R/pairs.R

# This file is part of RStan
# Copyright (C) 2012, 2013, 2014, 2015, 2016, 2017 Trustees of Columbia University
# Copyright (C) 1995-2012 The R Core Team
# Some parts  Copyright (C) 1999 Dr. Jens Oehlschlaegel-Akiyoshi
#
# RStan is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 3
# of the License, or (at your option) any later version.
#
# RStan is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.

pairs.stanfit <-
  function (x, labels = NULL, panel = NULL, ...,
            lower.panel = NULL, upper.panel = NULL, diag.panel = NULL, 
            text.panel = NULL, label.pos = 0.5 + 1/3, 
            cex.labels = NULL, font.labels = 1, 
            row1attop = TRUE, gap = 1, log = "",
            pars = NULL, include = TRUE,
            condition = "accept_stat__") {
    
    gsp <- get_sampler_params(x, inc_warmup = FALSE)
    if(is.null(pars)) {
      pars <- dimnames(x)[[3]]
      arr <- round(extract(x, pars = pars, permuted = FALSE), digits = 12)
      if ("energy__" %in% colnames(gsp[[1]])) {
        dims <- dim(arr)
        dims[3] <- dims[3] + 1L
        nms <- dimnames(arr)
        nms$parameters <- c(nms$parameters, "energy__")
        E <- sapply(gsp, FUN = function(y) y[,"energy__"])
        arr <- array(c(c(arr), c(E)), dim = dims, dimnames = nms)
      }
    }
    else if (!include) {
      incl_energy <- !any(pars == "energy__")
      pars <- setdiff(x@sim$pars_oi, pars)
      arr <- round(extract(x, pars = pars, permuted = FALSE), digits = 12)
      if (incl_energy && "energy__" %in% colnames(gsp[[1]])) {
        dims <- dim(arr)
        dims[3] <- dims[3] + 1L
        nms <- dimnames(arr)
        nms$parameters <- c(nms$parameters, "energy__")
        E <- sapply(gsp, FUN = function(y) y[,"energy__"])
        arr <- array(c(c(arr), c(E)), dim = dims, dimnames = nms)
      }
    }
    else arr <- round(extract(x, pars = pars, permuted = FALSE), digits = 12)
    sims <- nrow(arr)
    chains <- ncol(arr)
    varying <- apply(arr, 3, FUN = function(y) length(unique(c(y))) > 1)
    if (any(!varying)) {
      message("the following parameters were dropped because they are constant\n",
              paste(names(varying)[!varying], collapse = " "))
      arr <- arr[,,varying,drop = FALSE]
    }
    dupes <- duplicated(arr, MARGIN = 3)
    if (any(dupes)) {
      message("the following parameters were dropped because they are duplicative\n",
              paste(dimnames(arr)[[3]][dupes], collapse = " "))
      arr <- arr[,,!dupes,drop = FALSE]
    }
    divergent__ <- matrix(c(sapply(gsp, FUN = function(y) y[,"divergent__"])), 
                            nrow = sims * chains, ncol = dim(arr)[3])
    max_td <- x@stan_args[[1]]$control
    if (is.null(max_td)) max_td <- 10
    else {
      max_td <- max_td$max_treedepth
      if (is.null(max_td)) max_td <- 10
    }
    hit <- matrix(c(sapply(gsp, FUN = function(y) y[,"treedepth__"] >= max_td)), 
                    nrow = sims * chains, ncol = dim(arr)[3])
    
    if(is.list(condition)) {
      if(length(condition) != 2) stop("if a list, 'condition' must be of length 2")
      arr <- arr[,c(condition[[1]], condition[[2]]),,drop = FALSE]
      k <- length(condition[[1]])
      mark <- c(rep(TRUE, sims * k), rep(FALSE, sims * length(condition[[2]])))
    }
    else if(is.logical(condition)) {
      stopifnot(length(condition) == (sims * chains))
      mark <- !condition
    }
    else if(is.character(condition)) {
      condition <- match.arg(condition, several.ok = FALSE,
                             choices = c("accept_stat__", "stepsize__", "treedepth__", 
                                         "n_leapfrog__", "divergent__", "energy__",
                                         "lp__"))
      if (condition == "lp__") 
        mark <- simplify2array(get_logposterior(x, inc_warmup = FALSE))
      else mark <- sapply(gsp, FUN = function(y) y[,condition])
      
