0 and \code{k} to match dimension).
#' @param logger a logger function (only for advanced user, slows down the
#' computation).
#'
#' @inherit cve return
#'
@ -253,6 +279,7 @@ cve.call <- function(X, Y, method = "simple",
stop("Dimension missmatch of 'V.init' and 'X'")
}
min.dim <- max.dim <- ncol(X) - ncol(V.init)
storage.mode(V.init) <- "double"
attempts <- 0L
} else if (missing(k) || is.null(k)) {
min.dim <- as.integer(min.dim)
@ -320,6 +347,9 @@ cve.call <- function(X, Y, method = "simple",
}
}
# Convert numerical values to "double".
storage.mode(X) <- storage.mode(Y) <- "double"
if (is.function(logger)) {
loggerEnv <- environment(logger)
} else {

View File

@ -1,8 +1,10 @@
#' Gets estimated SDR basis.
#'
#' Returns the SDR basis matrix for SDR dimension(s).
#' Returns the SDR basis matrix for dimension \code{k}, i.e. returns the
#' cve-estimate with dimension \eqn{p\times k}{p x k}.
#'
#' @param object instance of \code{cve} as output from \code{\link{cve}} or
#' \code{\link{cve.call}}
#' \code{\link{cve.call}}.
#' @param k the SDR dimension.
#' @param ... ignored.
#'

279
CVE/R/datasets.R Normal file
View File

@ -0,0 +1,279 @@
#' Multivariate Normal Distribution.
#'
#' Random generation for the multivariate normal distribution.
#' \deqn{X \sim N_p(\mu, \Sigma)}{X ~ N_p(\mu, \Sigma)}
#'
#' @param n number of samples.
#' @param mu mean
#' @param sigma covariance matrix.
#'
#' @return a \eqn{n\times p}{n x p} matrix with samples in its rows.
#'
#' @examples
#' \dontrun{
#' rmvnorm(20, sigma = matrix(c(2, 1, 1, 2), 2))
#' rmvnorm(20, mu = c(3, -1, 2))
#' }
#' @keywords internal
rmvnorm <- function(n = 1, mu = rep(0, p), sigma = diag(p)) {
if (!missing(sigma)) {
p <- nrow(sigma)
} else if (!missing(mu)) {
mu <- matrix(mu, ncol = 1)
p <- nrow(mu)
} else {
stop("At least one of 'mu' or 'sigma' must be supplied.")
}
# See: https://en.wikipedia.org/wiki/Multivariate_normal_distribution
return(rep(mu, each = n) + matrix(rnorm(n * p), n) %*% chol(sigma))
}
#' Multivariate t distribution.
#'
#' Random generation from multivariate t distribution (student distribution).
#'
#' @param n number of samples.
#' @param mu mean
#' @param sigma a \eqn{k\times k}{k x k} positive definite matrix. If the degree
#' \eqn{\nu} if bigger than 2 the created covariance is
#' \deqn{var(x) = \Sigma\frac{\nu}{\nu - 2}}
#' for \eqn{\nu > 2}.
#' @param df degree of freedom \eqn{\nu}.
#'
#' @return a \eqn{n\times p}{n x p} matrix with samples in its rows.
#'
#' @examples
#' \dontrun{
#' rmvt(20, c(0, 1), matrix(c(3, 1, 1, 2), 2), 3)
#' rmvt(20, sigma = matrix(c(2, 1, 1, 2), 2), 3)
#' rmvt(20, mu = c(3, -1, 2), 3)
#' }
#' @keywords internal
rmvt <- function(n = 1, mu = rep(0, p), sigma = diag(p), df = Inf) {
if (!missing(sigma)) {
p <- nrow(sigma)
} else if (!missing(mu)) {
mu <- matrix(mu, ncol = 1)
p <- nrow(mu)
} else {
stop("At least one of 'mu' or 'sigma' must be supplied.")
}
if (df == Inf) {
Z <- 1
} else {
Z <- sqrt(df / rchisq(n, df))
}
return(rmvnorm(n, sigma = sigma) * Z + rep(mu, each = n))
}
#' Generalized Normal Distribution.
#'
#' Random generation for generalized Normal Distribution.
#'
#' @param n Number of generated samples.
#' @param mu mean.
#' @param alpha first shape parameter.
#' @param beta second shape parameter.
#'
#' @return numeric array of length \eqn{n}.
#'
#' @seealso https://en.wikipedia.org/wiki/Generalized_normal_distribution
#' @keywords internal
rgnorm <- function(n = 1, mu = 0, alpha = 1, beta = 1) {
if (alpha <= 0 | beta <= 0) {
stop("alpha and beta must be positive.")
}
lambda <- (1 / alpha)^beta
scales <- qgamma(runif(n), shape = 1 / beta, scale = 1 / lambda)^(1 / beta)
return(scales * ((-1)^rbinom(n, 1, 0.5)) + mu)
}
#' Laplace distribution
#'
#' Random generation for Laplace distribution.
#'
#' @param n Number of generated samples.
#' @param mu mean.
#' @param sd standard deviation.
#'
#' @return numeric array of length \eqn{n}.
#'
#' @seealso https://en.wikipedia.org/wiki/Laplace_distribution
#' @keywords internal
rlaplace <- function(n = 1, mu = 0, sd = 1) {
U <- runif(n, -0.5, 0.5)
scale <- sd / sqrt(2)
return(mu - scale * sign(U) * log(1 - 2 * abs(U)))
}
#' Generates test datasets.
#'
#' Provides sample datasets M1-M7 used in the paper Conditional variance
#' estimation for sufficient dimension reduction, Lukas Fertl, Efstathia Bura.
#' The general model is given by:
#' \deqn{Y = g(B'X) + \epsilon}
#'
#' @param name One of \code{"M1"}, \code{"M2"}, \code{"M3"}, \code{"M4",}
#' \code{"M5"}, \code{"M6"} or \code{"M7"}. Alternative just the dataset number
#' 1-7.
#' @param n number of samples.
#' @param p Dimension of random variable \eqn{X}.
#' @param sd standard diviation for error term \eqn{\epsilon}.
#' @param ... Additional parameters only for "M2" (namely \code{pmix} and
#' \code{lambda}), see: below.
#'
#' @return List with elements
#' \itemize{
#' \item{X}{data, a \eqn{n\times p}{n x p} matrix.}
#' \item{Y}{response.}
#' \item{B}{the dim-reduction matrix}
#' \item{name}{Name of the dataset (name parameter)}
#' }
#'
#' @section M1:
#' The predictors are distributed as
#' \eqn{X\sim N_p(0, \Sigma)}{X ~ N_p(0, \Sigma)} with
#' \eqn{\Sigma_{i, j} = 0.5^{|i - j|}}{\Sigma_ij = 0.5^|i - j|} for
#' \eqn{i, j = 1,..., p} for a subspace dimension of \eqn{k = 1} with a default
#' of \eqn{n = 100} data points. \eqn{p = 20},
#' \eqn{b_1 = (1,1,1,1,1,1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_1 = (1,1,1,1,1,1,0,...,0)' / sqrt(6)}, and \eqn{Y} is
#' given as \deqn{Y = cos(b_1'X) + \epsilon} where \eqn{\epsilon} is
#' distributed as generalized normal distribution with location 0,
#' shape-parameter 0.5, and the scale-parameter is chosen such that
#' \eqn{Var(\epsilon) = 0.5}.
#' @section M2:
#' The predictors are distributed as \eqn{X \sim Z 1_p \lambda + N_p(0, I_p)}{X ~ Z 1_p \lambda + N_p(0, I_p)}. with
#' \eqn{Z \sim 2 Binom(p_{mix}) - 1\in\{-1, 1\}}{Z~2Binom(pmix)-1} where
#' \eqn{1_p} is the \eqn{p}-dimensional vector of one's, for a subspace
#' dimension of \eqn{k = 1} with a default of \eqn{n = 100} data points.
#' \eqn{p = 20}, \eqn{b_1 = (1,1,1,1,1,1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_1 = (1,1,1,1,1,1,0,...,0)' / sqrt(6)},
#' and \eqn{Y} is \deqn{Y = cos(b_1'X) + 0.5\epsilon} where \eqn{\epsilon} is
#' standard normal.
#' Defaults for \code{pmix} is 0.3 and \code{lambda} defaults to 1.
#' @section M3:
#' The predictors are distributed as \eqn{X\sim N_p(0, I_p)}{X~N_p(0, I_p)}
#' for a subspace
#' dimension of \eqn{k = 1} with a default of \eqn{n = 100} data points.
#' \eqn{p = 20}, \eqn{b_1 = (1,1,1,1,1,1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_1 = (1,1,1,1,1,1,0,...,0)' / sqrt(6)},
#' and \eqn{Y} is
#' \deqn{Y = 2 log(|b_1'X| + 2) + 0.5\epsilon} where \eqn{\epsilon} is
#' standard normal.
#' @section M4:
#' The predictors are distributed as \eqn{X\sim N_p(0,\Sigma)}{X~N_p(0,\Sigma)}
#' with \eqn{\Sigma_{i, j} = 0.5^{|i - j|}}{\Sigma_ij = 0.5^|i - j|} for
#' \eqn{i, j = 1,..., p} for a subspace dimension of \eqn{k = 2} with a default
#' of \eqn{n = 100} data points. \eqn{p = 20},
#' \eqn{b_1 = (1,1,1,1,1,1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_1 = (1,1,1,1,1,1,0,...,0)' / sqrt(6)},
#' \eqn{b_2 = (1,-1,1,-1,1,-1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_2 = (1,-1,1,-1,1,-1,0,...,0)' / sqrt(6)}
#' and \eqn{Y} is given as \deqn{Y = \frac{b_1'X}{0.5 + (1.5 + b_2'X)^2} + 0.5\epsilon}{Y = (b_1'X) / (0.5 + (1.5 + b_2'X)^2) + 0.5\epsilon}
#' where \eqn{\epsilon} is standard normal.
#' @section M5:
#' The predictors are distributed as \eqn{X\sim U([0,1]^p)}{X~U([0, 1]^p)}
#' where \eqn{U([0, 1]^p)} is the uniform distribution with
#' independent components on the \eqn{p}-dimensional hypercube for a subspace
#' dimension of \eqn{k = 2} with a default of \eqn{n = 200} data points.
#' \eqn{p = 20},
#' \eqn{b_1 = (1,1,1,1,1,1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_1 = (1,1,1,1,1,1,0,...,0)' / sqrt(6)},
#' \eqn{b_2 = (1,-1,1,-1,1,-1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_2 = (1,-1,1,-1,1,-1,0,...,0)' / sqrt(6)}
#' and \eqn{Y} is given as \deqn{Y = cos(\pi b_1'X)(b_2'X + 1)^2 + 0.5\epsilon}
#' where \eqn{\epsilon} is standard normal.
#' @section M6:
#' The predictors are distributed as \eqn{X\sim N_p(0, I_p)}{X~N_p(0, I_p)}
#' for a subspace dimension of \eqn{k = 3} with a default of \eqn{n = 200} data
#' point. \eqn{p = 20, b_1 = e_1, b_2 = e_2}, and \eqn{b_3 = e_p}, where
#' \eqn{e_j} is the \eqn{j}-th unit vector in the \eqn{p}-dimensional space.
#' \eqn{Y} is given as \deqn{Y = (b_1'X)^2+(b_2'X)^2+(b_3'X)^2+0.5\epsilon}
#' where \eqn{\epsilon} is standard normal.
#' @section M7:
#' The predictors are distributed as \eqn{X\sim t_3(I_p)}{X~t_3(I_p)} where
#' \eqn{t_3(I_p)} is the standard multivariate t-distribution with 3 degrees of
#' freedom, for a subspace dimension of \eqn{k = 4} with a default of
#' \eqn{n = 200} data points.
#' \eqn{p = 20, b_1 = e_1, b_2 = e_2, b_3 = e_3}, and \eqn{b_4 = e_p}, where
#' \eqn{e_j} is the \eqn{j}-th unit vector in the \eqn{p}-dimensional space.
#' \eqn{Y} is given as \deqn{Y = (b_1'X)(b_2'X)^2+(b_3'X)(b_4'X)+0.5\epsilon}
#' where \eqn{\epsilon} is distributed as generalized normal distribution with
#' location 0, shape-parameter 1, and the scale-parameter is chosen such that
#' \eqn{Var(\epsilon) = 0.25}.
#'
#' @references Fertl Lukas, Bura Efstathia. (2019), Conditional Variance
#' Estimation for Sufficient Dimension Reduction. Working Paper.
#'
#' @import stats
#' @importFrom stats rnorm rbinom
#' @export
dataset <- function(name = "M1", n = NULL, p = 20, sd = 0.5, ...) {
name <- toupper(name)
if (nchar(name) == 1) { name <- paste0("M", name) }
if (name == "M1") {
if (missing(n)) { n <- 100 }
# B ... `p x 1`
B <- matrix(c(rep(1 / sqrt(6), 6), rep(0, p - 6)), ncol = 1)
X <- rmvnorm(n, sigma = 0.5^abs(outer(1:p, 1:p, FUN = `-`)))
beta <- 0.5
Y <- cos(X %*% B) + rgnorm(n, 0,
alpha = sqrt(sd^2 * gamma(1 / beta) / gamma(3 / beta)),
beta = beta
)
} else if (name == "M2") {
if (missing(n)) { n <- 100 }
params <- list(...)
pmix <- if (is.null(params$pmix)) { 0.3 } else { params$pmix }
lambda <- if (is.null(params$lambda)) { 1 } else { params$lambda }
# B ... `p x 1`
B <- matrix(c(rep(1 / sqrt(6), 6), rep(0, p - 6)), ncol = 1)
Z <- 2 * rbinom(n, 1, pmix) - 1
X <- matrix(rep(lambda * Z, p) + rnorm(n * p), n)
Y <- cos(X %*% B) + rnorm(n, 0, sd)
} else if (name == "M3") {
if (missing(n)) { n <- 100 }
# B ... `p x 1`
B <- matrix(c(rep(1 / sqrt(6), 6), rep(0, p - 6)), ncol = 1)
X <- matrix(rnorm(n * p), n)
Y <- 2 * log(2 + abs(X %*% B)) + rnorm(n, 0, sd)
} else if (name == "M4") {
if (missing(n)) { n <- 200 }
# B ... `p x 2`
B <- cbind(
c(rep(1 / sqrt(6), 6), rep(0, p - 6)),
c(rep(c(1, -1), 3) / sqrt(6), rep(0, p - 6))
)
X <- rmvnorm(n, sigma = 0.5^abs(outer(1:p, 1:p, FUN = `-`)))
XB <- X %*% B
Y <- (XB[, 1]) / (0.5 + (XB[, 2] + 1.5)^2) + rnorm(n, 0, sd)
} else if (name == "M5") {
if (missing(n)) { n <- 200 }
# B ... `p x 2`
B <- cbind(
c(rep(1, 6), rep(0, p - 6)),
c(rep(c(1, -1), 3), rep(0, p - 6))
) / sqrt(6)
X <- matrix(runif(n * p), n)
XB <- X %*% B
Y <- cos(XB[, 1] * pi) * (XB[, 2] + 1)^2 + rnorm(n, 0, sd)
} else if (name == "M6") {
if (missing(n)) { n <- 200 }
# B ... `p x 3`
B <- diag(p)[, -(3:(p - 1))]
X <- matrix(rnorm(n * p), n)
Y <- rowSums((X %*% B)^2) + rnorm(n, 0, sd)
} else if (name == "M7") {
if (missing(n)) { n <- 400 }
# B ... `p x 4`
B <- diag(p)[, -(4:(p - 1))]
# "R"andom "M"ulti"V"ariate "S"tudent
X <- rmvt(n = n, sigma = diag(p), df = 3)
XB <- X %*% B
Y <- (XB[, 1]) * (XB[, 2])^2 + (XB[, 3]) * (XB[, 4])
Y <- Y + rlaplace(n, 0, sd)
} else {
stop("Got unknown dataset name.")
}
return(list(X = X, Y = Y, B = B, name = name))
}

View File

@ -5,9 +5,15 @@ directions <- function(dr, k) {
#' Computes projected training data \code{X} for given dimension `k`.
#'
#' @param dr Instance of 'cve' as returned by \code{cve}.
#' Projects the dimensional design matrix \eqn{X} on the columnspace of the
#' cve-estimate for given dimension \eqn{k}.
#'
#' @param dr Instance of \code{'cve'} as returned by \code{\link{cve}}.
#' @param k SDR dimension to use for projection.
#'
#' @return the \eqn{n\times k}{n x k} dimensional matrix \eqn{X B} where \eqn{B}
#' is the cve-estimate for dimension \eqn{k}.
#'
#' @examples
#' # create B for simulation (k = 1)
#' B <- rep(1, 5) / sqrt(5)

View File

@ -1,27 +1,22 @@
#' Bandwidth estimation for CVE.
#'
#' Estimates a bandwidth \code{h} according
#' If no bandwidth or function for calculating it is supplied, the CVE method
#' defaults to using the following formula (version 1)
#' \deqn{%
#' h = (2 * tr(\Sigma) / p) * (1.2 * n^{-1 / (4 + k)})^2}{%
#' h = (2 * tr(\Sigma) / p) * (1.2 * n^(\frac{-1}{4 + k}))^2}
#' with \eqn{n} the sample size, \eqn{p} its dimension
#' (\code{n <- nrow(X); p <- ncol(X)}) and the covariance-matrix \eqn{\Sigma}
#' which is \code{(n-1)/n} times the sample covariance estimate.
#' h = \frac{2 tr(\Sigma)}{p} (1.2 n^{\frac{-1}{4 + k}})^2}{%
#' h = (2 * tr(\Sigma) / p) * (1.2 * n^(-1 / (4 + k)))^2}
#' Alternative version 2 is used for dimension prediction which is given by
#' \deqn{%
#' h = (2 * tr(\Sigma) / p) * \chi_k^{-1}(\frac{nObs - 1}{n - 1})}{%
#' h = (2 * tr(\Sigma) / p) * \chi_k^-1((nObs - 1) / (n - 1))}
#' with \eqn{n} the sample size, \eqn{p} its dimension and the
#' covariance-matrix \eqn{\Sigma}, which is \code{(n-1)/n} times the sample
#' covariance estimate.
#'
#' @param X data matrix with samples in its rows.
#' @param X a \eqn{n\times p}{n x p} matrix with samples in its rows.
#' @param k Dimension of lower dimensional projection.
#' @param nObs number of points in a slice, see \eqn{nObs} in CVE paper.
#' @param version either \code{1} or \code{2}, where
#' \itemize{
#' \item 1: uses the following formula:
#' \deqn{%
#' h = (2 * tr(\Sigma) / p) * (1.2 * n^{-1 / (4 + k)})^2}{%
#' h = (2 * tr(\Sigma) / p) * (1.2 * n^(\frac{-1}{4 + k}))^2}
#' \item 2: uses
#' \deqn{%
#' h = (2 * tr(\Sigma) / p) * \chi_k^-1((nObs - 1) / (n - 1))}{%
#' h = (2 * tr(\Sigma) / p) * \chi_k^{-1}(\frac{nObs - 1}{n - 1})}
#' }
#' @param nObs number of points in a slice, only for version 2.
#' @param version either \code{1} or \code{2}.
#'
#' @return Estimated bandwidth \code{h}.
#'

View File

@ -1,6 +1,9 @@
#' Loss distribution elbow plot.
#'
#' Boxplots of the loss from \code{min.dim} to \code{max.dim} \code{k} values.
#' Boxplots of the output \code{L} from \code{\link{cve}} over \code{k} from
#' \code{min.dim} to \code{max.dim}. For given \code{k}, \code{L} corresponds
#' to \eqn{L_n(V, X_i)} where \eqn{V \in S(p, p - k)}{V} is the minimizer of
#' \eqn{L_n(V)}, for further details see the paper.
#'
#' @param x Object of class \code{"cve"} (result of [\code{\link{cve}}]).
#' @param ... Pass through parameters to [\code{\link{plot}}] and
@ -31,6 +34,9 @@
#' # elbow plot
#' plot(cve.obj.simple)
#'
#' @references Fertl Lukas, Bura Efstathia. (2019), Conditional Variance
#' Estimation for Sufficient Dimension Reduction. Working Paper.
#'
#' @seealso see \code{\link{par}} for graphical parameters to pass through
#' as well as \code{\link{plot}}, the standard plot utility.
#' @method plot cve

View File

@ -1,6 +1,7 @@
#' Predict method for CVE Fits.
#'
#' Predict responces using reduced data with \code{\link{mars}}.
#' Predict response using projected data where the forward model \eqn{g(B'X)}
#' is estimated using \code{\link{mars}}.
#'
#' @param object instance of class \code{cve} (result of \code{cve},
#' \code{cve.call}).
@ -36,7 +37,7 @@
#'
#' # plot prediction against y.test
#' plot(yhat, y.test)
#' @seealso \code{\link{cve}}, \code{\link{cve.call}} or \pkg{\link{mars}}.
#' @seealso \code{\link{cve}}, \code{\link{cve.call}} and \pkg{\link{mars}}.
#'
#' @rdname predict.cve
#'

View File

@ -36,10 +36,6 @@ predict_dim_elbow <- function(object) {
# Get dimensions
n <- nrow(X)
p <- ncol(X)
# Compute persistent data.
i = rep(1:n, n)
j = rep(1:n, each = n)
D.eucl = matrix((X[i, ] - X[j, ])^2 %*% rep(1, p), n)
losses <- vector("double", length(object$res))
names(losses) <- names(object$res)
@ -48,16 +44,16 @@ predict_dim_elbow <- function(object) {
# extract dimension specific estimates and dimensions.
k <- dr.k$k
V <- dr.k$V
q <- ncol(V)
# estimate bandwidth according alternative formula (see: TODO: see)
# estimate bandwidth according alternative formula.
h <- estimate.bandwidth(X, k, sqrt(n), version = 2L)
# Projected `X`
XV <- X %*% V
# Devectorized distance matrix
# (inefficient in R but fast in C)
D <- matrix((XV[i, , drop = F] - XV[j, , drop = F])^2 %*% rep(1, q), n)
D <- D.eucl - D
XQ <- X %*% (diag(1, p) - tcrossprod(V)) # X (I - V V')
# Compute distances
d2 <- tcrossprod(XQ) # XQ XQ'
d1 <- matrix(diag(d2), n, n)
D <- d1 - 2 * d2 + t(d1)
# Apply kernel
# Note: CVE uses for d = ||Q(X_i - X_j)|| the kernel exp(-d^4 / (2 h^2))
K <- exp((-0.5 / h^2) * D^2)
# sum columns
colSumsK <- colSums(K)
@ -81,10 +77,6 @@ predict_dim_wilcoxon <- function(object, p.value = 0.05) {
# Get dimensions
n <- nrow(X)
p <- ncol(X)
# Compute persistent data.
i = rep(1:n, n)
j = rep(1:n, each = n)
D.eucl = matrix((X[i, ] - X[j, ])^2 %*% rep(1, p), n)
L <- matrix(NA, n, length(object$res))
colnames(L) <- names(object$res)
@ -93,16 +85,16 @@ predict_dim_wilcoxon <- function(object, p.value = 0.05) {
# extract dimension specific estimates and dimensions.
k <- dr.k$k
V <- dr.k$V
q <- ncol(V)
# estimate bandwidth according alternative formula (see: TODO: see)
# estimate bandwidth according alternative formula.
h <- estimate.bandwidth(X, k, sqrt(n), version = 2L)
# Projected `X`
XV <- X %*% V
# Devectorized distance matrix
# (inefficient in R but fast in C)
D <- matrix((XV[i, , drop = F] - XV[j, , drop = F])^2 %*% rep(1, q), n)
D <- D.eucl - D
XQ <- X %*% (diag(1, p) - tcrossprod(V)) # X (I - V V')
# Compute distances
d2 <- tcrossprod(XQ) # XQ XQ'
d1 <- matrix(diag(d2), n, n)
D <- d1 - 2 * d2 + t(d1)
# Apply kernel
# Note: CVE uses for d = ||Q(X_i - X_j)|| the kernel exp(-d^4 / (2 h^2))
K <- exp((-0.5 / h^2) * D^2)
# sum columns
colSumsK <- colSums(K)
@ -130,27 +122,24 @@ predict_dim_wilcoxon <- function(object, p.value = 0.05) {
))
}
#' Predicts SDR dimension using \code{\link[mda]{mars}} via a Cross-Validation.
#' TODO: rewrite!!!
#' \code{"TODO: @Lukas"}
#'
#' @param object instance of class \code{cve} (result of \code{cve},
#' \code{cve.call}).
#' @param object instance of class \code{cve} (result of \code{\link{cve}},
#' \code{\link{cve.call}}).
#' @param method one of \code{"CV"}, \code{"elbow"} or \code{"wilcoxon"}.
#' @param ... ignored.
#'
#' @return list with
#' \itemize{
#' \item MSE: Mean Square Error,
#' \item k: predicted dimensions.
#' }
#' @return list with \code{"k"} the predicted dimension and method dependent
#' informatoin.
#'
#' @section cv:
#' Cross-validation ... TODO:
#' @section Method cv:
#' TODO: \code{"TODO: @Lukas"}.
#'
#' @section elbow:
#' Cross-validation ... TODO:
#' @section Method elbow:
#' TODO: \code{"TODO: @Lukas"}.
#'
#' @section wilcoxon:
#' Cross-validation ... TODO:
#' @section Method wilcoxon:
#' TODO: \code{"TODO: @Lukas"}.
#'
#' @examples
#' # create B for simulation

View File

@ -1,5 +1,9 @@
#' Prints a summary of a \code{cve} result.
#' @param object Instance of 'cve' as returned by \code{cve}.
#'
#' Prints a summary statistics of output \code{L} from \code{cve} for
#' \code{k = min.dim, ..., max.dim}.
#'
#' @param object Instance of \code{"cve"} as returned by \code{\link{cve}}.
#' @param ... ignored.
#'
#' @examples

24
CVE/R/util.R Normal file
View File

@ -0,0 +1,24 @@
#' Draws a sample from the invariant measure on the Stiefel manifold
#' \eqn{S(p, q)}.
#'
#' @param p row dimension
#' @param q col dimension
#' @return \eqn{p \times q}{p x q} semi-orthogonal matrix.
#' @examples
#' V <- rStiefel(6, 4)
#' @export
rStiefel <- function(p, q) {
return(qr.Q(qr(matrix(rnorm(p * q, 0, 1), p, q))))
}
#' Null space basis of given matrix `V`
#'
#' @param V `(p, q)` matrix
#' @return Semi-orthogonal `(p, p - q)` matrix spaning the null space of `V`.
#' @keywords internal
#' @export
null <- function(V) {
tmp <- qr(V)
set <- if(tmp$rank == 0L) seq_len(ncol(V)) else -seq_len(tmp$rank)
return(qr.Q(tmp, complete = TRUE)[, set, drop = FALSE])
}

View File

@ -26,7 +26,7 @@ variance-covariance matrix \eqn{Var(X) = \Sigma_X}. \eqn{\epsilon} is a mean
zero random variable with finite \eqn{Var(\epsilon) = E(\epsilon^2)}, \eqn{g}
is an unknown, continuous non-constant function,
and \eqn{B = (b_1, ..., b_k)} is
a real \eqn{p \times k}{p x k} of rank \eqn{k <= p}{k \leq p}.
a real \eqn{p \times k}{p x k} of rank \eqn{k \leq p}{k <= p}.
Without loss of generality \eqn{B} is assumed to be orthonormal.
}
\references{

View File

@ -8,7 +8,7 @@
}
\arguments{
\item{object}{instance of \code{cve} as output from \code{\link{cve}} or
\code{\link{cve.call}}}
\code{\link{cve.call}}.}
\item{k}{the SDR dimension.}
@ -18,7 +18,8 @@
dir the matrix of CS or CMS of given dimension
}
\description{
Returns the SDR basis matrix for SDR dimension(s).
Returns the SDR basis matrix for dimension \code{k}, i.e. returns the
cve-estimate with dimension \eqn{p\times k}{p x k}.
}
\examples{
# set dimensions for simulation model

View File

@ -8,31 +8,51 @@ cve(formula, data, method = "simple", max.dim = 10L, ...)
}
\arguments{
\item{formula}{an object of class \code{"formula"} which is a symbolic
description of the model to be fitted.}
description of the model to be fitted like \eqn{Y\sim X}{Y ~ X} where
\eqn{Y} is a \eqn{n}-dimensional vector of the response variable and
\eqn{X} is a \eqn{n\times p}{n x p} matrix of the predictors.}
\item{data}{an optional data frame, containing the data for the formula if
supplied.}
supplied like \code{data <- data.frame(Y, X)} with dimension
\eqn{n \times (p + 1)}{n x (p + 1)}. By default the variables are taken from
the environment from which \code{cve} is called.}
\item{method}{specifies the CVE method variation as one of
\item{method}{This character string specifies the method of fitting. The
options are
\itemize{
\item "simple" exact implementation as described in the paper listed
below.
\item "weighted" variation with addaptive weighting of slices.
}}
\item "simple" implementation as described in the paper.
\item "weighted" variation with adaptive weighting of slices.
}
see paper.}
\item{max.dim}{upper bounds for \code{k}, (ignored if \code{k} is supplied).}
\item{...}{Parameters passed on to \code{cve.call}.}
\item{...}{optional parameters passed on to \code{cve.call}.}
}
\value{
an S3 object of class \code{cve} with components:
\describe{
\item{X}{Original training data,}
\item{Y}{Responce of original training data,}
\item{X}{design matrix of predictor vector used for calculating
cve-estimate,}
\item{Y}{\eqn{n}-dimensional vector of responses used for calculating
cve-estimate,}
\item{method}{Name of used method,}
\item{call}{the matched call,}
\item{res}{list of components \code{V, L, B, loss, h} and \code{k} for
each \eqn{k=min.dim,...,max.dim} (dimension).}
\item{res}{list of components \code{V, L, B, loss, h} for
each \code{k = min.dim, ..., max.dim}. If \code{k} was supplied in the
call \code{min.dim = max.dim = k}.
\itemize{
\item \code{B} is the cve-estimate with dimension
\eqn{p\times k}{p x k}.
\item \code{V} is the orthogonal complement of \eqn{B}.
\item \code{L} is the loss for each sample seperatels such that
it's mean is \code{loss}.
\item \code{loss} is the value of the target function that is
minimized, evaluated at \eqn{V}.
\item \code{h} bandwidth parameter used to calculate
\code{B, V, loss, L}.
}
}
}
}
\description{
@ -56,7 +76,7 @@ variance-covariance matrix \eqn{Var(X) = \Sigma_X}. \eqn{\epsilon} is a mean
zero random variable with finite \eqn{Var(\epsilon) = E(\epsilon^2)}, \eqn{g}
is an unknown, continuous non-constant function,
and \eqn{B = (b_1, ..., b_k)} is
a real \eqn{p \times k}{p x k} of rank \eqn{k <= p}{k \leq p}.
a real \eqn{p \times k}{p x k} of rank \eqn{k \leq p}{k <= p}.
Without loss of generality \eqn{B} is assumed to be orthonormal.
}
\examples{
@ -67,7 +87,7 @@ k <- 2
b1 <- rep(1 / sqrt(p), p)
b2 <- (-1)^seq(1, p) / sqrt(p)
B <- cbind(b1, b2)
# samplsize
# sample size
n <- 200
set.seed(21)
# creat predictor data x ~ N(0, I_p)

View File

@ -10,9 +10,9 @@ cve.call(X, Y, method = "simple", nObs = sqrt(nrow(X)), h = NULL,
max.iter = 50L, attempts = 10L, logger = NULL)
}
\arguments{
\item{X}{data matrix with samples in its rows.}
\item{X}{Design matrix with dimension \eqn{n\times p}{n x p}.}
\item{Y}{Responses (1 dimensional).}
\item{Y}{numeric array of length \eqn{n} of Responses.}
\item{method}{specifies the CVE method variation as one of
\itemize{
@ -34,38 +34,59 @@ estimated bandwidth.}
\item{k}{Dimension of lower dimensional projection, if \code{k} is given
only the specified dimension \code{B} matrix is estimated.}
\item{momentum}{number of [0, 1) giving the ration of momentum for eucledian
gradient update with a momentum term.}
\item{momentum}{number of \eqn{[0, 1)} giving the ration of momentum for
eucledian gradient update with a momentum term. \code{momentum = 0}
corresponds to normal gradient descend.}
\item{tau}{Initial step-size.}
\item{tol}{Tolerance for break condition.}
\item{slack}{Positive scaling to allow small increases of the loss while
optimizing.}
optimizing, i.e. \code{slack = 0.1} allows the target function to
increase up to \eqn{10 \%} in one optimization step.}
\item{gamma}{step-size reduction multiple.}
\item{gamma}{step-size reduction multiple. If gradient step with step size
\code{tau} is not accepted \code{gamma * tau} is set to the next step
size.}
\item{V.init}{Semi-orthogonal matrix of dimensions `(ncol(X), ncol(X) - k)
as optimization starting value. (If supplied, \code{attempts} is
set to 1 and \code{k} to match dimension)}
used as starting value in the optimization. (If supplied,
\code{attempts} is set to 0 and \code{k} to match dimension).}
\item{max.iter}{maximum number of optimization steps.}
\item{attempts}{number of arbitrary different starting points.}
\item{attempts}{If \code{V.init} not supplied, the optimization is carried
out \code{attempts} times with starting values drawn from the invariant
measure on the Stiefel manifold (see \code{\link{rStiefel}}).}
\item{logger}{a logger function (only for advanced user, significantly slows
down the computation).}
\item{logger}{a logger function (only for advanced user, slows down the
computation).}
}
\value{
an S3 object of class \code{cve} with components:
\describe{
\item{X}{Original training data,}
\item{Y}{Responce of original training data,}
\item{X}{design matrix of predictor vector used for calculating
cve-estimate,}
\item{Y}{\eqn{n}-dimensional vector of responses used for calculating
cve-estimate,}
\item{method}{Name of used method,}
\item{call}{the matched call,}
\item{res}{list of components \code{V, L, B, loss, h} and \code{k} for
each \eqn{k=min.dim,...,max.dim} (dimension).}
\item{res}{list of components \code{V, L, B, loss, h} for
each \code{k = min.dim, ..., max.dim}. If \code{k} was supplied in the
call \code{min.dim = max.dim = k}.
\itemize{
\item \code{B} is the cve-estimate with dimension
\eqn{p\times k}{p x k}.
\item \code{V} is the orthogonal complement of \eqn{B}.
\item \code{L} is the loss for each sample seperatels such that
it's mean is \code{loss}.
\item \code{loss} is the value of the target function that is
minimized, evaluated at \eqn{V}.
\item \code{h} bandwidth parameter used to calculate
\code{B, V, loss, L}.
}
}
}
}
\description{
@ -89,7 +110,7 @@ variance-covariance matrix \eqn{Var(X) = \Sigma_X}. \eqn{\epsilon} is a mean
zero random variable with finite \eqn{Var(\epsilon) = E(\epsilon^2)}, \eqn{g}
is an unknown, continuous non-constant function,
and \eqn{B = (b_1, ..., b_k)} is
a real \eqn{p \times k}{p x k} of rank \eqn{k <= p}{k \leq p}.
a real \eqn{p \times k}{p x k} of rank \eqn{k \leq p}{k <= p}.
Without loss of generality \eqn{B} is assumed to be orthonormal.
}
\examples{

127
CVE/man/dataset.Rd Normal file
View File

@ -0,0 +1,127 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/datasets.R
\name{dataset}
\alias{dataset}
\title{Generates test datasets.}
\usage{
dataset(name = "M1", n = NULL, p = 20, sd = 0.5, ...)
}
\arguments{
\item{name}{One of \code{"M1"}, \code{"M2"}, \code{"M3"}, \code{"M4",}
\code{"M5"}, \code{"M6"} or \code{"M7"}. Alternative just the dataset number
1-7.}
\item{n}{number of samples.}
\item{p}{Dimension of random variable \eqn{X}.}
\item{sd}{standard diviation for error term \eqn{\epsilon}.}
\item{...}{Additional parameters only for "M2" (namely \code{pmix} and
\code{lambda}), see: below.}
}
\value{
List with elements
\itemize{
\item{X}{data, a \eqn{n\times p}{n x p} matrix.}
\item{Y}{response.}
\item{B}{the dim-reduction matrix}
\item{name}{Name of the dataset (name parameter)}
}
}
\description{
Provides sample datasets M1-M7 used in the paper Conditional variance
estimation for sufficient dimension reduction, Lukas Fertl, Efstathia Bura.
The general model is given by:
\deqn{Y = g(B'X) + \epsilon}
}
\section{M1}{
The predictors are distributed as
\eqn{X\sim N_p(0, \Sigma)}{X ~ N_p(0, \Sigma)} with
\eqn{\Sigma_{i, j} = 0.5^{|i - j|}}{\Sigma_ij = 0.5^|i - j|} for
\eqn{i, j = 1,..., p} for a subspace dimension of \eqn{k = 1} with a default
of \eqn{n = 100} data points. \eqn{p = 20},
\eqn{b_1 = (1,1,1,1,1,1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_1 = (1,1,1,1,1,1,0,...,0)' / sqrt(6)}, and \eqn{Y} is
given as \deqn{Y = cos(b_1'X) + \epsilon} where \eqn{\epsilon} is
distributed as generalized normal distribution with location 0,
shape-parameter 0.5, and the scale-parameter is chosen such that
\eqn{Var(\epsilon) = 0.5}.
}
\section{M2}{
The predictors are distributed as \eqn{X \sim Z 1_p \lambda + N_p(0, I_p)}{X ~ Z 1_p \lambda + N_p(0, I_p)}. with
\eqn{Z \sim 2 Binom(p_{mix}) - 1\in\{-1, 1\}}{Z~2Binom(pmix)-1} where
\eqn{1_p} is the \eqn{p}-dimensional vector of one's, for a subspace
dimension of \eqn{k = 1} with a default of \eqn{n = 100} data points.
\eqn{p = 20}, \eqn{b_1 = (1,1,1,1,1,1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_1 = (1,1,1,1,1,1,0,...,0)' / sqrt(6)},
and \eqn{Y} is \deqn{Y = cos(b_1'X) + 0.5\epsilon} where \eqn{\epsilon} is
standard normal.
Defaults for \code{pmix} is 0.3 and \code{lambda} defaults to 1.
}
\section{M3}{
The predictors are distributed as \eqn{X\sim N_p(0, I_p)}{X~N_p(0, I_p)}
for a subspace
dimension of \eqn{k = 1} with a default of \eqn{n = 100} data points.
\eqn{p = 20}, \eqn{b_1 = (1,1,1,1,1,1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_1 = (1,1,1,1,1,1,0,...,0)' / sqrt(6)},
and \eqn{Y} is
\deqn{Y = 2 log(|b_1'X| + 2) + 0.5\epsilon} where \eqn{\epsilon} is
standard normal.
}
\section{M4}{
The predictors are distributed as \eqn{X\sim N_p(0,\Sigma)}{X~N_p(0,\Sigma)}
with \eqn{\Sigma_{i, j} = 0.5^{|i - j|}}{\Sigma_ij = 0.5^|i - j|} for
\eqn{i, j = 1,..., p} for a subspace dimension of \eqn{k = 2} with a default
of \eqn{n = 100} data points. \eqn{p = 20},
\eqn{b_1 = (1,1,1,1,1,1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_1 = (1,1,1,1,1,1,0,...,0)' / sqrt(6)},
\eqn{b_2 = (1,-1,1,-1,1,-1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_2 = (1,-1,1,-1,1,-1,0,...,0)' / sqrt(6)}
and \eqn{Y} is given as \deqn{Y = \frac{b_1'X}{0.5 + (1.5 + b_2'X)^2} + 0.5\epsilon}{Y = (b_1'X) / (0.5 + (1.5 + b_2'X)^2) + 0.5\epsilon}
where \eqn{\epsilon} is standard normal.
}
\section{M5}{
The predictors are distributed as \eqn{X\sim U([0,1]^p)}{X~U([0, 1]^p)}
where \eqn{U([0, 1]^p)} is the uniform distribution with
independent components on the \eqn{p}-dimensional hypercube for a subspace
dimension of \eqn{k = 2} with a default of \eqn{n = 200} data points.
\eqn{p = 20},
\eqn{b_1 = (1,1,1,1,1,1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_1 = (1,1,1,1,1,1,0,...,0)' / sqrt(6)},
\eqn{b_2 = (1,-1,1,-1,1,-1,0,...,0)' / \sqrt{6}\in\mathcal{R}^p}{b_2 = (1,-1,1,-1,1,-1,0,...,0)' / sqrt(6)}
and \eqn{Y} is given as \deqn{Y = cos(\pi b_1'X)(b_2'X + 1)^2 + 0.5\epsilon}
where \eqn{\epsilon} is standard normal.
}
\section{M6}{
The predictors are distributed as \eqn{X\sim N_p(0, I_p)}{X~N_p(0, I_p)}
for a subspace dimension of \eqn{k = 3} with a default of \eqn{n = 200} data
point. \eqn{p = 20, b_1 = e_1, b_2 = e_2}, and \eqn{b_3 = e_p}, where
\eqn{e_j} is the \eqn{j}-th unit vector in the \eqn{p}-dimensional space.
\eqn{Y} is given as \deqn{Y = (b_1'X)^2+(b_2'X)^2+(b_3'X)^2+0.5\epsilon}
where \eqn{\epsilon} is standard normal.
}
\section{M7}{
The predictors are distributed as \eqn{X\sim t_3(I_p)}{X~t_3(I_p)} where
\eqn{t_3(I_p)} is the standard multivariate t-distribution with 3 degrees of
freedom, for a subspace dimension of \eqn{k = 4} with a default of
\eqn{n = 200} data points.
\eqn{p = 20, b_1 = e_1, b_2 = e_2, b_3 = e_3}, and \eqn{b_4 = e_p}, where
\eqn{e_j} is the \eqn{j}-th unit vector in the \eqn{p}-dimensional space.
\eqn{Y} is given as \deqn{Y = (b_1'X)(b_2'X)^2+(b_3'X)(b_4'X)+0.5\epsilon}
where \eqn{\epsilon} is distributed as generalized normal distribution with
location 0, shape-parameter 1, and the scale-parameter is chosen such that
\eqn{Var(\epsilon) = 0.25}.
}
\references{
Fertl Lukas, Bura Efstathia. (2019), Conditional Variance
Estimation for Sufficient Dimension Reduction. Working Paper.
}

View File

@ -8,12 +8,17 @@
\method{directions}{cve}(dr, k)
}
\arguments{
\item{dr}{Instance of 'cve' as returned by \code{cve}.}
\item{dr}{Instance of \code{'cve'} as returned by \code{\link{cve}}.}
\item{k}{SDR dimension to use for projection.}
}
\value{
the \eqn{n\times k}{n x k} dimensional matrix \eqn{X B} where \eqn{B}
is the cve-estimate for dimension \eqn{k}.
}
\description{
Computes projected training data \code{X} for given dimension `k`.
Projects the dimensional design matrix \eqn{X} on the columnspace of the
cve-estimate for given dimension \eqn{k}.
}
\examples{
# create B for simulation (k = 1)

View File

@ -4,26 +4,33 @@
\alias{estimate.bandwidth}
\title{Bandwidth estimation for CVE.}
\usage{
estimate.bandwidth(X, k, nObs)
estimate.bandwidth(X, k, nObs, version = 1L)
}
\arguments{
\item{X}{data matrix with samples in its rows.}
\item{X}{a \eqn{n\times p}{n x p} matrix with samples in its rows.}
\item{k}{Dimension of lower dimensional projection.}
\item{nObs}{number of points in a slice, see \eqn{nObs} in CVE paper.}
\item{nObs}{number of points in a slice, only for version 2.}
\item{version}{either \code{1} or \code{2}.}
}
\value{
Estimated bandwidth \code{h}.
}
\description{
Estimates a bandwidth \code{h} according
If no bandwidth or function for calculating it is supplied, the CVE method
defaults to using the following formula (version 1)
\deqn{%
h = (2 * tr(\Sigma) / p) * (1.2 * n^{-1 / (4 + k)})^2}{%
h = (2 * tr(\Sigma) / p) * (1.2 * n^(\frac{-1}{4 + k}))^2}
with \eqn{n} the sample size, \eqn{p} its dimension
(\code{n <- nrow(X); p <- ncol(X)}) and the covariance-matrix \eqn{\Sigma}
which is \code{(n-1)/n} times the sample covariance estimate.
h = \frac{2 tr(\Sigma)}{p} (1.2 n^{\frac{-1}{4 + k}})^2}{%
h = (2 * tr(\Sigma) / p) * (1.2 * n^(-1 / (4 + k)))^2}
Alternative version 2 is used for dimension prediction which is given by
\deqn{%
h = (2 * tr(\Sigma) / p) * \chi_k^{-1}(\frac{nObs - 1}{n - 1})}{%
h = (2 * tr(\Sigma) / p) * \chi_k^-1((nObs - 1) / (n - 1))}
with \eqn{n} the sample size, \eqn{p} its dimension and the
covariance-matrix \eqn{\Sigma}, which is \code{(n-1)/n} times the sample
covariance estimate.
}
\examples{
# set dimensions for simulation model