      if(condition == "divergent__") mark <- as.logical(mark)
      else mark <- c(mark) >= median(mark)
      if (length(unique(mark)) == 1) 
        stop(paste(condition, "is constant so it cannot be used as a condition"))
    }
    else if(!is.null(condition)) {
      if(all(condition == as.integer(condition))) {
        arr <- arr[,condition,,drop = FALSE]
        k <- ncol(arr) %/% 2
        mark <- c(rep(FALSE, sims * k), rep(TRUE, sims * (chains - k)))
      }
      else if(condition > 0 && condition < 1) {
        mark <- rep(1:sims > (condition * sims), times = chains)
      }
      else stop("'condition' must be an integer (vector) or a number between 0 and 1 exclusive")
    }
    else {
      k <- ncol(arr) %/% 2
      mark <- c(rep(FALSE, sims * k), rep(TRUE, sims * (chains - k)))
    }
    
    x <- apply(arr, MARGIN = "parameters", FUN = function(y) y)
    nc <- ncol(x)
    
    if (isTRUE(log)) {
      xl <- apply(x >= 0, 2, FUN = all)
      if("lp__" %in% names(xl)) xl["lp__"] <- FALSE
    }
    else if (is.numeric(log)) xl <- log
    else xl <- grepl("x", log)

    if (is.numeric(xl) || any(xl)) {
      x[,xl] <- log(x[,xl])
      colnames(x)[xl] <- paste("log", colnames(x)[xl], sep = "-")
    }
    if(is.null(lower.panel)) {
      if(!is.null(panel)) lower.panel <- panel
      else lower.panel <- function(x,y, ...) {
        dots <- list(...)
        x_tmp <- x[!mark]
        y_tmp <- y[!mark]
        div_tmp <- divergent__[!mark]
        hit_tmp <- hit[!mark]
        dots$x <- x_tmp
        dots$y <- y_tmp
        dots$nrpoints <- 0
        dots$postPlotHook <- function() {
          if (!identical(condition, "divergent__")) {
            points(x_tmp[div_tmp==1], y_tmp[div_tmp == 1], col = "red", pch = ".")
            points(x_tmp[hit_tmp == 1], y_tmp[hit_tmp == 1], col = "yellow", pch = ".")
          }
        }
        dots$add <- TRUE
        do.call(smoothScatter, args = dots)
      }
    }
    if(is.null(upper.panel)) {
      if(!is.null(panel)) upper.panel <- panel
      else upper.panel <- function(x,y, ...) {
        dots <- list(...)
        x_tmp <- x[mark]
        y_tmp <- y[mark]
        div_tmp <- divergent__[mark]
        hit_tmp <- hit[mark]
        dots$x <- x_tmp
        dots$y <- y_tmp
        dots$nrpoints <- 0
        dots$postPlotHook <- function() {
          if (!identical(condition, "divergent__")) {
            points(x_tmp[div_tmp==1], y_tmp[div_tmp == 1], col = "red", pch = ".")
            points(x_tmp[hit_tmp == 1], y_tmp[hit_tmp == 1], col = "yellow", pch = ".")
          }
        }
        dots$add <- TRUE
        do.call(smoothScatter, args = dots)
      }
    }
    if(is.null(diag.panel)) diag.panel <- function(x, ...) {
        usr <- par("usr"); on.exit(par(usr))
        par(usr = c(usr[1:2], 0, 1.5) )
        h <- hist(x, plot = FALSE)
        breaks <- h$breaks
        y <- h$counts; y <- y/max(y)
        nB <- length(breaks)
        rect(breaks[-nB], 0, breaks[-1], y, col="cyan", ...)
    }
    if(is.null(panel)) panel <- points
    
    if(is.null(text.panel)) textPanel <- function(x = 0.5, y = 0.5, txt, cex, font) {
      text(x,y, txt, cex = cex, font = font)
    }
    else textPanel <- text.panel
    if(is.null(labels)) labels <- colnames(x)

    mc <- match.call(expand.dots = TRUE)
    mc[1] <- call("pairs")
    mc$x <- x
    mc$labels <- labels
    mc$panel <- panel
    mc$lower.panel <- lower.panel
    mc$upper.panel <- upper.panel
    mc$diag.panel <- diag.panel
    mc$text.panel <- textPanel
    mc$log <- ""
    mc$condition <- NULL
    mc$pars <- NULL
    mc$include <- NULL
    eval(mc)
  }

Try the rstan package in your browser

Any scripts or data that you put into this service are public.

rstan documentation built on May 29, 2024, 11:04 a.m.