View File

@ -13,7 +13,10 @@
[\code{\link{lines}}]}
}
\description{
Boxplots of the loss from \code{min.dim} to \code{max.dim} \code{k} values.
Boxplots of the output \code{L} from \code{\link{cve}} over \code{k} from
\code{min.dim} to \code{max.dim}. For given \code{k}, \code{L} corresponds
to \eqn{L_n(V, X_i)} where \eqn{V \in S(p, p - k)}{V} is the minimizer of
\eqn{L_n(V)}, for further details see the paper.
}
\examples{
# create B for simulation
@ -41,6 +44,10 @@ cve.obj.simple <- cve(Y ~ X, h = estimate.bandwidth, nObs = sqrt(nrow(X)))
# elbow plot
plot(cve.obj.simple)
}
\references{
Fertl Lukas, Bura Efstathia. (2019), Conditional Variance
Estimation for Sufficient Dimension Reduction. Working Paper.
}
\seealso{
see \code{\link{par}} for graphical parameters to pass through

View File

@ -20,7 +20,8 @@
prediced response of data \code{newdata}.
}
\description{
Predict responces using reduced data with \code{\link{mars}}.
Predict response using projected data where the forward model \eqn{g(B'X)}
is estimated using \code{\link{mars}}.
}
\examples{
# create B for simulation
@ -50,5 +51,5 @@ yhat <- predict(cve.obj.simple, x.test, 1)
plot(yhat, y.test)
}
\seealso{
\code{\link{cve}}, \code{\link{cve.call}} or \pkg{\link{mars}}.
\code{\link{cve}}, \code{\link{cve.call}} and \pkg{\link{mars}}.
}

View File

@ -2,26 +2,40 @@
% Please edit documentation in R/predict_dim.R
\name{predict_dim}
\alias{predict_dim}
\title{Predicts SDR dimension using \code{\link[mda]{mars}} via a Cross-Validation.}
\title{\code{"TODO: @Lukas"}}
\usage{
predict_dim(object, ...)
predict_dim(object, ..., method = "CV")
}
\arguments{
\item{object}{instance of class \code{cve} (result of \code{cve},
\code{cve.call}).}
\item{object}{instance of class \code{cve} (result of \code{\link{cve}},
\code{\link{cve.call}}).}
\item{...}{ignored.}
\item{method}{one of \code{"CV"}, \code{"elbow"} or \code{"wilcoxon"}.}
}
\value{
list with
\itemize{
\item MSE: Mean Square Error,
\item k: predicted dimensions.
}
list with \code{"k"} the predicted dimension and method dependent
informatoin.
}
\description{
Predicts SDR dimension using \code{\link[mda]{mars}} via a Cross-Validation.
\code{"TODO: @Lukas"}
}
\section{Method cv}{
TODO: \code{"TODO: @Lukas"}.
}
\section{Method elbow}{
TODO: \code{"TODO: @Lukas"}.
}
\section{Method wilcoxon}{
TODO: \code{"TODO: @Lukas"}.
}
\examples{
# create B for simulation
B <- rep(1, 5) / sqrt(5)

View File

@ -2,7 +2,8 @@
% Please edit documentation in R/util.R
\name{rStiefel}
\alias{rStiefel}
\title{Draws a sample from the invariant measure on the Stiefel manifold \eqn{S(p, q)}.}
\title{Draws a sample from the invariant measure on the Stiefel manifold
\eqn{S(p, q)}.}
\usage{
rStiefel(p, q)
}
@ -12,10 +13,11 @@ rStiefel(p, q)
\item{q}{col dimension}
}
\value{
\code{p} times \code{q} semi-orthogonal matrix.
\eqn{p \times q}{p x q} semi-orthogonal matrix.
}
\description{
Draws a sample from the invariant measure on the Stiefel manifold \eqn{S(p, q)}.
Draws a sample from the invariant measure on the Stiefel manifold
\eqn{S(p, q)}.
}
\examples{
V <- rStiefel(6, 4)

27
CVE/man/rgnorm.Rd Normal file
View File

@ -0,0 +1,27 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/datasets.R
\name{rgnorm}
\alias{rgnorm}
\title{Generalized Normal Distribution.}
\usage{
rgnorm(n = 1, mu = 0, alpha = 1, beta = 1)
}
\arguments{
\item{n}{Number of generated samples.}
\item{mu}{mean.}
\item{alpha}{first shape parameter.}
\item{beta}{second shape parameter.}
}
\value{
numeric array of length \eqn{n}.
}
\description{
Random generation for generalized Normal Distribution.
}
\seealso{
https://en.wikipedia.org/wiki/Generalized_normal_distribution
}
\keyword{internal}

25
CVE/man/rlaplace.Rd Normal file
View File

@ -0,0 +1,25 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/datasets.R
\name{rlaplace}
\alias{rlaplace}
\title{Laplace distribution}
\usage{
rlaplace(n = 1, mu = 0, sd = 1)
}
\arguments{
\item{n}{Number of generated samples.}
\item{mu}{mean.}
\item{sd}{standard deviation.}
}
\value{
numeric array of length \eqn{n}.
}
\description{
Random generation for Laplace distribution.
}
\seealso{
https://en.wikipedia.org/wiki/Laplace_distribution
}
\keyword{internal}

29
CVE/man/rmvnorm.Rd Normal file
View File

@ -0,0 +1,29 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/datasets.R
\name{rmvnorm}
\alias{rmvnorm}
\title{Multivariate Normal Distribution.}
\usage{
rmvnorm(n = 1, mu = rep(0, p), sigma = diag(p))
}
\arguments{
\item{n}{number of samples.}
\item{mu}{mean}
\item{sigma}{covariance matrix.}
}
\value{
a \eqn{n\times p}{n x p} matrix with samples in its rows.
}
\description{
Random generation for the multivariate normal distribution.
\deqn{X \sim N_p(\mu, \Sigma)}{X ~ N_p(\mu, \Sigma)}
}
\examples{
\dontrun{
rmvnorm(20, sigma = matrix(c(2, 1, 1, 2), 2))
rmvnorm(20, mu = c(3, -1, 2))
}
}
\keyword{internal}

34
CVE/man/rmvt.Rd Normal file
View File

@ -0,0 +1,34 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/datasets.R
\name{rmvt}
\alias{rmvt}
\title{Multivariate t distribution.}
\usage{
rmvt(n = 1, mu = rep(0, p), sigma = diag(p), df = Inf)
}
\arguments{
\item{n}{number of samples.}
\item{mu}{mean}
\item{sigma}{a \eqn{k\times k}{k x k} positive definite matrix. If the degree
\eqn{\nu} if bigger than 2 the created covariance is
\deqn{var(x) = \Sigma\frac{\nu}{\nu - 2}}
for \eqn{\nu > 2}.}
\item{df}{degree of freedom \eqn{\nu}.}
}
\value{
a \eqn{n\times p}{n x p} matrix with samples in its rows.
}
\description{
Random generation from multivariate t distribution (student distribution).
}
\examples{
\dontrun{
rmvt(20, c(0, 1), matrix(c(3, 1, 1, 2), 2), 3)
rmvt(20, sigma = matrix(c(2, 1, 1, 2), 2), 3)
rmvt(20, mu = c(3, -1, 2), 3)
}
}
\keyword{internal}

View File

@ -7,12 +7,13 @@
\method{summary}{cve}(object, ...)
}
\arguments{
\item{object}{Instance of 'cve' as returned by \code{cve}.}
\item{object}{Instance of \code{"cve"} as returned by \code{\link{cve}}.}
\item{...}{ignored.}
}
\description{
Prints a summary of a \code{cve} result.
Prints a summary statistics of output \code{L} from \code{cve} for
\code{k = min.dim, ..., max.dim}.
}
\examples{
# create B for simulation

View File

@ -161,97 +161,3 @@ mat* adjacence(const mat *vec_L, const mat *vec_Y, const mat *vec_y1,
return mat_S;
}
// int getWorkLen(const int n, const int p, const int q) {
// int mpq; /**< Max of p and q */
// int nn = ((n - 1) * n) / 2;
// if (p > q) {
// mpq = p;
// } else {
// mpq = q;
// }
// if (nn * p < (mpq + 1) * mpq) {
// return 2 * (mpq + 1) * mpq;
// } else {
// return (nn + mpq) * mpq;
// }
// }
// double cost(const unsigned int method,
// const int n,
// const double *Y,
// const double *vecK,
// const double *colSums,
// double *y1, double *L) {
// int i, j, k;
// double tmp, sum;
// for (i = 0; i < n; ++i) {
// y1[i] = Y[i];
// L[i] = Y[i] * Y[i];
// }
// for (k = j = 0; j < n; ++j) {
// for (i = j + 1; i < n; ++i, ++k) {
// y1[i] += Y[j] * vecK[k];
// y1[j] += Y[i] * vecK[k];
// L[i] += Y[j] * Y[j] * vecK[k];
// L[j] += Y[i] * Y[i] * vecK[k];
// }
// }
// for (i = 0; i < n; ++i) {
// y1[i] /= colSums[i];
// L[i] /= colSums[i];
// }
// tmp = 0.0;
// if (method == CVE_METHOD_WEIGHTED) {
// sum = 0.0;
// for (i = 0; i < n; ++i) {
// tmp += (colSums[i] - 1.0) * (L[i] -= y1[i] * y1[i]);
// sum += colSums[i];
// }
// return tmp / (sum - (double)n); // TODO: check for division by zero!
// } else {
// for (i = 0; i < n; ++i) {
// tmp += (L[i] -= y1[i] * y1[i]);
// }
// return tmp / (double)n;
// }
// }
// void scaling(const unsigned int method,
// const int n,
// const double *Y, const double *y1, const double *L,
// const double *vecD, const double *vecK,
// const double *colSums,
// double *vecS) {
// int i, j, k, nn = (n * (n - 1)) / 2;
// double tmp;
// if (method == CVE_METHOD_WEIGHTED) {
// for (k = j = 0; j < n; ++j) {
// for (i = j + 1; i < n; ++i, ++k) {
// tmp = Y[j] - y1[i];
// vecS[k] = (L[i] - (tmp * tmp));
// tmp = Y[i] - y1[j];
// vecS[k] += (L[j] - (tmp * tmp));
// }
// }
// } else {
// for (k = j = 0; j < n; ++j) {
// for (i = j + 1; i < n; ++i, ++k) {
// tmp = Y[j] - y1[i];
// vecS[k] = (L[i] - (tmp * tmp)) / colSums[i];
// tmp = Y[i] - y1[j];
// vecS[k] += (L[j] - (tmp * tmp)) / colSums[j];
// }
// }
// }
// for (k = 0; k < nn; ++k) {
// vecS[k] *= vecK[k] * vecD[k];
// }
// }

View File

@ -17,8 +17,8 @@ static const R_CallMethodDef CallEntries[] = {
{NULL, NULL, 0}
};
/* Restrict C entrypoints to registered routines. */
void R_initCVE(DllInfo *dll) {
/* Restrict C entry points to registered routines. */
void R_init_CVE(DllInfo *dll) {
R_registerRoutines(dll, NULL, CallEntries, NULL, NULL);
R_useDynamicSymbols(dll, FALSE);
}

View File

@ -914,7 +914,7 @@ mat* laplace(mat *A, double *workMem) {
* \_____/ \_____/
* IpA C = ImA B
* \_______/
* IpA C = Y ==> C = IpA^-1 Y
* IpA C = Y ==> C = IpA^-1 Y
*
* @param A Skew-Symmetric matrix of dimension `(n, n)`.
* @param B Matrix of dimensions `(n, m)` with `m <= n`.

View File

@ -1,193 +0,0 @@
#'
#' @param n number of samples.
#' @param mu mean
#' @param sigma covariance matrix.
#'
#' @returns a \eqn{n\times p} matrix with samples in its rows.
#'
#' @examples
#' rmvnorm(20, sigma = matrix(c(2, 1, 1, 2), 2))
#' rmvnorm(20, mu = c(3, -1, 2))
rmvnorm <- function(n = 1, mu = rep(0, p), sigma = diag(p)) {
if (!missing(sigma)) {
p <- nrow(sigma)
} else if (!missing(mu)) {
mu <- matrix(mu, ncol = 1)
p <- nrow(mu)
} else {
stop("At least one of 'mu' or 'sigma' must be supplied.")
}
# See: https://en.wikipedia.org/wiki/Multivariate_normal_distribution
return(rep(mu, each = n) + matrix(rnorm(n * p), n) %*% chol(sigma))
}
#' Samples from the multivariate t distribution (student distribution).
#'
#' @param n number of samples.
#' @param mu mean, ... TODO:
#' @param sigma a \eqn{k\times k} positive definite matrix. If the degree
#' \eqn{\nu} if bigger than 2 the created covariance is
#' \deqn{var(x) = \Sigma\frac{\nu}{\nu - 2}}
#' for \eqn{\nu > 2}.
#' @param df degree of freedom \eqn{\nu}.
#'
#' @returns a \eqn{n\times p} matrix with samples in its rows.
#'
#' @examples
#' rmvt(20, c(0, 1), matrix(c(3, 1, 1, 2), 2), 3)
#' rmvt(20, sigma = matrix(c(2, 1, 1, 2), 2), 3)
#' rmvt(20, mu = c(3, -1, 2), 3)
rmvt <- function(n = 1, mu = rep(0, p), sigma = diag(p), df = Inf) {
if (!missing(sigma)) {
p <- nrow(sigma)
} else if (!missing(mu)) {
mu <- matrix(mu, ncol = 1)
p <- nrow(mu)
} else {
stop("At least one of 'mu' or 'sigma' must be supplied.")
}
if (df == Inf) {
Z <- 1
} else {
Z <- sqrt(df / rchisq(n, df))
}
return(rmvnorm(n, sigma = sigma) * Z + rep(mu, each = n))
}
#' Generalized Normal Distribution.
#' see: https://en.wikipedia.org/wiki/Generalized_normal_distribution
rgnorm <- function(n = 1, mu = 0, alpha = 1, beta = 1) {
if (alpha <= 0 | beta <= 0) {
stop("alpha and beta must be positive.")
}
lambda <- (1 / alpha)^beta
scales <- qgamma(runif(n), shape = 1 / beta, scale = 1 / lambda)^(1 / beta)
return(scales * ((-1)^rbinom(n, 1, 0.5)) + mu)
}
#' Laplace distribution
#' see: https://en.wikipedia.org/wiki/Laplace_distribution
rlaplace <- function(n = 1, mu = 0, sigma = 1) {
U <- runif(n, -0.5, 0.5)
scale <- sigma / sqrt(2)
return(mu - scale * sign(U) * log(1 - 2 * abs(U)))
}
#' Generates test datasets.
#'
#' Provides sample datasets. There are 5 different datasets named
#' M1, M2, M3, M4 and M5 described in the paper references below.
#' The general model is given by:
#' \deqn{Y = g(B'X) + \epsilon}
#'
#' @param name One of \code{"M1"}, \code{"M2"}, \code{"M3"}, \code{"M4"} or \code{"M5"}
#' @param n nr samples
#' @param B SDR basis used for dataset creation if supplied.
#' @param p Dim. of random variable \code{X}.
#' @param p.mix Only for \code{"M4"}, see: below.
#' @param lambda Only for \code{"M4"}, see: below.
#'
#' @return List with elements
#' \itemize{
#' \item{X}{data}
#' \item{Y}{response}
#' \item{B}{Used dim-reduction matrix}
#' \item{name}{Name of the dataset (name parameter)}
#' }
#'
#' @section M1:
#' The data follows \eqn{X\sim N_p(0, \Sigma)}{X ~ N_p(0, Sigma)} for a subspace
#' dimension of \eqn{k = 2} with a default of \eqn{n = 200} data points.
#' The link function \eqn{g} is given as
#' \deqn{g(x) = \frac{x_1}{0.5 + (x_2 + 1.5)^2} + \epsilon / 2}{%
#' g(x) = x_1 / (0.5 + (x_2 + 1.5)^2) + epsilon / 2}
#' @section M2:
#' \eqn{X\sim N_p(0, \Sigma)}{X ~ N_p(0, Sigma)} with \eqn{k = 2} with a
#' default of \eqn{n = 200} data points.
#' The link function \eqn{g} is given as
#' \deqn{g(x) = (b_1^T X) (b_2^T X)^2 + \epsilon / 2}
#' @section M3:
#' \deqn{g(x) = cos(b_1^T X) + \epsilon / 2}
#' @section M4:
#' TODO:
#' @section M5:
#' TODO:
#'
#' @import stats
#' @importFrom stats rnorm rbinom
#' @export
dataset <- function(name = "M1", n = NULL, p = 20, sigma = 0.5, ...) {
name <- toupper(name)
if (nchar(name) == 1) { name <- paste0("M", name) }
if (name == "M1") {
if (missing(n)) { n <- 100 }
# B ... `p x 1`
B <- matrix(c(rep(1 / sqrt(6), 6), rep(0, p - 6)), ncol = 1)
X <- rmvnorm(n, sigma = sigma^abs(outer(1:p, 1:p, FUN = `-`)))
beta <- 0.5
Y <- cos(X %*% B) + rgnorm(n, 0,
alpha = sqrt(0.25 * gamma(1 / beta) / gamma(3 / beta)),
beta = beta
)
} else if (name == "M2") {
if (missing(n)) { n <- 100 }
prob <- 0.3
lambda <- 1 # dispersion
# B ... `p x 1`
B <- matrix(c(rep(1 / sqrt(6), 6), rep(0, p - 6)), ncol = 1)
Z <- 2 * rbinom(n, 1, prob) - 1
X <- matrix(rep(lambda * Z, p) + rnorm(n * p), n)
Y <- cos(X %*% B) + rnorm(n, 0, sigma)
} else if (name == "M3") {
if (missing(n)) { n <- 200 }
# B ... `p x 1`
B <- matrix(c(rep(1 / sqrt(6), 6), rep(0, p - 6)), ncol = 1)
X <- matrix(rnorm(n * p), n)
Y <- 1.5 * log(2 + abs(X %*% B)) + rnorm(n, 0, sigma^2)
} else if (name == "M4") {
if (missing(n)) { n <- 200 }
# B ... `p x 2`
B <- cbind(
c(rep(1 / sqrt(6), 6), rep(0, p - 6)),
c(rep(c(1, -1), 3) / sqrt(6), rep(0, p - 6))
)
X <- rmvnorm(n, sigma = sigma^abs(outer(1:p, 1:p, FUN = `-`)))
XB <- X %*% B
Y <- (XB[, 1]) / (0.5 + (XB[, 2] + 1.5)^2) + rnorm(n, 0, sigma^2)
} else if (name == "M5") {
if (missing(n)) { n <- 200 }
# B ... `p x 2`
B <- cbind(
c(rep(1, 6), rep(0, p - 6)),
c(rep(c(1, -1), 3), rep(0, p - 6))
) / sqrt(6)
X <- matrix(runif(n * p), n)
XB <- X %*% B
Y <- cos(XB[, 1] * pi) * (XB[, 2] + 1)^2 + rnorm(n, 0, sigma^2)
} else if (name == "M6") {
if (missing(n)) { n <- 200 }
# B ... `p x 3`
B <- diag(p)[, -(3:(p - 1))]
X <- matrix(rnorm(n * p), n)
Y <- rowSums((X %*% B)^2) + rnorm(n, 0, sigma^2)
} else if (name == "M7") {
if (missing(n)) { n <- 400 }
# B ... `p x 4`
B <- diag(p)[, -(4:(p - 1))]
# "R"andom "M"ulti"V"ariate "S"tudent
X <- rmvt(n = n, sigma = diag(p), df = 3)
XB <- X %*% B
Y <- (XB[, 1]) * (XB[, 2])^2 + (XB[, 3]) * (XB[, 4])
Y <- Y + rlaplace(n, 0, sigma)
} else {
stop("Got unknown dataset name.")
}
return(list(X = X, Y = Y, B = B, name = name))
}

View File

@ -1,82 +0,0 @@
#' Draws a sample from the invariant measure on the Stiefel manifold \eqn{S(p, q)}.
#'
#' @param p row dimension
#' @param q col dimension
#' @return \code{p} times \code{q} semi-orthogonal matrix.
#' @examples
#' V <- rStiefel(6, 4)
#' @export
rStiefel <- function(p, q) {
return(qr.Q(qr(matrix(rnorm(p * q, 0, 1), p, q))))
}
#' Retraction to the manifold.
#'
#' @param A matrix.
#' @return `(p, q)` semi-orthogonal matrix, aka element of the Stiefel manifold.
#' @keywords internal
#' @export
retractStiefel <- function(A) {
return(qr.Q(qr(A)))
}
#' Skew-Symmetric matrix computed from `A` as
#' \eqn{1/2 (A - A^T)}.
#' @param A Matrix of dim `(p, q)`
#' @return Skew-Symmetric matrix of dim `(p, p)`.
#' @keywords internal
#' @export
skew <- function(A) {
0.5 * (A - t(A))
}
#' Symmetric matrix computed from `A` as
#' \eqn{1/2 (A + A^T)}.
#' @param A Matrix of dim `(p, q)`
#' @return Symmetric matrix of dim `(p, p)`.
#' @keywords internal
#' @export
sym <- function(A) {
0.5 * (A + t(A))
}
#' Orthogonal Projection onto the tangent space of the stiefel manifold.
#'
#' @param V Point on the stiefel manifold.
#' @param G matrix to be projected onto the tangent space at `V`.
#' @return `(p, q)` matrix as element of the tangent space at `V`.
#' @keywords internal
#' @export
projTangentStiefel <- function(V, G) {
Q <- diag(1, nrow(V)) - V %*% t(V)
return(Q %*% G + V %*% skew(t(V) %*% G))
}
#' Null space basis of given matrix `V`
#'
#' @param V `(p, q)` matrix
#' @return Semi-orthogonal `(p, p - q)` matrix spaning the null space of `V`.
#' @keywords internal
#' @export
null <- function(V) {
tmp <- qr(V)
set <- if(tmp$rank == 0L) seq_len(ncol(V)) else -seq_len(tmp$rank)
return(qr.Q(tmp, complete = TRUE)[, set, drop = FALSE])
}
#' Creates a (numeric) matrix where each column contains
#' an element to element matching.
#' @param elements numeric vector of elements to match
#' @return matrix of size `(2, n * (n - 1) / 2)` for a argument of lenght `n`.
#' @keywords internal
#' @examples
#' elem.pairs(seq.int(2, 5))
#' @export
elem.pairs <- function(elements) {
# Number of elements to match.
n <- length(elements)
# Create all combinations.
pairs <- rbind(rep(elements, each=n), rep(elements, n))
# Select unique combinations without self interaction.
return(pairs[, pairs[1, ] < pairs[2, ]])
}

View File

@ -1,68 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/datasets.R
\name{dataset}
\alias{dataset}
\title{Generates test datasets.}
\usage{
dataset(name = "M1", n, B, p.mix = 0.3, lambda = 1)
}
\arguments{
\item{name}{One of \code{"M1"}, \code{"M2"}, \code{"M3"}, \code{"M4"} or \code{"M5"}}
\item{n}{nr samples}
\item{B}{SDR basis used for dataset creation if supplied.}
\item{p.mix}{Only for \code{"M4"}, see: below.}
\item{lambda}{Only for \code{"M4"}, see: below.}
\item{p}{Dim. of random variable \code{X}.}
}
\value{
List with elements
\itemize{
\item{X}{data}
\item{Y}{response}
\item{B}{Used dim-reduction matrix}
\item{name}{Name of the dataset (name parameter)}
}
}
\description{
Provides sample datasets. There are 5 different datasets named
M1, M2, M3, M4 and M5 described in the paper references below.
The general model is given by:
\deqn{Y = g(B'X) + \epsilon}
}
\section{M1}{
The data follows \eqn{X\sim N_p(0, \Sigma)}{X ~ N_p(0, Sigma)} for a subspace
dimension of \eqn{k = 2} with a default of \eqn{n = 200} data points.
The link function \eqn{g} is given as
\deqn{g(x) = \frac{x_1}{0.5 + (x_2 + 1.5)^2} + \epsilon / 2}{%
g(x) = x_1 / (0.5 + (x_2 + 1.5)^2) + epsilon / 2}
}
\section{M2}{
\eqn{X\sim N_p(0, \Sigma)}{X ~ N_p(0, Sigma)} with \eqn{k = 2} with a
default of \eqn{n = 200} data points.
The link function \eqn{g} is given as
\deqn{g(x) = (b_1^T X) (b_2^T X)^2 + \epsilon / 2}
}
\section{M3}{
\deqn{g(x) = cos(b_1^T X) + \epsilon / 2}
}
\section{M4}{
TODO:
}
\section{M5}{
TODO:
}

View File

@ -1,23 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/util.R
\name{elem.pairs}
\alias{elem.pairs}
\title{Creates a (numeric) matrix where each column contains
an element to element matching.}
\usage{
elem.pairs(elements)
}
\arguments{
\item{elements}{numeric vector of elements to match}
}
\value{
matrix of size `(2, n * (n - 1) / 2)` for a argument of lenght `n`.
}
\description{
Creates a (numeric) matrix where each column contains
an element to element matching.
}
\examples{
elem.pairs(seq.int(2, 5))
}
\keyword{internal}

View File

@ -1,20 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/util.R
\name{projTangentStiefel}
\alias{projTangentStiefel}
\title{Orthogonal Projection onto the tangent space of the stiefel manifold.}
\usage{
projTangentStiefel(V, G)
}
\arguments{
\item{V}{Point on the stiefel manifold.}
\item{G}{matrix to be projected onto the tangent space at `V`.}
}
\value{
`(p, q)` matrix as element of the tangent space at `V`.
}
\description{
Orthogonal Projection onto the tangent space of the stiefel manifold.
}
\keyword{internal}

View File

@ -1,18 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/util.R
\name{retractStiefel}
\alias{retractStiefel}
\title{Retraction to the manifold.}
\usage{
retractStiefel(A)
}
\arguments{
\item{A}{matrix.}
}
\value{
`(p, q)` semi-orthogonal matrix, aka element of the Stiefel manifold.
}
\description{
Retraction to the manifold.
}
\keyword{internal}

View File

@ -1,20 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/util.R
\name{skew}
\alias{skew}
\title{Skew-Symmetric matrix computed from `A` as
\eqn{1/2 (A - A^T)}.}
\usage{
skew(A)
}
\arguments{
\item{A}{Matrix of dim `(p, q)`}
}
\value{
Skew-Symmetric matrix of dim `(p, p)`.
}
\description{
Skew-Symmetric matrix computed from `A` as
\eqn{1/2 (A - A^T)}.
}
\keyword{internal}

View File

@ -1,20 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/util.R
\name{sym}
\alias{sym}
\title{Symmetric matrix computed from `A` as
\eqn{1/2 (A + A^T)}.}
\usage{
sym(A)
}
\arguments{
\item{A}{Matrix of dim `(p, q)`}
}
\value{
Symmetric matrix of dim `(p, p)`.
}
\description{
Symmetric matrix computed from `A` as
\eqn{1/2 (A + A^T)}.
}
\keyword{internal}

View File

@ -1,11 +0,0 @@
Package: CVEpureR
Type: Package
Title: Conditional Variance Estimator for Sufficient Dimension Reduction
Version: 0.1
Date: 2019-08-29
Author: Loki
Maintainer: Loki <loki@no.mail>
Description: Implementation of the Conditional Variance Estimation (CVE) method. This package version is writen in pure R.
License: GPL-3
Encoding: UTF-8
RoxygenNote: 6.1.1

View File

@ -1,23 +0,0 @@
# Generated by roxygen2: do not edit by hand
S3method(plot,cve)
S3method(summary,cve)
export(cve)
export(cve.call)
export(cve.grid.search)
export(cve_linesearch)
export(cve_sgd)
export(cve_simple)
export(dataset)
export(elem.pairs)
export(estimate.bandwidth)
export(grad)
export(null)
export(rStiefl)
import(stats)
importFrom(graphics,lines)
importFrom(graphics,plot)
importFrom(graphics,points)
importFrom(stats,model.frame)
importFrom(stats,rbinom)
importFrom(stats,rnorm)

View File

@ -1,265 +0,0 @@
#' Conditional Variance Estimator (CVE)
#'
#' Conditional Variance Estimator for Sufficient Dimension
#' Reduction
#'
#' TODO: And some details
#'
#'
#' @references Fertl Likas, Bura Efstathia. Conditional Variance Estimation for Sufficient Dimension Reduction, 2019
#'
#' @docType package
#' @author Loki
#' @useDynLib CVE, .registration = TRUE
"_PACKAGE"
#' Implementation of the CVE method.
#'
#' Conditional Variance Estimator (CVE) is a novel sufficient dimension
#' reduction (SDR) method assuming a model
#' \deqn{Y \sim g(B'X) + \epsilon}{Y ~ g(B'X) + epsilon}
#' where B'X is a lower dimensional projection of the predictors.
#'
#' @param formula Formel for the regression model defining `X`, `Y`.
#' See: \code{\link{formula}}.
#' @param data data.frame holding data for formula.
#' @param method The different only differe in the used optimization.
#' All of them are Gradient based optimization on a Stiefel manifold.
#' \itemize{
#' \item "simple" Simple reduction of stepsize.
#' \item "sgd" stocastic gradient decent.
#' \item TODO: further
#' }
#' @param ... Further parameters depending on the used method.
#' @examples
#' library(CVE)
#'
#' # sample dataset
#' ds <- dataset("M5")
#'
#' # call ´cve´ with default method (aka "simple")
#' dr.simple <- cve(ds$Y ~ ds$X, k = ncol(ds$B))
#' # plot optimization history (loss via iteration)
#' plot(dr.simple, main = "CVE M5 simple")
#'
#' # call ´cve´ with method "linesearch" using ´data.frame´ as data.
#' data <- data.frame(Y = ds$Y, X = ds$X)
#' # Note: ´Y, X´ are NOT defined, they are extracted from ´data´.
#' dr.linesearch <- cve(Y ~ ., data, method = "linesearch", k = ncol(ds$B))
#' plot(dr.linesearch, main = "CVE M5 linesearch")
#'
#' @references Fertl L., Bura E. Conditional Variance Estimation for Sufficient Dimension Reduction, 2019
#'
#' @seealso \code{\link{formula}}. For a complete parameters list (dependent on
#' the method) see \code{\link{cve_simple}}, \code{\link{cve_sgd}}
#' @import stats
#' @importFrom stats model.frame
#' @export
cve <- function(formula, data, method = "simple", max.dim = 10L, ...) {
# check for type of `data` if supplied and set default
if (missing(data)) {
data <- environment(formula)
} else if (!is.data.frame(data)) {
stop("Parameter 'data' must be a 'data.frame' or missing.")
}
# extract `X`, `Y` from `formula` with `data`
model <- stats::model.frame(formula, data)
X <- as.matrix(model[ ,-1L, drop = FALSE])
Y <- as.double(model[ , 1L])
# pass extracted data on to [cve.call()]
dr <- cve.call(X, Y, method = method, max.dim = max.dim, ...)
# overwrite `call` property from [cve.call()]
dr$call <- match.call()
return(dr)
}
#' @param nObs as described in the Paper.
#' @param X Data
#' @param Y Responces
#' @param nObs Like in the paper.
#' @param k guess for SDR dimension.
#' @param ... Method specific parameters.
#' @rdname cve
#' @export
cve.call <- function(X, Y, method = "simple",
nObs = sqrt(nrow(X)), h = NULL,
min.dim = 1L, max.dim = 10L, k = NULL,
tau = 1.0, tol = 1e-3,
epochs = 50L, attempts = 10L,
logger = NULL) {
# parameter checking
if (!(is.matrix(X) && is.numeric(X))) {
stop("Parameter 'X' should be a numeric matrices.")
}
if (!is.numeric(Y)) {
stop("Parameter 'Y' must be numeric.")
}
if (is.matrix(Y) || !is.double(Y)) {
Y <- as.double(Y)
}
if (nrow(X) != length(Y)) {
stop("Rows of 'X' and 'Y' elements are not compatible.")
}
if (ncol(X) < 2) {
stop("'X' is one dimensional, no need for dimension reduction.")
}
if (missing(k) || is.null(k)) {
min.dim <- as.integer(min.dim)
max.dim <- as.integer(min(max.dim, ncol(X) - 1L))
} else {
min.dim <- as.integer(k)
max.dim <- as.integer(k)
}
if (min.dim > max.dim) {
stop("'min.dim' bigger 'max.dim'.")
}
if (max.dim >= ncol(X)) {
stop("'max.dim' (or 'k') must be smaller than 'ncol(X)'.")
}
if (is.function(h)) {
estimate.bandwidth <- h
h <- NULL
}
if (!is.numeric(tau) || length(tau) > 1L || tau <= 0.0) {
stop("Initial step-width 'tau' must be positive number.")
} else {
tau <- as.double(tau)
}
if (!is.numeric(tol) || length(tol) > 1L || tol < 0.0) {
stop("Break condition tolerance 'tol' must be not negative number.")
} else {
tol <- as.double(tol)
}
if (!is.numeric(epochs) || length(epochs) > 1L) {
stop("Parameter 'epochs' must be positive integer.")
} else if (!is.integer(epochs)) {
epochs <- as.integer(epochs)
}
if (epochs < 1L) {
stop("Parameter 'epochs' must be at least 1L.")
}
if (!is.numeric(attempts) || length(attempts) > 1L) {
stop("Parameter 'attempts' must be positive integer.")
} else if (!is.integer(attempts)) {
attempts <- as.integer(attempts)
}
if (attempts < 1L) {
stop("Parameter 'attempts' must be at least 1L.")
}
if (is.function(logger)) {
loggerEnv <- environment(logger)
} else {
loggerEnv <- NULL
}
# Call specified method.
method <- tolower(method)
call <- match.call()
dr <- list()
for (k in min.dim:max.dim) {
if (missing(h) || is.null(h)) {
h <- estimate.bandwidth(X, k, nObs)
} else if (is.numeric(h) && h > 0.0) {
h <- as.double(h)
} else {
stop("Bandwidth 'h' must be positive numeric.")
}
if (method == 'simple') {
dr.k <- .Call('cve_simple', PACKAGE = 'CVE',
X, Y, k, h,
tau, tol,
epochs, attempts,
logger, loggerEnv)
# dr.k <- cve_simple(X, Y, k, nObs = nObs, ...)
# } else if (method == 'linesearch') {
# dr.k <- cve_linesearch(X, Y, k, nObs = nObs, ...)
# } else if (method == 'rcg') {
# dr.k <- cve_rcg(X, Y, k, nObs = nObs, ...)
# } else if (method == 'momentum') {
# dr.k <- cve_momentum(X, Y, k, nObs = nObs, ...)
# } else if (method == 'rmsprob') {
# dr.k <- cve_rmsprob(X, Y, k, nObs = nObs, ...)
# } else if (method == 'sgdrmsprob') {
# dr.k <- cve_sgdrmsprob(X, Y, k, nObs = nObs, ...)
# } else if (method == 'sgd') {
# dr.k <- cve_sgd(X, Y, k, nObs = nObs, ...)
} else {
stop('Got unknown method.')
}
dr.k$B <- null(dr.k$V)
dr.k$loss <- mean(dr.k$L)
dr.k$h <- h
dr.k$k <- k
class(dr.k) <- "cve.k"
dr[[k]] <- dr.k
}
# augment result information
dr$method <- method
dr$call <- call
class(dr) <- "cve"
return(dr)
}
#' Ploting helper for objects of class \code{cve}.
#'
#' @param x Object of class \code{cve} (result of [cve()]).
#' @param content Specifies what to plot:
#' \itemize{
#' \item "history" Plots the loss history from stiefel optimization
#' (default).
#' \item ... TODO: add (if there are any)
#' }
#' @param ... Pass through parameters to [plot()] and [lines()]
#'
#' @usage ## S3 method for class 'cve'
#' plot(x, content = "history", ...)
#' @seealso see \code{\link{par}} for graphical parameters to pass through
#' as well as \code{\link{plot}} for standard plot utility.
#' @importFrom graphics plot lines points
#' @method plot cve
#' @export
plot.cve <- function(x, ...) {
L <- c()
k <- c()
for (dr.k in x) {
if (class(dr.k) == 'cve.k') {
k <- c(k, paste0(dr.k$k))
L <- c(L, dr.k$L)
}
}
L <- matrix(L, ncol = length(k))
boxplot(L, main = "Loss ...",
xlab = "SDR dimension k",
ylab = expression(L(V, X[i])),
names = k)
}
#' Prints a summary of a \code{cve} result.
#' @param object Instance of 'cve' as return of \code{cve}.
#' @method summary cve
#' @export
summary.cve <- function(object, ...) {
cat('Summary of CVE result - Method: "', object$method, '"\n',
'\n',
'Dataset size: ', nrow(object$X), '\n',
'Data Dimension: ', ncol(object$X), '\n',
'SDR Dimension: ', object$k, '\n',
'loss: ', object$loss, '\n',
'\n',
'Called via:\n',
' ',
sep='')
print(object$call)
}

View File

@ -1,169 +0,0 @@
#' Implementation of the CVE method using curvilinear linesearch with Armijo-Wolfe
#' conditions.
#'
#' @keywords internal
#' @export
cve_linesearch <- function(X, Y, k,
nObs = sqrt(nrow(X)),
h = NULL,
tau = 1.0,
tol = 1e-3,
rho1 = 0.1,
rho2 = 0.9,
slack = 0,
epochs = 50L,
attempts = 10L,
max.linesearch.iter = 10L,
logger = NULL
) {
# Set `grad` functions environment to enable if to find this environments
# local variabels, needed to enable the manipulation of this local variables
# from within `grad`.
environment(grad) <- environment()
# Get dimensions.
n <- nrow(X)
p <- ncol(X)
q <- p - k
# Save initial learning rate `tau`.
tau.init <- tau
# Addapt tolearance for break condition.
tol <- sqrt(2 * q) * tol
# Estaimate bandwidth if not given.
if (missing(h) | !is.numeric(h)) {
h <- estimate.bandwidth(X, k, nObs)
}
# Compute persistent data.
# Compute lookup indexes for symmetrie, lower/upper
# triangular parts and vectorization.
pair.index <- elem.pairs(seq(n))
i <- pair.index[1, ] # `i` indices of `(i, j)` pairs
j <- pair.index[2, ] # `j` indices of `(i, j)` pairs
# Matrix of vectorized indices. (vec(index) -> seq)
index <- matrix(seq(n * n), n, n)
lower <- index[lower.tri(index)]
upper <- t(index)[lower]
# Create all pairewise differences of rows of `X`.
X_diff <- X[i, , drop = F] - X[j, , drop = F]
# Identity matrix.
I_p <- diag(1, p)
# Init tracking of current best (according multiple attempts).
V.best <- NULL
loss.best <- Inf
# Start loop for multiple attempts.
for (attempt in 1:attempts) {
# Sample a `(p, q)` dimensional matrix from the stiefel manifold as
# optimization start value.
V <- rStiefl(p, q)
# Initial loss and gradient.
loss <- Inf
G <- grad(X, Y, V, h, loss.out = TRUE, persistent = TRUE)
# Set last loss (aka, loss after applying the step).
loss.last <- loss
# Call logger with initial values before starting optimization.
if (is.function(logger)) {
epoch <- 0 # Set epoch count to 0 (only relevant for logging).
error <- NA
logger(environment())
}
## Start optimization loop.
for (epoch in 1:epochs) {
# Cayley transform matrix `A`
A <- (G %*% t(V)) - (V %*% t(G))
# Directional derivative of the loss at current position, given
# as `Tr(G^T \cdot A \cdot V)`.
loss.prime <- -0.5 * norm(A, type = 'F')^2
# Linesearch
tau.upper <- Inf
tau.lower <- 0
tau <- tau.init
for (iter in 1:max.linesearch.iter) {
# Apply learning rate `tau`.
A.tau <- (tau / 2) * A
# Parallet transport (on Stiefl manifold) into direction of `G`.
inv <- solve(I_p + A.tau)
V.tau <- inv %*% ((I_p - A.tau) %*% V)
# Loss at position after a step.
loss <- Inf # aka loss.tau
G.tau <- grad(X, Y, V.tau, h, loss.out = TRUE, persistent = TRUE)
# Armijo condition.
if (loss > loss.last + (rho1 * tau * loss.prime)) {
tau.upper <- tau
tau <- (tau.lower + tau.upper) / 2
next()
}
V.prime.tau <- -0.5 * inv %*% A %*% (V + V.tau)
loss.prime.tau <- sum(G * V.prime.tau) # Tr(grad(tau)^T \cdot Y^'(tau))
# Wolfe condition.
if (loss.prime.tau < rho2 * loss.prime) {
tau.lower <- tau
if (tau.upper == Inf) {
tau <- 2 * tau.lower
} else {
tau <- (tau.lower + tau.upper) / 2
}
} else {
break()
}
}
# Compute error.
error <- norm(V %*% t(V) - V.tau %*% t(V.tau), type = "F")
# Check break condition (epoch check to skip ignored gradient calc).
# Note: the devision by `sqrt(2 * k)` is included in `tol`.
if (error < tol | epoch >= epochs) {
# take last step and stop optimization.
V <- V.tau
# Final call to the logger before stopping optimization
if (is.function(logger)) {
G <- G.tau
logger(environment())
}
break()
}
# Perform the step and remember previous loss.
V <- V.tau
loss.last <- loss
G <- G.tau
# Log after taking current step.
if (is.function(logger)) {
logger(environment())
}
}
# Check if current attempt improved previous ones
if (loss < loss.best) {
loss.best <- loss
V.best <- V
}
}
return(list(
loss = loss.best,
V = V.best,
B = null(V.best),
h = h
))
}

View File

@ -1,139 +0,0 @@
#' Implementation of the CVE method as a Riemann Conjugated Gradient method.
#'
#' @references A Riemannian Conjugate Gradient Algorithm with Implicit Vector
#' Transport for Optimization on the Stiefel Manifold
#' @keywords internal
#' @export
cve_momentum <- function(X, Y, k,
nObs = sqrt(nrow(X)),
h = NULL,
tau = 1.0,
tol = 1e-4,
rho = 0.1, # Momentum update.
slack = 0,
epochs = 50L,
attempts = 10L,
logger = NULL
) {
# Set `grad` functions environment to enable if to find this environments
# local variabels, needed to enable the manipulation of this local variables
# from within `grad`.
environment(grad) <- environment()
# Get dimensions.
n <- nrow(X) # Number of samples.
p <- ncol(X) # Data dimensions
q <- p - k # Complement dimension of the SDR space.
# Save initial learning rate `tau`.
tau.init <- tau
# Addapt tolearance for break condition.
tol <- sqrt(2 * q) * tol
# Estaimate bandwidth if not given.
if (missing(h) || !is.numeric(h)) {
h <- estimate.bandwidth(X, k, nObs)
}
# Compute persistent data.
# Compute lookup indexes for symmetrie, lower/upper
# triangular parts and vectorization.
pair.index <- elem.pairs(seq(n))
i <- pair.index[1, ] # `i` indices of `(i, j)` pairs
j <- pair.index[2, ] # `j` indices of `(i, j)` pairs
# Index of vectorized matrix, for lower and upper triangular part.
lower <- ((i - 1) * n) + j
upper <- ((j - 1) * n) + i
# Create all pairewise differences of rows of `X`.
X_diff <- X[i, , drop = F] - X[j, , drop = F]
# Identity matrix.
I_p <- diag(1, p)
# Init tracking of current best (according multiple attempts).
V.best <- NULL
loss.best <- Inf
# Start loop for multiple attempts.
for (attempt in 1:attempts) {
# Reset learning rate `tau`.
tau <- tau.init
# Sample a `(p, q)` dimensional matrix from the stiefel manifold as
# optimization start value.
V <- rStiefl(p, q)
# Initial loss and gradient.
loss <- Inf
G <- grad(X, Y, V, h, loss.out = TRUE, persistent = TRUE)
# Set last loss (aka, loss after applying the step).
loss.last <- loss
# Call logger with initial values before starting optimization.
if (is.function(logger)) {
epoch <- 0 # Set epoch count to 0 (only relevant for logging).
error <- NA
logger(environment())
}
M <- matrix(0, p, q)
## Start optimization loop.
for (epoch in 1:epochs) {
# Apply learning rate `tau`.
A <- projTangentStiefl(V, G)
# Momentum update.
M <- A + rho * projTangentStiefl(V, M)
# Parallet transport (on Stiefl manifold) into direction of `G`.
V.tau <- retractStiefl(V - tau * M)
# Loss at position after a step.
loss <- grad(X, Y, V.tau, h, loss.only = TRUE, persistent = TRUE)
# Check if step is appropriate, iff not reduce learning rate.
if ((loss - loss.last) > slack * loss.last) {
tau <- tau / 2
next() # Keep position and try with smaller `tau`.
}
# Compute error.
error <- norm(V %*% t(V) - V.tau %*% t(V.tau), type = "F")
# Check break condition (epoch check to skip ignored gradient calc).
# Note: the devision by `sqrt(2 * k)` is included in `tol`.
if (error < tol || epoch >= epochs) {
# take last step and stop optimization.
V <- V.tau
# Call logger last time befor stoping.
if (is.function(logger)) {
logger(environment())
}
break()
}
# Perform the step and remember previous loss.
V <- V.tau
loss.last <- loss
# Call logger after taking a step.
if (is.function(logger)) {
logger(environment())
}
# Compute gradient at new position.
G <- grad(X, Y, V, h, persistent = TRUE)
}
# Check if current attempt improved previous ones
if (loss < loss.best) {
loss.best <- loss
V.best <- V
}
}
return(list(
loss = loss.best,
V = V.best,
B = null(V.best),
h = h
))
}

View File

@ -1,179 +0,0 @@
#' Implementation of the CVE method as a Riemann Conjugated Gradient method.
#'
#' @references A Riemannian Conjugate Gradient Algorithm with Implicit Vector
#' Transport for Optimization on the Stiefel Manifold
#' @keywords internal
#' @export
cve_rcg <- function(X, Y, k,
nObs = sqrt(nrow(X)),
h = NULL,
tau = 1.0,
tol = 1e-4,
rho = 1e-4, # For Armijo condition.
slack = 0,
epochs = 50L,
attempts = 10L,
max.linesearch.iter = 20L,
logger = NULL
) {
# Set `grad` functions environment to enable if to find this environments
# local variabels, needed to enable the manipulation of this local variables
# from within `grad`.
environment(grad) <- environment()
# Get dimensions.
n <- nrow(X) # Number of samples.
p <- ncol(X) # Data dimensions
q <- p - k # Complement dimension of the SDR space.
# Save initial learning rate `tau`.
tau.init <- tau
# Addapt tolearance for break condition.
tol <- sqrt(2 * q) * tol
# Estaimate bandwidth if not given.
if (missing(h) || !is.numeric(h)) {
h <- estimate.bandwidth(X, k, nObs)
}
# Compute persistent data.
# Compute lookup indexes for symmetrie, lower/upper
# triangular parts and vectorization.
pair.index <- elem.pairs(seq(n))
i <- pair.index[1, ] # `i` indices of `(i, j)` pairs
j <- pair.index[2, ] # `j` indices of `(i, j)` pairs
# Index of vectorized matrix, for lower and upper triangular part.
lower <- ((i - 1) * n) + j
upper <- ((j - 1) * n) + i
# Create all pairewise differences of rows of `X`.
X_diff <- X[i, , drop = F] - X[j, , drop = F]
# Identity matrix.
I_p <- diag(1, p)
# Init tracking of current best (according multiple attempts).
V.best <- NULL
loss.best <- Inf
# Start loop for multiple attempts.
for (attempt in 1:attempts) {
# Reset learning rate `tau`.
tau <- tau.init
# Sample a `(p, q)` dimensional matrix from the stiefel manifold as
# optimization start value.
V <- rStiefl(p, q)
# Initial loss and gradient.
loss <- Inf
G <- grad(X, Y, V, h, loss.out = TRUE, persistent = TRUE)
# Set last loss (aka, loss after applying the step).
loss.last <- loss
# Cayley transform matrix `A`
A <- (G %*% t(V)) - (V %*% t(G))
A.last <- A
W <- -A
Z <- W %*% V
# Compute directional derivative.
loss.prime <- sum(G * Z) # Tr(G^T Z)
# Call logger with initial values before starting optimization.
if (is.function(logger)) {
epoch <- 0 # Set epoch count to 0 (only relevant for logging).
error <- NA
logger(environment())
}
## Start optimization loop.
for (epoch in 1:epochs) {
# New directional derivative.
loss.prime <- sum(G * Z)
# Reset `tau` for step-size selection.
tau <- tau.init
for (iter in 1:max.linesearch.iter) {
V.tau <- retractStiefl(V + tau * Z)
# Loss at position after a step.
loss <- grad(X, Y, V.tau, h,
loss.only = TRUE, persistent = TRUE)
# Check Armijo condition.
if (loss <= loss.last + (rho * tau * loss.prime)) {
break() # Iff fulfilled stop linesearch.
}
# Reduce step-size and continue linesearch.
tau <- tau / 2
}
# Compute error.
error <- norm(V %*% t(V) - V.tau %*% t(V.tau), type = "F")
# Perform step with found step-size
V <- V.tau
loss.last <- loss
# Call logger.
if (is.function(logger)) {
logger(environment())
}
# Check break condition.
# Note: the devision by `sqrt(2 * k)` is included in `tol`.
if (error < tol) {
break()
}
# Compute Gradient at new position.
G <- grad(X, Y, V, h, persistent = TRUE)
# Store last `A` for `beta` computation.
A.last <- A
# Cayley transform matrix `A`
A <- (G %*% t(V)) - (V %*% t(G))
# Check 2. break condition.
if (norm(A, type = 'F') < tol) {
break()
}
# New directional derivative.
loss.prime <- sum(G * Z)
# Reset beta if needed.
if (loss.prime < 0) {
# Compute `beta` as described in paper.
beta.FR <- (norm(A, type = 'F') / norm(A.last, type = 'F'))^2
beta.PR <- sum(A * (A - A.last)) / norm(A.last, type = 'F')^2
if (beta.PR < -beta.FR) {
beta <- -beta.FR
} else if (abs(beta.PR) < beta.FR) {
beta <- beta.PR
} else if (beta.PR > beta.FR) {
beta <- beta.FR
} else {
beta <- 0
}
} else {
beta <- 0
}
# Update direction.
W <- -A + beta * W
Z <- W %*% V
}
# Check if current attempt improved previous ones
if (loss < loss.best) {
loss.best <- loss
V.best <- V
}
}
return(list(
loss = loss.best,
V = V.best,
B = null(V.best),
h = h
))
}

View File

@ -1,121 +0,0 @@
#' Implementation of the CVE method as a Riemann Conjugated Gradient method.
#'
#' @references A Riemannian Conjugate Gradient Algorithm with Implicit Vector
#' Transport for Optimization on the Stiefel Manifold
#' @keywords internal
#' @export
cve_rmsprob <- function(X, Y, k,
nObs = sqrt(nrow(X)),
h = NULL,
tau = 0.1,
tol = 1e-4,
rho = 0.1, # Momentum update.
slack = 0,
epochs = 50L,
attempts = 10L,
epsilon = 1e-7,
max.linesearch.iter = 20L,
logger = NULL
) {
# Set `grad` functions environment to enable if to find this environments
# local variabels, needed to enable the manipulation of this local variables
# from within `grad`.
environment(grad) <- environment()
# Get dimensions.
n <- nrow(X) # Number of samples.
p <- ncol(X) # Data dimensions
q <- p - k # Complement dimension of the SDR space.
# Save initial learning rate `tau`.
tau.init <- tau
# Addapt tolearance for break condition.
tol <- sqrt(2 * q) * tol
# Estaimate bandwidth if not given.
if (missing(h) || !is.numeric(h)) {
h <- estimate.bandwidth(X, k, nObs)
}
# Compute persistent data.
# Compute lookup indexes for symmetrie, lower/upper
# triangular parts and vectorization.
pair.index <- elem.pairs(seq(n))
i <- pair.index[1, ] # `i` indices of `(i, j)` pairs
j <- pair.index[2, ] # `j` indices of `(i, j)` pairs
# Index of vectorized matrix, for lower and upper triangular part.
lower <- ((i - 1) * n) + j
upper <- ((j - 1) * n) + i
# Create all pairewise differences of rows of `X`.
X_diff <- X[i, , drop = F] - X[j, , drop = F]
# Identity matrix.
I_p <- diag(1, p)
# Init tracking of current best (according multiple attempts).
V.best <- NULL
loss.best <- Inf
# Start loop for multiple attempts.
for (attempt in 1:attempts) {
# Sample a `(p, q)` dimensional matrix from the stiefel manifold as
# optimization start value.
V <- rStiefl(p, q)
# Call logger with initial values before starting optimization.
if (is.function(logger)) {
loss <- grad(X, Y, V, h, loss.only = TRUE, persistent = TRUE)
epoch <- 0 # Set epoch count to 0 (only relevant for logging).
error <- NA
logger(environment())
}
M <- matrix(0, p, q)
## Start optimization loop.
for (epoch in 1:epochs) {
# Compute gradient and loss at current position.
loss <- Inf
G <- grad(X, Y, V, h, loss.out = TRUE, persistent = TRUE)
# Projectd Gradient.
A <- projTangentStiefl(V, G)
# Projected element squared gradient.
Asq <- projTangentStiefl(V, G * G)
# Momentum update.
M <- (1 - rho) * Asq + rho * projTangentStiefl(V, M)
# Parallet transport (on Stiefl manifold) into direction of `G`.
V.tau <- retractStiefl(V - tau.init * A / (sqrt(abs(M)) + epsilon))
# Compute error.
error <- norm(V %*% t(V) - V.tau %*% t(V.tau), type = "F")
# Perform step.
V <- V.tau
# Call logger after taking a step.
if (is.function(logger)) {
# Set tau to an step size estimate (only for logging)
tau <- tau.init / mean(sqrt(abs(M)) + epsilon)
logger(environment())
}
# Check break condition.
# Note: the devision by `sqrt(2 * k)` is included in `tol`.
if (error < tol) {
break()
}
}
# Check if current attempt improved previous ones
if (loss < loss.best) {
loss.best <- loss
V.best <- V
}
}
return(list(
loss = loss.best,
V = V.best,
B = null(V.best),
h = h
))
}

View File

@ -1,129 +0,0 @@
#' Simple implementation of the CVE method. 'Simple' means that this method is
#' a classic GD method unsing no further tricks.
#'
#' @keywords internal
#' @export
cve_sgd <- function(X, Y, k,
nObs = sqrt(nrow(X)),
h = NULL,
tau = 0.01,
tol = 1e-3,
epochs = 50L,
batch.size = 16L,
attempts = 10L,
logger = NULL
) {
# Set `grad` functions environment to enable if to find this environments
# local variabels, needed to enable the manipulation of this local variables
# from within `grad`.
environment(grad) <- environment()
# Get dimensions.
n <- nrow(X) # Number of samples.
p <- ncol(X) # Data dimensions
q <- p - k # Complement dimension of the SDR space.
# Save initial learning rate `tau`.
tau.init <- tau
# Addapt tolearance for break condition.
tol <- sqrt(2 * q) * tol
# Estaimate bandwidth if not given.
if (missing(h) || !is.numeric(h)) {
h <- estimate.bandwidth(X, k, nObs)
}
# Compute persistent data.
# Compute lookup indexes for symmetrie, lower/upper
# triangular parts and vectorization.
pair.index <- elem.pairs(seq(n))
i <- pair.index[1, ] # `i` indices of `(i, j)` pairs
j <- pair.index[2, ] # `j` indices of `(i, j)` pairs
# Index of vectorized matrix, for lower and upper triangular part.
lower <- ((i - 1) * n) + j
upper <- ((j - 1) * n) + i
# Create all pairewise differences of rows of `X`.
X_diff <- X[i, , drop = F] - X[j, , drop = F]
# Identity matrix.
I_p <- diag(1, p)
# Init a list of data indices (shuffled for batching).
indices <- seq(n)
# Init tracking of current best (according multiple attempts).
V.best <- NULL
loss.best <- Inf
# Start loop for multiple attempts.
for (attempt in 1:attempts) {
# Reset learning rate `tau`.
tau <- tau.init
# Sample a `(p, q)` dimensional matrix from the stiefel manifold as
# optimization start value.
V <- rStiefl(p, q)
# Keep track of last `V` for computing error after an epoch.
V.last <- V
if (is.function(logger)) {
loss <- grad(X, Y, V, h, loss.only = TRUE, persistent = TRUE)
error <- NA
epoch <- 0
logger(environment())
}
# Repeat `epochs` times
for (epoch in 1:epochs) {
# Shuffle batches
batch.shuffle <- sample(indices)
# Make a step for each batch.
for (batch.start in seq(1, n, batch.size)) {
# Select batch data indices.
batch.end <- min(batch.start + batch.size - 1, length(batch.shuffle))
batch <- batch.shuffle[batch.start:batch.end]
# Compute batch gradient.
loss <- NULL
G <- grad(X[batch, ], Y[batch], V, h, loss.out = TRUE)
# Cayley transform matrix.
A <- (G %*% t(V)) - (V %*% t(G))
# Apply learning rate `tau`.
A.tau <- tau * A
# Parallet transport (on Stiefl manifold) into direction of `G`.
V <- solve(I_p + A.tau) %*% ((I_p - A.tau) %*% V)
}
# And the error for the history.
error <- norm(V.last %*% t(V.last) - V %*% t(V), type = "F")
V.last <- V
if (is.function(logger)) {
# Compute loss at end of epoch for logging.
loss <- grad(X, Y, V, h, loss.only = TRUE, persistent = TRUE)
logger(environment())
}
# Check break condition.
if (error < tol) {
break()
}
}
# Compute actual loss after finishing for comparing multiple attempts.
loss <- grad(X, Y, V, h, loss.only = TRUE, persistent = TRUE)
# After each attempt, check if last attempt reached a better result.
if (loss < loss.best) {
loss.best <- loss
V.best <- V
}
}
return(list(
loss = loss.best,
V = V.best,
B = null(V.best),
h = h
))
}

View File

@ -1,133 +0,0 @@
#' Simple implementation of the CVE method. 'Simple' means that this method is
#' a classic GD method unsing no further tricks.
#'
#' @keywords internal
#' @export
cve_sgdrmsprob <- function(X, Y, k,
nObs = sqrt(nrow(X)),
h = NULL,
tau = 0.1,
tol = 1e-4,
rho = 0.1,
epochs = 50L,
batch.size = 16L,
attempts = 10L,
epsilon = 1e-7,
logger = NULL
) {
# Set `grad` functions environment to enable if to find this environments
# local variabels, needed to enable the manipulation of this local variables
# from within `grad`.
environment(grad) <- environment()
# Get dimensions.
n <- nrow(X) # Number of samples.
p <- ncol(X) # Data dimensions
q <- p - k # Complement dimension of the SDR space.
# Save initial learning rate `tau`.
tau.init <- tau
# Addapt tolearance for break condition.
tol <- sqrt(2 * q) * tol
# Estaimate bandwidth if not given.
if (missing(h) || !is.numeric(h)) {
h <- estimate.bandwidth(X, k, nObs)
}
# Compute persistent data.
# Compute lookup indexes for symmetrie, lower/upper
# triangular parts and vectorization.
pair.index <- elem.pairs(seq(n))
i <- pair.index[1, ] # `i` indices of `(i, j)` pairs
j <- pair.index[2, ] # `j` indices of `(i, j)` pairs
# Index of vectorized matrix, for lower and upper triangular part.
lower <- ((i - 1) * n) + j
upper <- ((j - 1) * n) + i
# Create all pairewise differences of rows of `X`.
X_diff <- X[i, , drop = F] - X[j, , drop = F]
# Identity matrix.
I_p <- diag(1, p)
# Init a list of data indices (shuffled for batching).
indices <- seq(n)
# Init tracking of current best (according multiple attempts).
V.best <- NULL
loss.best <- Inf
# Start loop for multiple attempts.
for (attempt in 1:attempts) {
# Reset learning rate `tau`.
tau <- tau.init
# Sample a `(p, q)` dimensional matrix from the stiefel manifold as
# optimization start value.
V <- rStiefl(p, q)
# Keep track of last `V` for computing error after an epoch.
V.last <- V
if (is.function(logger)) {
loss <- grad(X, Y, V, h, loss.only = TRUE, persistent = TRUE)
error <- NA
epoch <- 0
logger(environment())
}
M <- matrix(0, p, q)
# Repeat `epochs` times
for (epoch in 1:epochs) {
# Shuffle batches
batch.shuffle <- sample(indices)
# Make a step for each batch.
for (batch.start in seq(1, n, batch.size)) {
# Select batch data indices.
batch.end <- min(batch.start + batch.size - 1, length(batch.shuffle))
batch <- batch.shuffle[batch.start:batch.end]
# Compute batch gradient.
loss <- NULL
G <- grad(X[batch, ], Y[batch], V, h, loss.out = TRUE)
# Projectd Gradient.
A <- projTangentStiefl(V, G)
# Projected element squared gradient.
Asq <- projTangentStiefl(V, G * G)
# Momentum update.
M <- (1 - rho) * Asq + rho * projTangentStiefl(V, M)
# Parallet transport (on Stiefl manifold) into direction of `G`.
V <- retractStiefl(V - tau.init * A / (sqrt(abs(M)) + epsilon))
}
# And the error for the history.
error <- norm(V.last %*% t(V.last) - V %*% t(V), type = "F")
V.last <- V
if (is.function(logger)) {
# Compute loss at end of epoch for logging.
loss <- grad(X, Y, V, h, loss.only = TRUE, persistent = TRUE)
logger(environment())
}
# Check break condition.
if (error < tol) {
break()
}
}
# Compute actual loss after finishing for comparing multiple attempts.
loss <- grad(X, Y, V, h, loss.only = TRUE, persistent = TRUE)
# After each attempt, check if last attempt reached a better result.
if (loss < loss.best) {
loss.best <- loss
V.best <- V
}
}
return(list(
loss = loss.best,
V = V.best,
B = null(V.best),
h = h
))
}

View File

@ -1,139 +0,0 @@
#' Simple implementation of the CVE method. 'Simple' means that this method is
#' a classic GD method unsing no further tricks.
#'
#' @keywords internal
#' @export
cve_simple <- function(X, Y, k,
nObs = sqrt(nrow(X)),
h = NULL,
tau = 1.0,
tol = 1e-3,
slack = 0,
epochs = 50L,
attempts = 10L,
logger = NULL
) {
# Set `grad` functions environment to enable if to find this environments
# local variabels, needed to enable the manipulation of this local variables
# from within `grad`.
environment(grad) <- environment()
# Get dimensions.
n <- nrow(X) # Number of samples.
p <- ncol(X) # Data dimensions
q <- p - k # Complement dimension of the SDR space.
# Save initial learning rate `tau`.
tau.init <- tau
# Addapt tolearance for break condition.
tol <- sqrt(2 * q) * tol
# Estaimate bandwidth if not given.
if (missing(h) || !is.numeric(h)) {
h <- estimate.bandwidth(X, k, nObs)
}
# Compute persistent data.
# Compute lookup indexes for symmetrie, lower/upper
# triangular parts and vectorization.
pair.index <- elem.pairs(seq(n))
i <- pair.index[1, ] # `i` indices of `(i, j)` pairs
j <- pair.index[2, ] # `j` indices of `(i, j)` pairs
# Index of vectorized matrix, for lower and upper triangular part.
lower <- ((i - 1) * n) + j
upper <- ((j - 1) * n) + i
# Create all pairewise differences of rows of `X`.
X_diff <- X[i, , drop = F] - X[j, , drop = F]
# Identity matrix.
I_p <- diag(1, p)
# Init tracking of current best (according multiple attempts).
V.best <- NULL
loss.best <- Inf
# Start loop for multiple attempts.
for (attempt in 1:attempts) {
# Reset learning rate `tau`.
tau <- tau.init
# Sample a `(p, q)` dimensional matrix from the stiefel manifold as
# optimization start value.
V <- rStiefl(p, q)
# Initial loss and gradient.
loss <- Inf
G <- grad(X, Y, V, h, loss.out = TRUE, persistent = TRUE)
# Set last loss (aka, loss after applying the step).
loss.last <- loss
# Cayley transform matrix `A`
A <- (G %*% t(V)) - (V %*% t(G))
# Call logger with initial values before starting optimization.
if (is.function(logger)) {
logger(0L, attempt, loss, V, tau)
}
## Start optimization loop.
for (epoch in 1:epochs) {
# Apply learning rate `tau`.
A.tau <- tau * A
# Parallet transport (on Stiefl manifold) into direction of `G`.
V.tau <- solve(I_p + A.tau) %*% ((I_p - A.tau) %*% V)
# Loss at position after a step.
loss <- grad(X, Y, V.tau, h, loss.only = TRUE, persistent = TRUE)
# Check if step is appropriate, iff not reduce learning rate.
if ((loss - loss.last) > slack * loss.last) {
tau <- tau / 2
next() # Keep position and try with smaller `tau`.
}
# Compute error.
error <- norm(V %*% t(V) - V.tau %*% t(V.tau), type = "F")
# Check break condition (epoch check to skip ignored gradient calc).
# Note: the devision by `sqrt(2 * k)` is included in `tol`.
if (error < tol || epoch >= epochs) {
# take last step and stop optimization.
V <- V.tau
# Call logger last time befor stoping.
if (is.function(logger)) {
logger(epoch, attempt, loss, V, tau)
}
break()
}
# Perform the step and remember previous loss.
V <- V.tau
loss.last <- loss
# Call logger after taking a step.
if (is.function(logger)) {
logger(epoch, attempt, loss, V, tau)
}
# Compute gradient at new position.
G <- grad(X, Y, V, h, persistent = TRUE)
# Cayley transform matrix `A`
A <- (G %*% t(V)) - (V %*% t(G))
}
# Check if current attempt improved previous ones
if (loss < loss.best) {
loss.best <- loss
V.best <- V
}
}
return(list(
loss = loss.best,
V = V.best,
B = null(V.best),
h = h
))
}

View File

@ -1,109 +0,0 @@
#' Generates test datasets.
#'
#' Provides sample datasets. There are 5 different datasets named
#' M1, M2, M3, M4 and M5 described in the paper references below.
#' The general model is given by:
#' \deqn{Y ~ g(B'X) + \epsilon}
#'
#' @param name One of \code{"M1"}, \code{"M2"}, \code{"M3"}, \code{"M4"} or \code{"M5"}
#' @param n nr samples
#' @param p Dim. of random variable \code{X}.
#' @param p.mix Only for \code{"M4"}, see: below.
#' @param lambda Only for \code{"M4"}, see: below.
#'
#' @return List with elements
#' \itemize{
#' \item{X}{data}
#' \item{Y}{response}
#' \item{B}{Used dim-reduction matrix}
#' \item{name}{Name of the dataset (name parameter)}
#' }
#'
#' @section M1:
#' The data follows \eqn{X\sim N_p(0, \Sigma)}{X ~ N_p(0, Sigma)} for a subspace
#' dimension of \eqn{k = 2} with a default of \eqn{n = 200} data points.
#' The link function \eqn{g} is given as
#' \deqn{g(x) = \frac{x_1}{0.5 + (x_2 + 1.5)^2} + 0.5\epsilon}{g(x) = x_1 / (0.5 + (x_2 + 1.5)^2) + 0.5 epsilon}
#' @section M2:
#' \eqn{X\sim N_p(0, \Sigma)}{X ~ N_p(0, Sigma)} with \eqn{k = 2} with a default of \eqn{n = 200} data points.
#' The link function \eqn{g} is given as
#' \deqn{g(x) = x_1 x_2^2 + 0.5\epsilon}{g(x) = x_1 x_2^2 + 0.5 epsilon}
#' @section M3:
#' TODO:
#' @section M4:
#' TODO:
#' @section M5:
#' TODO:
#'
#' @import stats
#' @importFrom stats rnorm rbinom
#' @export
dataset <- function(name = "M1", n, B, p.mix = 0.3, lambda = 1.0) {
# validate parameters
stopifnot(name %in% c("M1", "M2", "M3", "M4", "M5"))
# set default values if not supplied
if (missing(n)) {
n <- if (name %in% c("M1", "M2")) 200 else if (name != "M5") 100 else 42
}
if (missing(B)) {
p <- 12
if (name == "M1") {
B <- cbind(
c( 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0),
c( 1,-1, 1,-1, 1,-1, 0, 0, 0, 0, 0, 0)
) / sqrt(6)
} else if (name == "M2") {
B <- cbind(
c(c(1, 0), rep(0, 10)),
c(c(0, 1), rep(0, 10))
)
} else {
B <- matrix(c(rep(1 / sqrt(6), 6), rep(0, 6)), 12, 1)
}
} else {
p <- dim(B)[1]
# validate col. nr to match dataset `k = dim(B)[2]`
stopifnot(
name %in% c("M1", "M2") && dim(B)[2] == 2,
name %in% c("M3", "M4", "M5") && dim(B)[2] == 1
)
}
# set link function `g` for model `Y ~ g(B'X) + epsilon`
if (name == "M1") {
g <- function(BX) { BX[1] / (0.5 + (BX[2] + 1.5)^2) }
} else if (name == "M2") {
g <- function(BX) { BX[1] * BX[2]^2 }
} else if (name %in% c("M3", "M4")) {
g <- function(BX) { cos(BX[1]) }
} else { # name == "M5"
g <- function(BX) { 2 * log(abs(BX[1]) + 1) }
}
# compute X
if (name != "M4") {
# compute root of the covariance matrix according the dataset
if (name %in% c("M1", "M3")) {
# Variance-Covariance structure for `X ~ N_p(0, \Sigma)` with
# `\Sigma_{i, j} = 0.5^{|i - j|}`.
Sigma <- matrix(0.5^abs(kronecker(1:p, 1:p, '-')), p, p)
# decompose Sigma to Sigma.root^T Sigma.root = Sigma for usage in creation of `X`
Sigma.root <- chol(Sigma)
} else { # name %in% c("M2", "M5")
Sigma.root <- diag(rep(1, p)) # d-dim identity
}
# data `X` as multivariate random normal variable with
# variance matrix `Sigma`.
X <- replicate(p, rnorm(n, 0, 1)) %*% Sigma.root
} else { # name == "M4"
X <- t(replicate(100, rep((1 - 2 * rbinom(1, 1, p.mix)) * lambda, p) + rnorm(p, 0, 1)))
}
# responce `y ~ g(B'X) + epsilon` with `epsilon ~ N(0, 1 / 2)`
Y <- apply(X, 1, function(X_i) {
g(t(B) %*% X_i) + rnorm(1, 0, 0.5)
})
return(list(X = X, Y = Y, B = B, name = name))
}

View File

@ -1,27 +0,0 @@
#' Estimated bandwidth for CVE.
#'
#' Estimates a propper bandwidth \code{h} according
#' \deqn{%
#' h = \chi_{p-q}^{-1}\left(\frac{nObs - 1}{n-1}\right)\frac{2 tr(\Sigma)}{p}}{%
#' h = qchisq( (nObs - 1)/(n - 1), p - q ) 2 tr(Sigma) / p}
#'
#' @param X data matrix of dimension (n x p) with n data points X_i of dimension
#' q. Therefor each row represents a datapoint of dimension p.
#' @param k Guess for rank(B).
#' @param nObs Ether numeric of a function. If specified as numeric value
#' its used in the computation of the bandwidth directly. If its a function
#' `nObs` is evaluated as \code{nObs(nrow(x))}. The default behaviou if not
#' supplied at all is to use \code{nObs <- nrow(x)^0.5}.
#'
#' @seealso [\code{\link{qchisq}}]
#' @export
estimate.bandwidth <- function(X, k, nObs) {
n <- nrow(X)
p <- ncol(X)
X_centered <- scale(X, center=TRUE, scale=FALSE)
Sigma <- (1 / n) * t(X_centered) %*% X_centered
quantil <- qchisq((nObs - 1) / (n - 1), k)
return(2 * quantil * sum(diag(Sigma)) / p)
}

View File

@ -1,82 +0,0 @@
#' Compute get gradient of `L(V)` given a dataset `X`.
#'
#' @param X Data matrix.
#' @param Y Responce.
#' @param V Position to compute the gradient at, aka point on Stiefl manifold.
#' @param h Bandwidth
#' @param loss.out Iff \code{TRUE} loss will be written to parent environment.
#' @param loss.only Boolean to only compute the loss, of \code{TRUE} a single
#' value loss is returned and \code{envir} is ignored.
#' @param persistent Determines if data indices and dependent calculations shall
#' be reused from the parent environment. ATTENTION: Do NOT set this flag, only
#' intended for internal usage by carefully aligned functions!
#' @keywords internal
#' @export
grad <- function(X, Y, V, h,
loss.out = FALSE,
loss.only = FALSE,
persistent = FALSE) {
# Get number of samples and dimension.
n <- nrow(X)
p <- ncol(X)
if (!persistent) {
# Compute lookup indexes for symmetrie, lower/upper
# triangular parts and vectorization.
pair.index <- elem.pairs(seq(n))
i <- pair.index[1, ] # `i` indices of `(i, j)` pairs
j <- pair.index[2, ] # `j` indices of `(i, j)` pairs
# Index of vectorized matrix, for lower and upper triangular part.
lower <- ((i - 1) * n) + j
upper <- ((j - 1) * n) + i
# Create all pairewise differences of rows of `X`.
X_diff <- X[i, , drop = F] - X[j, , drop = F]
}
# Projection matrix onto `span(V)`
Q <- diag(1, p) - tcrossprod(V, V)
# Vectorized distance matrix `D`.
vecD <- colSums(tcrossprod(Q, X_diff)^2)
# Create Kernel matrix (aka. apply kernel to distances)
K <- matrix(1, n, n) # `exp(0) == 1`
K[lower] <- exp((-0.5 / h) * vecD^2) # Set lower tri. part
K[upper] <- t(K)[upper] # Mirror lower tri. to upper
# Weighted `Y` momentums
colSumsK <- colSums(K)
y1 <- (K %*% Y) / colSumsK
y2 <- (K %*% Y^2) / colSumsK
# Per example loss `L(V, X_i)`
L <- y2 - y1^2
if (loss.only) {
return(mean(L))
}
if (loss.out) {
loss <<- mean(L)
}
# Compute scaling vector `vecS` for `X_diff`.
tmp <- kronecker(matrix(y1, n, 1), matrix(Y, 1, n), `-`)^2
tmp <- as.vector(L) - tmp
tmp <- tmp * K / colSumsK
vecS <- (tmp + t(tmp))[lower] * vecD
# The gradient.
# 1. The `crossprod(A, B)` is equivalent to `t(A) %*% B`,
# 2. `(X_diff %*% V) * vecS` is first a marix matrix mult. and then using
# recycling to scale each row with the values of `vecS`.
# Note that `vecS` is a vector and that `R` uses column-major ordering
# of matrices.
# (See: notes for more details)
# TODO: Depending on n, p, q decide which version to take (for current
# datasets "inner" is faster, see: notes).
# inner = crossprod(X_diff, X_diff * vecS) %*% V,
# outer = crossprod(X_diff, (X_diff %*% V) * vecS)
G <- crossprod(X_diff, X_diff * vecS) %*% V
G <- (-2 / (n * h^2)) * G
return(G)
}

View File

@ -1,43 +0,0 @@
#' Performs a grid search for parameters over a parameter grid.
#' @examples
#' args <- list(
#' h = c(0.05, 0.1, 0.2),
#' method = c("simple", "sgd"),
#' tau = c(0.5, 0.1, 0.01)
#' )
#' cve.grid.search(args)
#' @export
cve.grid.search <- function(X, Y, k, args) {
args$stringsAsFactors = FALSE
args$KEEP.OUT.ATTRS = FALSE
grid <- do.call(expand.grid, args)
grid.length <- length(grid[[1]])
print(grid)
for (i in 1:grid.length) {
arguments <- as.list(grid[i, ])
# Set required arguments
arguments$X <- X
arguments$Y <- Y
arguments$k <- k
# print(arguments)
dr <- do.call(cve.call, arguments)
print(dr$loss)
}
}
# ds <- dataset()
# X <- ds$X
# Y <- ds$Y
# (k <- ncol(ds$B))
# args <- list(
# h = c(0.05, 0.1, 0.2),
# method = c("simple", "sgd"),
# tau = c(0.5, 0.1, 0.01),
# attempts = c(1L)
# )
# cve.grid.search(X, Y, k, args)

View File

@ -1,82 +0,0 @@
#' Samples uniform from the Stiefl Manifold.
#'
#' @param p row dim.
#' @param q col dim.
#' @return `(p, q)` semi-orthogonal matrix
#' @examples
#' V <- rStiefel(6, 4)
#' @export
rStiefl <- function(p, q) {
return(qr.Q(qr(matrix(rnorm(p * q, 0, 1), p, q))))
}
#' Retraction to the manifold.
#'
#' @param A matrix.
#' @return `(p, q)` semi-orthogonal matrix, aka element of the Stiefl manifold.
#' @keywords internal
#' @export
retractStiefl <- function(A) {
return(qr.Q(qr(A)))
}
#' Skew-Symmetric matrix computed from `A` as
#' \eqn{1/2 (A - A^T)}.
#' @param A Matrix of dim `(p, q)`
#' @return Skew-Symmetric matrix of dim `(p, p)`.
#' @keywords internal
#' @export
skew <- function(A) {
0.5 * (A - t(A))
}
#' Symmetric matrix computed from `A` as
#' \eqn{1/2 (A + A^T)}.
#' @param A Matrix of dim `(p, q)`
#' @return Symmetric matrix of dim `(p, p)`.
#' @keywords internal
#' @export
sym <- function(A) {
0.5 * (A + t(A))
}
#' Orthogonal Projection onto the tangent space of the stiefl manifold.
#'
#' @param V Point on the stiefl manifold.
#' @param G matrix to be projected onto the tangent space at `V`.
#' @return `(p, q)` matrix as element of the tangent space at `V`.
#' @keywords internal
#' @export
projTangentStiefl <- function(V, G) {
Q <- diag(1, nrow(V)) - V %*% t(V)
return(Q %*% G + V %*% skew(t(V) %*% G))
}
#' Null space basis of given matrix `V`
#'
#' @param V `(p, q)` matrix
#' @return Semi-orthogonal `(p, p - q)` matrix spaning the null space of `V`.
#' @keywords internal
#' @export
null <- function(V) {
tmp <- qr(V)
set <- if(tmp$rank == 0L) seq_len(ncol(V)) else -seq_len(tmp$rank)
return(qr.Q(tmp, complete=TRUE)[, set, drop=FALSE])
}
#' Creates a (numeric) matrix where each column contains
#' an element to element matching.
#' @param elements numeric vector of elements to match
#' @return matrix of size `(2, n * (n - 1) / 2)` for a argument of lenght `n`.
#' @keywords internal
#' @examples
#' elem.pairs(seq.int(2, 5))
#' @export
elem.pairs <- function(elements) {
# Number of elements to match.
n <- length(elements)
# Create all combinations.
pairs <- rbind(rep(elements, each=n), rep(elements, n))
# Select unique combinations without self interaction.
return(pairs[, pairs[1, ] < pairs[2, ]])
}

View File

@ -1,2 +0,0 @@
runtime_test Runtime comparison of CVE against MAVE for M1 - M5 datasets.
logging Example of a logger function for cve algorithm analysis.

View File

@ -1,43 +0,0 @@
library(CVEpureR)
# Setup histories.
(epochs <- 50)
(attempts <- 10)
loss.history <- matrix(NA, epochs + 1, attempts)
error.history <- matrix(NA, epochs + 1, attempts)
tau.history <- matrix(NA, epochs + 1, attempts)
true.error.history <- matrix(NA, epochs + 1, attempts)
# Create a dataset
ds <- dataset("M1")
X <- ds$X
Y <- ds$Y
B <- ds$B # the true `B`
(k <- ncol(ds$B))
# True projection matrix.
P <- B %*% solve(t(B) %*% B) %*% t(B)
# Define the logger for the `cve()` method.
logger <- function(env) {
# Note the `<<-` assignement!
loss.history[env$epoch + 1, env$attempt] <<- env$loss
error.history[env$epoch + 1, env$attempt] <<- env$error
tau.history[env$epoch + 1, env$attempt] <<- env$tau
# Compute true error by comparing to the true `B`
B.est <- null(env$V) # Function provided by CVE
P.est <- B.est %*% solve(t(B.est) %*% B.est) %*% t(B.est)
true.error <- norm(P - P.est, 'F') / sqrt(2 * k)
true.error.history[env$epoch + 1, env$attempt] <<- true.error
}
# Performe SDR for ONE `k`.
dr <- cve(Y ~ X, k = k, logger = logger, epochs = epochs, attempts = attempts)
# Plot history's
par(mfrow = c(2, 2))
matplot(loss.history, type = 'l', log = 'y', xlab = 'iter',
main = 'loss', ylab = expression(L(V[iter])))
matplot(error.history, type = 'l', log = 'y', xlab = 'iter',
main = 'error', ylab = 'error')
matplot(tau.history, type = 'l', log = 'y', xlab = 'iter',
main = 'tau', ylab = 'tau')
matplot(true.error.history, type = 'l', log = 'y', xlab = 'iter',
main = 'true error', ylab = 'true error')

View File

@ -1,89 +0,0 @@
# Usage:
# ~$ Rscript runtime_test.R
library(CVEpureR) # load CVE
#' Writes progress to console.
tell.user <- function(name, start.time, i, length) {
cat("\rRunning Test (", name, "):",
i, "/", length,
" - elapsed:", format(Sys.time() - start.time), "\033[K")
}
subspace.dist <- function(B1, B2){
P1 <- B1 %*% solve(t(B1) %*% B1) %*% t(B1)
P2 <- B2 %*% solve(t(B2) %*% B2) %*% t(B2)
return(norm(P1 - P2, type = 'F'))
}
# Number of simulations
SIM.NR <- 50
# maximal number of iterations in curvilinear search algorithm
MAXIT <- 50
# number of arbitrary starting values for curvilinear optimization
ATTEMPTS <- 10
# set names of datasets
dataset.names <- c("M1", "M2", "M3", "M4", "M5")
# Set used CVE method
methods <- c("simple") # c("legacy", "simple", "sgd", "linesearch")
# Setup error and time tracking variables
error <- matrix(NA, SIM.NR, length(methods) * length(dataset.names))
time <- matrix(NA, SIM.NR, ncol(error))
colnames(error) <- kronecker(paste0(dataset.names, '-'), methods, paste0)
colnames(time) <- colnames(error)
# only for telling user (to stdout)
count <- 0
start.time <- Sys.time()
# Start simulation loop.
for (sim in 1:SIM.NR) {
# Repeat for each dataset.
for (name in dataset.names) {
count <- count + 1
tell.user(name, start.time, count, SIM.NR * length(dataset.names))
# Create a new dataset
ds <- dataset(name)
# Prepare X, Y and combine to data matrix
Y <- ds$Y
X <- ds$X
data <- cbind(Y, X)
# get dimensions
dim <- ncol(X)
truedim <- ncol(ds$B)
for (method in methods) {
dr.time <- system.time(
dr <- cve.call(X, Y,
method = method,
k = truedim,
attempts = ATTEMPTS
)
)
dr <- dr[[truedim]]
key <- paste0(name, '-', method)
error[sim, key] <- subspace.dist(dr$B, ds$B) / sqrt(2 * truedim)
time[sim, key] <- dr.time["elapsed"]
}
}
}
cat("\n\n## Time [sec] Means:\n")
print(colMeans(time))
cat("\n## Error Means:\n")
print(colMeans(error))
at <- seq(ncol(error)) + rep(seq(ncol(error) / length(methods)) - 1, each = length(methods))
boxplot(error,
main = paste0("Error (Nr of simulations ", SIM.NR, ")"),
ylab = "Error",
las = 2,
at = at
)
boxplot(time,
main = paste0("Time (Nr of simulations ", SIM.NR, ")"),
ylab = "Time [sec]",
las = 2,
at = at
)

Binary file not shown.

View File

@ -1,20 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/CVE.R
\docType{package}
\name{CVEpureR-package}
\alias{CVEpureR}
\alias{CVEpureR-package}
\title{Conditional Variance Estimator (CVE)}
\description{
Conditional Variance Estimator for Sufficient Dimension
Reduction
}
\details{
TODO: And some details
}
\references{
Fertl Likas, Bura Efstathia. Conditional Variance Estimation for Sufficient Dimension Reduction, 2019
}
\author{
Loki
}

View File

@ -1,71 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/CVE.R
\name{cve}
\alias{cve}
\alias{cve.call}
\title{Implementation of the CVE method.}
\usage{
cve(formula, data, method = "simple", max.dim = 10, ...)
cve.call(X, Y, method = "simple", nObs = nrow(X)^0.5, min.dim = 1,
max.dim = 10, k, ...)
}
\arguments{
\item{formula}{Formel for the regression model defining `X`, `Y`.
See: \code{\link{formula}}.}
\item{data}{data.frame holding data for formula.}
\item{method}{The different only differe in the used optimization.
All of them are Gradient based optimization on a Stiefel manifold.
\itemize{
\item "simple" Simple reduction of stepsize.
\item "sgd" stocastic gradient decent.
\item TODO: further
}}
\item{...}{Further parameters depending on the used method.}
\item{X}{Data}
\item{Y}{Responces}
\item{nObs}{as described in the Paper.}
\item{k}{guess for SDR dimension.}
\item{nObs}{Like in the paper.}
\item{...}{Method specific parameters.}
}
\description{
Conditional Variance Estimator (CVE) is a novel sufficient dimension
reduction (SDR) method assuming a model
\deqn{Y \sim g(B'X) + \epsilon}{Y ~ g(B'X) + epsilon}
where B'X is a lower dimensional projection of the predictors.
}
\examples{
library(CVE)
# sample dataset
ds <- dataset("M5")
# call ´cve´ with default method (aka "simple")
dr.simple <- cve(ds$Y ~ ds$X, k = ncol(ds$B))
# plot optimization history (loss via iteration)
plot(dr.simple, main = "CVE M5 simple")
# call ´cve´ with method "linesearch" using ´data.frame´ as data.
data <- data.frame(Y = ds$Y, X = ds$X)
# Note: ´Y, X´ are NOT defined, they are extracted from ´data´.
dr.linesearch <- cve(Y ~ ., data, method = "linesearch", k = ncol(ds$B))
plot(dr.linesearch, main = "CVE M5 linesearch")
}
\references{
Fertl L., Bura E. Conditional Variance Estimation for Sufficient Dimension Reduction, 2019
}
\seealso{
\code{\link{formula}}. For a complete parameters list (dependent on
the method) see \code{\link{cve_simple}}, \code{\link{cve_sgd}}
}

View File

@ -1,19 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/gridSearch.R
\name{cve.grid.search}
\alias{cve.grid.search}
\title{Performs a grid search for parameters over a parameter grid.}
\usage{
cve.grid.search(X, Y, k, args)
}
\description{
Performs a grid search for parameters over a parameter grid.
}
\examples{
args <- list(
h = c(0.05, 0.1, 0.2),
method = c("simple", "sgd"),
tau = c(0.5, 0.1, 0.01)
)
cve.grid.search(args)
}

View File

@ -1,16 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/cve_linesearch.R
\name{cve_linesearch}
\alias{cve_linesearch}
\title{Implementation of the CVE method using curvilinear linesearch with Armijo-Wolfe
conditions.}
\usage{
cve_linesearch(X, Y, k, nObs = sqrt(nrow(X)), h = NULL, tau = 1,
tol = 0.001, rho1 = 0.1, rho2 = 0.9, slack = 0, epochs = 50L,
attempts = 10L, max.linesearch.iter = 10L, logger = NULL)
}
\description{
Implementation of the CVE method using curvilinear linesearch with Armijo-Wolfe
conditions.
}
\keyword{internal}

View File

@ -1,16 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/cve_sgd.R
\name{cve_sgd}
\alias{cve_sgd}
\title{Simple implementation of the CVE method. 'Simple' means that this method is
a classic GD method unsing no further tricks.}
\usage{
cve_sgd(X, Y, k, nObs = sqrt(nrow(X)), h = NULL, tau = 0.01,
tol = 0.001, epochs = 50L, batch.size = 16L, attempts = 10L,
logger = NULL)
}
\description{
Simple implementation of the CVE method. 'Simple' means that this method is
a classic GD method unsing no further tricks.
}
\keyword{internal}

View File

@ -1,16 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/cve_simple.R
\name{cve_simple}
\alias{cve_simple}
\title{Simple implementation of the CVE method. 'Simple' means that this method is
a classic GD method unsing no further tricks.}
\usage{
cve_simple(X, Y, k, nObs = sqrt(nrow(X)), h = NULL, tau = 1,
tol = 0.001, slack = 0, epochs = 50L, attempts = 10L,
logger = NULL)
}
\description{
Simple implementation of the CVE method. 'Simple' means that this method is
a classic GD method unsing no further tricks.
}
\keyword{internal}

View File

@ -1,64 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/datasets.R
\name{dataset}
\alias{dataset}
\title{Generates test datasets.}
\usage{
dataset(name = "M1", n, B, p.mix = 0.3, lambda = 1)
}
\arguments{
\item{name}{One of \code{"M1"}, \code{"M2"}, \code{"M3"}, \code{"M4"} or \code{"M5"}}
\item{n}{nr samples}
\item{p.mix}{Only for \code{"M4"}, see: below.}
\item{lambda}{Only for \code{"M4"}, see: below.}
\item{p}{Dim. of random variable \code{X}.}
}
\value{
List with elements
\itemize{
\item{X}{data}
\item{Y}{response}
\item{B}{Used dim-reduction matrix}
\item{name}{Name of the dataset (name parameter)}
}
}
\description{
Provides sample datasets. There are 5 different datasets named
M1, M2, M3, M4 and M5 described in the paper references below.
The general model is given by:
\deqn{Y ~ g(B'X) + \epsilon}
}
\section{M1}{
The data follows \eqn{X\sim N_p(0, \Sigma)}{X ~ N_p(0, Sigma)} for a subspace
dimension of \eqn{k = 2} with a default of \eqn{n = 200} data points.
The link function \eqn{g} is given as
\deqn{g(x) = \frac{x_1}{0.5 + (x_2 + 1.5)^2} + 0.5\epsilon}{g(x) = x_1 / (0.5 + (x_2 + 1.5)^2) + 0.5 epsilon}
}
\section{M2}{
\eqn{X\sim N_p(0, \Sigma)}{X ~ N_p(0, Sigma)} with \eqn{k = 2} with a default of \eqn{n = 200} data points.
The link function \eqn{g} is given as
\deqn{g(x) = x_1 x_2^2 + 0.5\epsilon}{g(x) = x_1 x_2^2 + 0.5 epsilon}
}
\section{M3}{
TODO:
}
\section{M4}{
TODO:
}
\section{M5}{
TODO:
}

View File

@ -1,23 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/util.R
\name{elem.pairs}
\alias{elem.pairs}
\title{Creates a (numeric) matrix where each column contains
an element to element matching.}
\usage{
elem.pairs(elements)
}
\arguments{
\item{elements}{numeric vector of elements to match}
}
\value{
matrix of size `(2, n * (n - 1) / 2)` for a argument of lenght `n`.
}
\description{
Creates a (numeric) matrix where each column contains
an element to element matching.
}
\examples{
elem.pairs(seq.int(2, 5))
}
\keyword{internal}

View File

@ -1,28 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/estimateBandwidth.R
\name{estimate.bandwidth}
\alias{estimate.bandwidth}
\title{Estimated bandwidth for CVE.}
\usage{
estimate.bandwidth(X, k, nObs)
}
\arguments{
\item{X}{data matrix of dimension (n x p) with n data points X_i of dimension
q. Therefor each row represents a datapoint of dimension p.}
\item{k}{Guess for rank(B).}
\item{nObs}{Ether numeric of a function. If specified as numeric value
its used in the computation of the bandwidth directly. If its a function
`nObs` is evaluated as \code{nObs(nrow(x))}. The default behaviou if not
supplied at all is to use \code{nObs <- nrow(x)^0.5}.}
}
\description{
Estimates a propper bandwidth \code{h} according
\deqn{%
h = \chi_{p-q}^{-1}\left(\frac{nObs - 1}{n-1}\right)\frac{2 tr(\Sigma)}{p}}{%
h = qchisq( (nObs - 1)/(n - 1), p - q ) 2 tr(Sigma) / p}
}
\seealso{
[\code{\link{qchisq}}]
}

View File

@ -1,31 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/gradient.R
\name{grad}
\alias{grad}
\title{Compute get gradient of `L(V)` given a dataset `X`.}
\usage{
grad(X, Y, V, h, loss.out = FALSE, loss.only = FALSE,
persistent = FALSE)
}
\arguments{
\item{X}{Data matrix.}
\item{Y}{Responce.}
\item{V}{Position to compute the gradient at, aka point on Stiefl manifold.}
\item{h}{Bandwidth}
\item{loss.out}{Iff \code{TRUE} loss will be written to parent environment.}
\item{loss.only}{Boolean to only compute the loss, of \code{TRUE} a single
value loss is returned and \code{envir} is ignored.}
\item{persistent}{Determines if data indices and dependent calculations shall
be reused from the parent environment. ATTENTION: Do NOT set this flag, only
intended for internal usage by carefully aligned functions!}
}
\description{
Compute get gradient of `L(V)` given a dataset `X`.
}
\keyword{internal}

View File

@ -1,18 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/util.R
\name{null}
\alias{null}
\title{Null space basis of given matrix `V`}
\usage{
null(V)
}
\arguments{
\item{V}{`(p, q)` matrix}
}
\value{
Semi-orthogonal `(p, p - q)` matrix spaning the null space of `V`.
}
\description{
Null space basis of given matrix `V`
}
\keyword{internal}

View File

@ -1,28 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/CVE.R
\name{plot.cve}
\alias{plot.cve}
\title{Ploting helper for objects of class \code{cve}.}
\usage{
## S3 method for class 'cve'
plot(x, content = "history", ...)
}
\arguments{
\item{x}{Object of class \code{cve} (result of [cve()]).}
\item{...}{Pass through parameters to [plot()] and [lines()]}
\item{content}{Specifies what to plot:
\itemize{
\item "history" Plots the loss history from stiefel optimization
(default).
\item ... TODO: add (if there are any)
}}
}
\description{
Ploting helper for objects of class \code{cve}.
}
\seealso{
see \code{\link{par}} for graphical parameters to pass through
as well as \code{\link{plot}} for standard plot utility.
}

View File

@ -1,22 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/util.R
\name{rStiefl}
\alias{rStiefl}
\title{Samples uniform from the Stiefel Manifold}
\usage{
rStiefl(p, q)
}
\arguments{
\item{p}{row dim.}
\item{q}{col dim.}
}
\value{
`(p, q)` semi-orthogonal matrix
}
\description{
Samples uniform from the Stiefel Manifold
}
\examples{
V <- rStiefel(6, 4)
}

View File

@ -1,14 +0,0 @@
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/CVE.R
\name{summary.cve}
\alias{summary.cve}
\title{Prints a summary of a \code{cve} result.}
\usage{
\method{summary}{cve}(object, ...)
}
\arguments{
\item{object}{Instance of 'cve' as return of \code{cve}.}
}
\description{
Prints a summary of a \code{cve} result.
}

View File

@ -1,479 +0,0 @@
library(microbenchmark)
dyn.load("benchmark.so")
## rowSum* .call --------------------------------------------------------------
rowSums.c <- function(M) {
stopifnot(
is.matrix(M),
is.numeric(M)
)
if (!is.double(M)) {
M <- matrix(as.double(M), nrow = nrow(M))
}
.Call('R_rowSums', PACKAGE = 'benchmark', M)
}
rowSumsV2.c <- function(M) {
stopifnot(
is.matrix(M),
is.numeric(M)
)
if (!is.double(M)) {
M <- matrix(as.double(M), nrow = nrow(M))
}
.Call('R_rowSumsV2', PACKAGE = 'benchmark', M)
}
rowSumsV3.c <- function(M) {
stopifnot(
is.matrix(M),
is.numeric(M)
)
if (!is.double(M)) {
M <- matrix(as.double(M), nrow = nrow(M))
}
.Call('R_rowSumsV3', PACKAGE = 'benchmark', M)
}
colSums.c <- function(M) {
stopifnot(
is.matrix(M),
is.numeric(M)
)
if (!is.double(M)) {
M <- matrix(as.double(M), nrow = nrow(M))
}
.Call('R_colSums', PACKAGE = 'benchmark', M)
}
rowSquareSums.c <- function(M) {
stopifnot(
is.matrix(M),
is.numeric(M)
)
if (!is.double(M)) {
M <- matrix(as.double(M), nrow = nrow(M))
}
.Call('R_rowSquareSums', PACKAGE = 'benchmark', M)
}
rowSumsSymVec.c <- function(vecA, nrow, diag = 0.0) {
stopifnot(
is.vector(vecA),
is.numeric(vecA),
is.numeric(diag),
nrow * (nrow - 1) == length(vecA) * 2
)
if (!is.double(vecA)) {
vecA <- as.double(vecA)
}
.Call('R_rowSumsSymVec', PACKAGE = 'benchmark',
vecA, as.integer(nrow), as.double(diag))
}
rowSweep.c <- function(A, v, op = '-') {
stopifnot(
is.matrix(A),
is.numeric(v)
)
if (!is.double(A)) {
A <- matrix(as.double(A), nrow = nrow(A))
}
if (!is.vector(v) || !is.double(v)) {
v <- as.double(v)
}
stopifnot(
nrow(A) == length(v),
op %in% c('+', '-', '*', '/')
)
.Call('R_rowSweep', PACKAGE = 'benchmark', A, v, op)
}
## row*, col* tests ------------------------------------------------------------
n <- 3000
M <- matrix(runif(n * 12), n, 12)
stopifnot(
all.equal(rowSums(M^2), rowSums.c(M^2)),
all.equal(colSums(M), colSums.c(M)),
all.equal(rowSums(M), rowSumsV2.c(M)),
all.equal(rowSums(M), rowSumsV3.c(M))
)
microbenchmark(
rowSums = rowSums(M),
rowSums.c = rowSums.c(M),
rowSumsV2.c = rowSumsV2.c(M),
rowSumsV3.c = rowSumsV3.c(M)
)
microbenchmark(
rowSums = rowSums(M^2),
rowSums.c = rowSums.c(M^2),
rowSqSums.c = rowSquareSums.c(M)
)
microbenchmark(
colSums = colSums(M),
colSums.c = colSums.c(M)
)
sum = rowSums(M)
stopifnot(all.equal(
sweep(M, 1, sum, FUN = `/`),
rowSweep.c(M, sum, '/') # Col-Normalize)
), all.equal(
sweep(M, 1, sum, FUN = `/`),
M / sum
))
microbenchmark(
sweep = sweep(M, 1, sum, FUN = `/`),
M / sum,
rowSweep.c = rowSweep.c(M, sum, '/') # Col-Normalize)
)
# Create symmetric matrix with constant diagonal entries.
nrow <- 200
diag <- 1.0
Sym <- tcrossprod(runif(nrow))
diag(Sym) <- diag
# Get vectorized lower triangular part of `Sym` matrix.
SymVec <- Sym[lower.tri(Sym)]
stopifnot(all.equal(
rowSums(Sym),
rowSumsSymVec.c(SymVec, nrow, diag)
))
microbenchmark(
rowSums = rowSums(Sym),
rowSums.c = rowSums.c(Sym),
rowSumsSymVec.c = rowSumsSymVec.c(SymVec, nrow, diag)
)
## Matrix-Matrix opperation .call ---------------------------------------------
transpose.c <- function(A) {
stopifnot(
is.matrix(A), is.numeric(A)
)
if (!is.double(A)) {
A <- matrix(as.double(A), nrow(A), ncol(A))
}
.Call('R_transpose', PACKAGE = 'benchmark', A)
}
sympMV.c <- function(vecA, x) {
stopifnot(
is.vector(vecA), is.numeric(vecA),
is.vector(x), is.numeric(x),
length(x) * (length(x) + 1) == 2 * length(vecA)
)
if (!is.double(vecA)) {
vecA <- as.double(vecA)
}
if (!is.double(x)) {
x <- as.double(x)
}
.Call('R_sympMV', PACKAGE = 'benchmark', vecA, x)
}
matrixprod.c <- function(A, B) {
stopifnot(
is.matrix(A), is.numeric(A),
is.matrix(B), is.numeric(B),
ncol(A) == nrow(B)
)
if (!is.double(A)) {
A <- matrix(as.double(A), nrow = nrow(A))
}
if (!is.double(B)) {
B <- matrix(as.double(B), nrow = nrow(B))
}
.Call('R_matrixprod', PACKAGE = 'benchmark', A, B)
}
crossprod.c <- function(A, B) {
stopifnot(
is.matrix(A), is.numeric(A),
is.matrix(B), is.numeric(B),
nrow(A) == nrow(B)
)
if (!is.double(A)) {
A <- matrix(as.double(A), nrow = nrow(A))
}
if (!is.double(B)) {
B <- matrix(as.double(B), nrow = nrow(B))
}
.Call('R_crossprod', PACKAGE = 'benchmark', A, B)
}
kronecker.c <- function(A, B, op = '*') {
stopifnot(
is.matrix(A), is.numeric(A),
is.matrix(B), is.numeric(B),
is.character(op), op %in% c('*', '+', '/', '-')
)
if (!is.double(A)) {
A <- matrix(as.double(A), nrow = nrow(A))
}
if (!is.double(B)) {
B <- matrix(as.double(B), nrow = nrow(B))
}
.Call('R_kronecker', PACKAGE = 'benchmark', A, B, op)
}
skewSymRank2k.c <- function(A, B, alpha = 1, beta = 0) {
stopifnot(
is.matrix(A), is.numeric(A),
is.matrix(B), is.numeric(B),
all(dim(A) == dim(B)),
is.numeric(alpha), length(alpha) == 1L,
is.numeric(beta), length(beta) == 1L
)
if (!is.double(A)) {
A <- matrix(as.double(A), nrow = nrow(A))
}
if (!is.double(B)) {
B <- matrix(as.double(B), nrow = nrow(B))
}
.Call('R_skewSymRank2k', PACKAGE = 'benchmark', A, B,
as.double(alpha), as.double(beta))
}
## Matrix-Matrix opperation tests ---------------------------------------------
n <- 200
k <- 100
m <- 300
A <- matrix(runif(n * k), n, k)
B <- matrix(runif(k * m), k, m)
stopifnot(
all.equal(t(A), transpose.c(A))
)
microbenchmark(
t(A),
transpose.c(A)
)
Sym <- tcrossprod(runif(n))
vecSym <- Sym[lower.tri(Sym, diag = T)]
x <- runif(n)
stopifnot(all.equal(
as.double(Sym %*% x),
sympMV.c(vecSym, x)
))
microbenchmark(
Sym %*% x,
sympMV.c = sympMV.c(vecSym, x)
)
stopifnot(
all.equal(A %*% B, matrixprod.c(A, B))
)
microbenchmark(
"%*%" = A %*% B,
matrixprod.c = matrixprod.c(A, B)
)
A <- matrix(runif(k * n), k, n)
B <- matrix(runif(k * m), k, m)
stopifnot(
all.equal(crossprod(A, B), crossprod.c(A, B))
)
microbenchmark(
crossprod = crossprod(A, B),
crossprod.c = crossprod.c(A, B)
)
n <- 100L
m <- 12L
p <- 11L
q <- 10L
A <- matrix(runif(n * m), n, m)
B <- matrix(runif(p * q), p, q)
stopifnot(all.equal(
kronecker(A, B),
kronecker.c(A, B)
))
microbenchmark(
kronecker = kronecker(A, B),
kronecker.c = kronecker.c(A, B)
)
n <- 12
k <- 11
A <- matrix(runif(n * k), n, k)
B <- matrix(runif(n * k), n, k)
stopifnot(all.equal(
A %*% t(B) - B %*% t(A), skewSymRank2k.c(A, B)
))
microbenchmark(
A %*% t(B) - B %*% t(A),
skewSymRank2k.c(A, B)
)
## Orthogonal projection onto null space .Call --------------------------------
nullProj.c <- function(B) {
stopifnot(
is.matrix(B), is.numeric(B)
)
if (!is.double(B)) {
B <- matrix(as.double(B), nrow = nrow(B))
}
.Call('R_nullProj', PACKAGE = 'benchmark', B)
}
## Orthogonal projection onto null space tests --------------------------------
p <- 12
q <- 10
V <- qr.Q(qr(matrix(rnorm(p * q, 0, 1), p, q)))
# Projection matrix onto `span(V)`
Q <- diag(1, p) - tcrossprod(V, V)
stopifnot(
all.equal(Q, nullProj.c(V))
)
microbenchmark(
nullProj = diag(1, p) - tcrossprod(V, V),
nullProj.c = nullProj.c(V)
)
# ## Kronecker optimizations ----------------------------------------------------
# library(microbenchmark)
# dist.1 <- function(X_diff, Q) {
# rowSums((X_diff %*% Q)^2)
# }
# dist.2 <- function(X, Q) {
# ones <- rep(1, nrow(X))
# proj <- X %*% Q
# rowSums((kronecker(proj, ones) - kronecker(ones, proj))^2)
# }
# n <- 400L
# p <- 12L
# k <- 2L
# q <- p - k
# X <- matrix(rnorm(n * p), n, p)
# Q <- diag(1, p) - tcrossprod(rnorm(p))
# ones <- rep(1, n)
# X_diff <- kronecker(X, ones) - kronecker(ones, X)
# stopifnot(all.equal(dist.1(X_diff, Q), dist.2(X, Q)))
# microbenchmark(
# dist.1(X_diff, Q),
# dist.2(X, Q),
# times = 10L
# )
# # if (!persistent) {
# # pair.index <- elem.pairs(seq(n))
# # i <- pair.index[, 1] # `i` indices of `(i, j)` pairs
# # j <- pair.index[, 2] # `j` indices of `(i, j)` pairs
# # lower <- ((i - 1) * n) + j
# # upper <- ((j - 1) * n) + i
# # X_diff <- X[i, , drop = F] - X[j, , drop = F]
# # }
# # # Projection matrix onto `span(V)`
# # Q <- diag(1, p) - tcrossprod(V, V)
# # # Vectorized distance matrix `D`.
# # vecD <- rowSums((X_diff %*% Q)^2)
# ## WIP for gradient. ----------------------------------------------------------
grad.c <- function(X, X_diff, Y, V, h) {
stopifnot(
is.matrix(X), is.double(X),
is.matrix(X_diff), is.double(X_diff),
ncol(X_diff) == ncol(X), nrow(X_diff) == nrow(X) * (nrow(X) - 1) / 2,
is.vector(Y) || (is.matrix(Y) && pmin(dim(Y)) == 1L), is.double(Y),
length(Y) == nrow(X),
is.matrix(V), is.double(V),
nrow(V) == ncol(X),
is.vector(h), is.numeric(h), length(h) == 1
)
.Call('R_grad', PACKAGE = 'benchmark',
X, X_diff, as.double(Y), V, as.double(h));
}
elem.pairs <- function(elements) {
# Number of elements to match.
n <- length(elements)
# Create all combinations.
pairs <- rbind(rep(elements, each=n), rep(elements, n))
# Select unique combinations without self interaction.
return(pairs[, pairs[1, ] < pairs[2, ]])
}
grad <- function(X, Y, V, h, persistent = TRUE) {
n <- nrow(X)
p <- ncol(X)
if (!persistent) {
pair.index <- elem.pairs(seq(n))
i <- pair.index[, 1] # `i` indices of `(i, j)` pairs
j <- pair.index[, 2] # `j` indices of `(i, j)` pairs
lower <- ((i - 1) * n) + j
upper <- ((j - 1) * n) + i
X_diff <- X[i, , drop = F] - X[j, , drop = F]
}
# Projection matrix onto `span(V)`
Q <- diag(1, p) - tcrossprod(V, V)
# Vectorized distance matrix `D`.
vecD <- rowSums((X_diff %*% Q)^2)
# Create Kernel matrix (aka. apply kernel to distances)
K <- matrix(1, n, n) # `exp(0) == 1`
K[lower] <- exp((-0.5 / h) * vecD^2) # Set lower tri. part
K[upper] <- t(K)[upper] # Mirror lower tri. to upper
# Weighted `Y` momentums
colSumsK <- colSums(K)
y1 <- (K %*% Y) / colSumsK
y2 <- (K %*% Y^2) / colSumsK
# Per example loss `L(V, X_i)`
L <- y2 - y1^2
# Compute scaling vector `vecS` for `X_diff`.
tmp <- kronecker(matrix(y1, n, 1), matrix(Y, 1, n), `-`)^2
tmp <- as.vector(L) - tmp
tmp <- tmp * K / colSumsK
vecS <- (tmp + t(tmp))[lower] * vecD
G <- crossprod(X_diff, X_diff * vecS) %*% V
G <- (-2 / (n * h^2)) * G
return(G)
}
rStiefel <- function(p, q) {
return(qr.Q(qr(matrix(rnorm(p * q, 0, 1), p, q))))
}
n <- 200
p <- 12
q <- 10
X <- matrix(runif(n * p), n, p)
Y <- runif(n)
V <- rStiefel(p, q)
h <- 0.1
pair.index <- elem.pairs(seq(n))
i <- pair.index[1, ] # `i` indices of `(i, j)` pairs
j <- pair.index[2, ] # `j` indices of `(i, j)` pairs
lower <- ((i - 1) * n) + j
upper <- ((j - 1) * n) + i
X_diff <- X[i, , drop = F] - X[j, , drop = F]
stopifnot(all.equal(
grad(X, Y, V, h),
grad.c(X, X_diff, Y, V, h)
))
microbenchmark(
grad = grad(X, Y, V, h),
grad.c = grad.c(X, X_diff, Y, V, h)
)

View File

@ -1,510 +0,0 @@
#include <stdlib.h>
#include <string.h> // for `mem*` functions.
#include <R_ext/BLAS.h>
#include <R_ext/Lapack.h>
#include <R_ext/Error.h>
// #include <Rmath.h>
#include "benchmark.h"
void rowSums(const double *A,
const int nrow, const int ncol,
double *sum) {
int i, j, block_size, block_size_i;
const double *A_block = A;
const double *A_end = A + nrow * ncol;
if (nrow > 508) {
block_size = 508;
} else {
block_size = nrow;
}
// Iterate `(block_size_i, ncol)` submatrix blocks.
for (i = 0; i < nrow; i += block_size_i) {
// Reset `A` to new block beginning.
A = A_block;
// Take block size of eveything left and reduce to max size.
block_size_i = nrow - i;
if (block_size_i > block_size) {
block_size_i = block_size;
}
// Copy blocks first column.
for (j = 0; j < block_size_i; j += 4) {
sum[j] = A[j];
sum[j + 1] = A[j + 1];
sum[j + 2] = A[j + 2];
sum[j + 3] = A[j + 3];
}
for (; j < block_size_i; ++j) {
sum[j] = A[j];
}
// Sum following columns to the first one.
for (A += nrow; A < A_end; A += nrow) {
for (j = 0; j < block_size_i; j += 4) {
sum[j] += A[j];
sum[j + 1] += A[j + 1];
sum[j + 2] += A[j + 2];
sum[j + 3] += A[j + 3];
}
for (; j < block_size_i; ++j) {
sum[j] += A[j];
}
}
// Step one block forth.
A_block += block_size_i;
sum += block_size_i;
}
}
void rowSumsV2(const double *A,
const int nrow, const int ncol,
double *sum) {
int i, j, block_size, block_size_i;
const double *A_block = A;
const double *A_end = A + nrow * ncol;
if (nrow > CVE_MEM_CHUNK_SIZE) {
block_size = CVE_MEM_CHUNK_SIZE;
} else {
block_size = nrow;
}
// Iterate `(block_size_i, ncol)` submatrix blocks.
for (i = 0; i < nrow; i += block_size_i) {
// Reset `A` to new block beginning.
A = A_block;
// Take block size of eveything left and reduce to max size.
block_size_i = nrow - i;
if (block_size_i > block_size) {
block_size_i = block_size;
}
// Compute first blocks column,
for (j = 0; j < block_size_i; ++j) {
sum[j] = A[j];
}
// and sum the following columns to the first one.
for (A += nrow; A < A_end; A += nrow) {
for (j = 0; j < block_size_i; ++j) {
sum[j] += A[j];
}
}
// Step one block forth.
A_block += block_size_i;
sum += block_size_i;
}
}
void rowSumsV3(const double *A,
const int nrow, const int ncol,
double *sum) {
int i, onei = 1;
double* ones = (double*)malloc(ncol * sizeof(double));
const double one = 1.0;
const double zero = 0.0;
for (i = 0; i < ncol; ++i) {
ones[i] = 1.0;
}
matrixprod(A, nrow, ncol, ones, ncol, 1, sum);
free(ones);
}
void colSums(const double *A, const int nrow, const int ncol,
double *sums) {
int i, j, nrowb = 4 * (nrow / 4); // 4 dividable nrow block, biggest 4*k <= nrow.
double sum;
for (j = 0; j < ncol; ++j) {
sum = 0.0;
for (i = 0; i < nrowb; i += 4) {
sum += A[i]
+ A[i + 1]
+ A[i + 2]
+ A[i + 3];
}
for (; i < nrow; ++i) {
sum += A[i];
}
*(sums++) = sum;
A += nrow;
}
}
void rowSquareSums(const double *A,
const int nrow, const int ncol,
double *sum) {
int i, j, block_size, block_size_i;
const double *A_block = A;
const double *A_end = A + nrow * ncol;
if (nrow > 508) {
block_size = 508;
} else {
block_size = nrow;
}
// Iterate `(block_size_i, ncol)` submatrix blocks.
for (i = 0; i < nrow; i += block_size_i) {
// Reset `A` to new block beginning.
A = A_block;
// Take block size of eveything left and reduce to max size.
block_size_i = nrow - i;
if (block_size_i > block_size) { // TODO: contains BUG!!! floor last one !!!
block_size_i = block_size;
} /// ...
// TODO:
// Copy blocks first column.
for (j = 0; j < block_size_i; j += 4) {
sum[j] = A[j] * A[j];
sum[j + 1] = A[j + 1] * A[j + 1];
sum[j + 2] = A[j + 2] * A[j + 2];
sum[j + 3] = A[j + 3] * A[j + 3];
}
for (; j < block_size_i; ++j) {
sum[j] = A[j] * A[j];
}
// Sum following columns to the first one.
for (A += nrow; A < A_end; A += nrow) {
for (j = 0; j < block_size_i; j += 4) {
sum[j] += A[j] * A[j];
sum[j + 1] += A[j + 1] * A[j + 1];
sum[j + 2] += A[j + 2] * A[j + 2];
sum[j + 3] += A[j + 3] * A[j + 3];
}
for (; j < block_size_i; ++j) {
sum[j] += A[j] * A[j];
}
}
// Step one block forth.
A_block += block_size_i;
sum += block_size_i;
}
}
void rowSumsSymVec(const double *Avec, const int nrow,
const double diag,
double *sum) {
int i, j;
if (diag == 0.0) {
memset(sum, 0, nrow * sizeof(double));
} else {
for (i = 0; i < nrow; ++i) {
sum[i] = diag;
}
}
for (j = 0; j < nrow; ++j) {
for (i = j + 1; i < nrow; ++i, ++Avec) {
sum[j] += *Avec;
sum[i] += *Avec;
}
}
}
#define ROW_SWEEP_ALG(op) \
/* Iterate `(block_size_i, ncol)` submatrix blocks. */ \
for (i = 0; i < nrow; i += block_size_i) { \
/* Set `A` and `C` to block beginning. */ \
A = A_block; \
C = C_block; \
/* Get current block's row size. */ \
block_size_i = nrow - i; \
if (block_size_i > block_size) { \
block_size_i = block_size; \
} \
/* Perform element wise operation for block. */ \
for (; A < A_end; A += nrow, C += nrow) { \
for (j = 0; j < block_size_i; ++j) { \
C[j] = (A[j]) op (v[j]); \
} \
} \
/* Step one block forth. */ \
A_block += block_size_i; \
C_block += block_size_i; \
v += block_size_i; \
}
/* C[, j] = A[, j] * v for each j = 1 to ncol */
void rowSweep(const double *A, const int nrow, const int ncol,
const char* op,
const double *v, // vector of length nrow
double *C) {
int i, j, block_size, block_size_i;
const double *A_block = A;
double *C_block = C;
const double *A_end = A + nrow * ncol;
if (nrow > CVE_MEM_CHUNK_SMALL) { // small because 3 vectors in cache
block_size = CVE_MEM_CHUNK_SMALL;
} else {
block_size = nrow;
}
if (*op == '+') {
ROW_SWEEP_ALG(+)
} else if (*op == '-') {
ROW_SWEEP_ALG(-)
} else if (*op == '*') {
ROW_SWEEP_ALG(*)
} else if (*op == '/') {
ROW_SWEEP_ALG(/)
} else {
error("Got unknown 'op' (opperation) argument.");
}
}
void transpose(const double *A, const int nrow, const int ncol, double* T) {
int i, j, len = nrow * ncol;
// Filling column-wise and accessing row-wise.
for (i = 0, j = 0; i < len; ++i, j += nrow) {
if (j >= len) {
j -= len - 1;
}
T[i] = A[j];
}
}
// Symmetric Packed matrix vector product.
// Computes
// y <- Ax
// where A is supplied as packed lower triangular part of a symmetric
// matrix A. Meaning that `vecA` is `vec_ltri(A)`.
void sympMV(const double* vecA, const int nrow, const double* x, double* y) {
double one = 1.0;
double zero = 0.0;
int onei = 1;
F77_NAME(dspmv)("L", &nrow, &one, vecA, x, &onei, &zero, y, &onei);
}
void matrixprod(const double *A, const int nrowA, const int ncolA,
const double *B, const int nrowB, const int ncolB,
double *C) {
const double one = 1.0;
const double zero = 0.0;
// DGEMM with parameterization:
// C <- A %*% B
F77_NAME(dgemm)("N", "N", &nrowA, &ncolB, &ncolA,
&one, A, &nrowA, B, &nrowB,
&zero, C, &nrowA);
}
void crossprod(const double *A, const int nrowA, const int ncolA,
const double *B, const int nrowB, const int ncolB,
double *C) {
const double one = 1.0;
const double zero = 0.0;
// DGEMM with parameterization:
// C <- t(A) %*% B
F77_NAME(dgemm)("T", "N", &ncolA, &ncolB, &nrowA,
&one, A, &nrowA, B, &nrowB,
&zero, C, &ncolA);
}
#define KRONECKER_ALG(op) \
for (j = 0; j < ncolA; ++j) { \
for (l = 0; l < ncolB; ++l) { \
colB = B + (l * nrowB); \
for (i = 0; i < nrowA; ++i) { \
for (k = 0; k < nrowB; ++k) { \
*(C++) = (A[i]) op (colB[k]); \
} \
} \
} \
A += nrowA; \
}
void kronecker(const double * restrict A, const int nrowA, const int ncolA,
const double * restrict B, const int nrowB, const int ncolB,
const char* op,
double * restrict C) {
int i, j, k, l;
const double *colB;
if (*op == '+') {
KRONECKER_ALG(+)
} else if (*op == '-') {
KRONECKER_ALG(-)
} else if (*op == '*') {
KRONECKER_ALG(*)
} else if (*op == '/') {
KRONECKER_ALG(/)
} else {
error("Got unknown 'op' (opperation) argument.");
}
}
void nullProj(const double *B, const int nrowB, const int ncolB,
double *Q) {
const double minusOne = -1.0;
const double one = 1.0;
// Initialize Q as identity matrix.
memset(Q, 0, sizeof(double) * nrowB * nrowB);
double *Q_diag, *Q_end = Q + nrowB * nrowB;
for (Q_diag = Q; Q_diag < Q_end; Q_diag += nrowB + 1) {
*Q_diag = 1.0;
}
// DGEMM with parameterization:
// Q <- (-1.0 * B %*% t(B)) + Q
F77_NAME(dgemm)("N", "T", &nrowB, &nrowB, &ncolB,
&minusOne, B, &nrowB, B, &nrowB,
&one, Q, &nrowB);
}
void rangePairs(const int from, const int to, int *pairs) {
int i, j;
for (i = from; i < to; ++i) {
for (j = i + 1; j < to; ++j) {
pairs[0] = i;
pairs[1] = j;
pairs += 2;
}
}
}
// A dence skwe-symmetric rank 2 update.
// Perform the update
// C := alpha (A * B^T - B * A^T) + beta C
void skewSymRank2k(const int nrow, const int ncol,
double alpha, const double *A, const double *B,
double beta,
double *C) {
F77_NAME(dgemm)("N", "T",
&nrow, &nrow, &ncol,
&alpha, A, &nrow, B, &nrow,
&beta, C, &nrow);
alpha *= -1.0;
beta = 1.0;
F77_NAME(dgemm)("N", "T",
&nrow, &nrow, &ncol,
&alpha, B, &nrow, A, &nrow,
&beta, C, &nrow);
}
// TODO: clarify
static inline double gaussKernel(const double x, const double scale) {
return exp(scale * x * x);
}
// TODO: mutch potential for optimization!!!
static void weightedYandLoss(const int n,
const double *Y,
const double *vecD,
const double *vecW,
const double *colSums,
double *y1, double *L, double *vecS,
double *loss) {
int i, j, k, N = n * (n - 1) / 2;
double l;
for (i = 0; i < n; ++i) {
y1[i] = Y[i];
L[i] = Y[i] * Y[i];
}
for (k = j = 0; j < n; ++j) {
for (i = j + 1; i < n; ++i, ++k) {
y1[i] += Y[j] * vecW[k];
y1[j] += Y[i] * vecW[k];
L[i] += Y[j] * Y[j] * vecW[k];
L[j] += Y[i] * Y[i] * vecW[k];
}
}
for (i = 0; i < n; ++i) {
y1[i] /= colSums[i];
L[i] /= colSums[i];
}
l = 0.0;
for (i = 0; i < n; ++i) {
l += (L[i] -= y1[i] * y1[i]);
}
*loss = l / (double)n;
for (k = j = 0; j < n; ++j) {
for (i = j + 1; i < n; ++i, ++k) {
l = Y[j] - y1[i];
vecS[k] = (L[i] - (l * l)) / colSums[i];
l = Y[i] - y1[j];
vecS[k] += (L[j] - (l * l)) / colSums[j];
}
}
for (k = 0; k < N; ++k) {
vecS[k] *= vecW[k] * vecD[k];
}
}
void grad(const int n, const int p, const int q,
const double *X,
const double *X_diff,
const double *Y,
const double *V,
const double h,
double *G, double *loss) {
// Number of X_i to X_j not trivial pairs.
int i, N = (n * (n - 1)) / 2;
double scale = -0.5 / h;
if (X_diff == (void*)0) {
// TODO: ...
}
// Allocate and compute projection matrix `Q = I_p - V * V^T`
double *Q = (double*)malloc(p * p * sizeof(double));
nullProj(V, p, q, Q);
// allocate and compute vectorized distance matrix with a temporary
// projection of `X_diff`.
double *vecD = (double*)malloc(N * sizeof(double));
double *X_proj;
if (p < 5) { // TODO: refine that!
X_proj = (double*)malloc(N * 5 * sizeof(double));
} else {
X_proj = (double*)malloc(N * p * sizeof(double));
}
matrixprod(X_diff, N, p, Q, p, p, X_proj);
rowSquareSums(X_proj, N, p, vecD);
// Apply kernel to distence vector for weights computation.
double *vecK = X_proj; // reuse memory area, no longer needed.
for (i = 0; i < N; ++i) {
vecK[i] = gaussKernel(vecD[i], scale);
}
double *colSums = X_proj + N; // still allocated!
rowSumsSymVec(vecK, n, 1.0, colSums); // rowSums = colSums cause Sym
// compute weighted responces of first end second momontum, aka y1, y2.
double *y1 = X_proj + N + n;
double *L = X_proj + N + (2 * n);
// Allocate X_diff scaling vector `vecS`, not in `X_proj` mem area because
// used symultanious to X_proj in final gradient computation.
double *vecS = (double*)malloc(N * sizeof(double));
weightedYandLoss(n, Y, vecD, vecK, colSums, y1, L, vecS, loss);
// compute the gradient using X_proj for intermidiate scaled X_diff.
rowSweep(X_diff, N, p, "*", vecS, X_proj);
// reuse Q which has the required dim (p, p).
crossprod(X_diff, N, p, X_proj, N, p, Q);
// Product with V
matrixprod(Q, p, p, V, p, q, G);
// And final scaling (TODO: move into matrixprod!)
scale = -2.0 / (((double)n) * h * h);
N = p * q;
for (i = 0; i < N; ++i) {
G[i] *= scale;
}
free(vecS);
free(X_proj);
free(vecD);
free(Q);
}

View File

@ -1,219 +0,0 @@
#ifndef CVE_INCLUDE_GUARD_
#define CVE_INCLUDE_GUARD_
#include <Rinternals.h>
#define CVE_MEM_CHUNK_SMALL 1016
#define CVE_MEM_CHUNK_SIZE 2032
void rowSums(const double *A,
const int nrow, const int ncol,
double *sum);
SEXP R_rowSums(SEXP A) {
SEXP sums = PROTECT(allocVector(REALSXP, nrows(A)));
rowSums(REAL(A), nrows(A), ncols(A), REAL(sums));
UNPROTECT(1);
return sums;
}
void rowSumsV2(const double *A,
const int nrow, const int ncol,
double *sum);
SEXP R_rowSumsV2(SEXP A) {
SEXP sums = PROTECT(allocVector(REALSXP, nrows(A)));
rowSumsV2(REAL(A), nrows(A), ncols(A), REAL(sums));
UNPROTECT(1);
return sums;
}
void rowSumsV3(const double *A,
const int nrow, const int ncol,
double *sum);
SEXP R_rowSumsV3(SEXP A) {
SEXP sums = PROTECT(allocVector(REALSXP, nrows(A)));
rowSumsV3(REAL(A), nrows(A), ncols(A), REAL(sums));
UNPROTECT(1);
return sums;
}
void colSums(const double *A,
const int nrow, const int ncol,
double *sum);
SEXP R_colSums(SEXP A) {
SEXP sums = PROTECT(allocVector(REALSXP, ncols(A)));
colSums(REAL(A), nrows(A), ncols(A), REAL(sums));
UNPROTECT(1);
return sums;
}
void rowSquareSums(const double*, const int, const int, double*);
SEXP R_rowSquareSums(SEXP A) {
SEXP result = PROTECT(allocVector(REALSXP, nrows(A)));
rowSquareSums(REAL(A), nrows(A), ncols(A), REAL(result));
UNPROTECT(1);
return result;
}
void rowSumsSymVec(const double *Avec, const int nrow,
const double diag,
double *sum);
SEXP R_rowSumsSymVec(SEXP Avec, SEXP nrow, SEXP diag) {
SEXP sum = PROTECT(allocVector(REALSXP, *INTEGER(nrow)));
rowSumsSymVec(REAL(Avec), *INTEGER(nrow), *REAL(diag), REAL(sum));
UNPROTECT(1);
return sum;
}
void rowSweep(const double *A, const int nrow, const int ncol,
const char* op,
const double *v, // vector of length nrow
double *C);
SEXP R_rowSweep(SEXP A, SEXP v, SEXP op) {
SEXP C = PROTECT(allocMatrix(REALSXP, nrows(A), ncols(A)));
rowSweep(REAL(A), nrows(A), ncols(A),
CHAR(STRING_ELT(op, 0)),
REAL(v), REAL(C));
UNPROTECT(1);
return C;
}
void transpose(const double *A, const int nrow, const int ncol, double* T);
SEXP R_transpose(SEXP A) {
SEXP T = PROTECT(allocMatrix(REALSXP, ncols(A), nrows(A)));
transpose(REAL(A), nrows(A), ncols(A), REAL(T));
UNPROTECT(1); /* T */
return T;
}
void sympMV(const double* vecA, const int nrow, const double* x, double* y);
SEXP R_sympMV(SEXP vecA, SEXP x) {
SEXP y = PROTECT(allocVector(REALSXP, length(x)));
sympMV(REAL(vecA), length(x), REAL(x), REAL(y));
UNPROTECT(1); /* y */
return y;
}
void matrixprod(const double *A, const int nrowA, const int ncolA,
const double *B, const int nrowB, const int ncolB,
double *C);
SEXP R_matrixprod(SEXP A, SEXP B) {
SEXP C = PROTECT(allocMatrix(REALSXP, nrows(A), ncols(B)));
matrixprod(REAL(A), nrows(A), ncols(A),
REAL(B), nrows(B), ncols(B),
REAL(C));
UNPROTECT(1);
return C;
}
void crossprod(const double* A, const int nrowA, const int ncolA,
const double* B, const int ncolB, const int nrowB,
double* C);
SEXP R_crossprod(SEXP A, SEXP B) {
SEXP C = PROTECT(allocMatrix(REALSXP, ncols(A), ncols(B)));
crossprod(REAL(A), nrows(A), ncols(A),
REAL(B), nrows(B), ncols(B),
REAL(C));
UNPROTECT(1);
return C;
}
void kronecker(const double *A, const int nrowA, const int ncolA,
const double *B, const int nrowB, const int ncolB,
const char *op,
double *C);
SEXP R_kronecker(SEXP A, SEXP B, SEXP op) {
SEXP C = PROTECT(allocMatrix(REALSXP,
nrows(A) * nrows(B),
ncols(A) * ncols(B)));
kronecker(REAL(A), nrows(A), ncols(A),
REAL(B), nrows(B), ncols(B),
CHAR(STRING_ELT(op, 0)),
REAL(C));
UNPROTECT(1);
return C;
}
void skewSymRank2k(const int n, const int k,
double alpha, const double *A, const double *B,
double beta,
double *C);
SEXP R_skewSymRank2k(SEXP A, SEXP B, SEXP alpha, SEXP beta) {
SEXP C = PROTECT(allocMatrix(REALSXP, nrows(A), nrows(A)));
memset(REAL(C), 0, nrows(A) * nrows(A) * sizeof(double));
skewSymRank2k(nrows(A), ncols(A),
*REAL(alpha), REAL(A), REAL(B),
*REAL(beta), REAL(C));
UNPROTECT(1);
return C;
}
void nullProj(const double* B, const int nrowB, const int ncolB,
double* Q);
SEXP R_nullProj(SEXP B) {
SEXP Q = PROTECT(allocMatrix(REALSXP, nrows(B), nrows(B)));
nullProj(REAL(B), nrows(B), ncols(B), REAL(Q));
UNPROTECT(1);
return Q;
}
void rangePairs(const int from, const int to, int *pairs);
SEXP R_rangePairs(SEXP from, SEXP to) {
int start = asInteger(from);
int end = asInteger(to) + 1;
int n = end - start;
SEXP out = PROTECT(allocMatrix(INTSXP, 2, n * (n - 1) / 2));
rangePairs(start, end, INTEGER(out));
UNPROTECT(1);
return out;
}
void grad(const int n, const int p, const int q,
const double *X,
const double *X_diff,
const double *Y,
const double *V,
const double h,
double *G, double *const loss);
SEXP R_grad(SEXP X, SEXP X_diff, SEXP Y, SEXP V, SEXP h) {
int N = (nrows(X) * (nrows(X) - 1)) / 2;
SEXP G = PROTECT(allocMatrix(REALSXP, nrows(V), ncols(V)));
SEXP loss = PROTECT(allocVector(REALSXP, 1));
grad(nrows(X), ncols(X), ncols(V),
REAL(X), REAL(X_diff), REAL(Y), REAL(V), *REAL(h),
REAL(G), REAL(loss));
UNPROTECT(2);
return G;
}
#endif /* CVE_INCLUDE_GUARD_ */