plotTrace: Plot trace

View source: R/plotTrace.R

plotTraceR Documentation

Plot trace

Description

Plots the posterior distributions of variables from trace file.

Usage

plotTrace(trace, color = "default", vars = NULL, match = NULL)

Arguments

trace

(list of data frames; no default) Name of a list of data frames, such as produced by readTrace(). If the readTrace() output contains multiple traces (such as from multiple runs), summarizeTrace() will provide summaries for each trace individually, as well as the combined trace.

color

("character"; "default") Colors for parameters. Defaults to default RevGadgets colors. For non-default colors, provide a named vector of length of the number of parameters.

vars

(character or character vector; NULL) The specific name(s) of the variable(s) to be summarized.

match

(character; NULL) A string to match to a group of parameters. For example, match = "er" will plot the variables "er[1]", "er[2]", "er[3]", etc.. match will only work if your search string is followed by brackets in one or more of the column names of the provided trace file. match = "er" will only return the exchangeability parameters, but will not plot "Posterior".

Details

Plots the posterior distributions of continuous variables from one or multiple traces (as in, from multiple runs). Shaded regions under the curve represent the 95% credible interval. If multiple traces are provided, plotTrace() will plot each run independently as well as plot the combined output. Note that for variables ith very different distributions, overlaying the plots may result in illegible figures. In these cases, we recommend plotting each parameter separately.

Value

plotTrace() returns a list of the length of provided trace object, plus one combined trace. Each element of the list contains a ggplot object with plots of the provided parameters. These plots may be modified in typical ggplot fashion.

Examples




# example with quantitative parameters

# download the example dataset to working directory
url_gtr <-
   "https://revbayes.github.io/tutorials/intro/data/primates_cytb_GTR.log"
dest_path_gtr <- "primates_cytb_GTR.log"
download.file(url_gtr, dest_path_gtr)

# to run on your own data, change this to the path to your data file
file <- dest_path_gtr

one_trace <- readTrace(paths = file)
plots <- plotTrace(trace = one_trace,
                    vars = c("pi[1]","pi[2]","pi[3]","pi[4]"))
plots[[1]]

# add custom colors
plots <- plotTrace(trace = one_trace,
                   vars = c("pi[3]","pi[4]","pi[1]","pi[2]"),
                   color = c("pi[1]" = "green",
                             "pi[2]"= "red",
                             "pi[3]"= "blue",
                             "pi[4]"= "orange"))
plots[[1]]

# make the same plot, using match
plots <- plotTrace(trace = one_trace, match = "pi")
plots[[1]]

#' # remove file
# WARNING: only run for example dataset!
# otherwise you might delete your data!
file.remove(dest_path_gtr)

# plot some qualitative variables

# download the example dataset to working directory
url_rj <- "https://revbayes.github.io/tutorials/intro/data/freeK_RJ.log"
dest_path_rj <- "freeK_RJ.log"
download.file(url_rj, dest_path_rj)

file <- dest_path_rj
trace <- readTrace(path = file)

plots <- plotTrace(trace = trace,
                   vars = c("prob_rate_12", "prob_rate_13",
                            "prob_rate_31", "prob_rate_32"))
plots[[1]]

# with custom colors
plots <- plotTrace(trace = trace,
                   vars = c("prob_rate_12", "prob_rate_13",
                            "prob_rate_31", "prob_rate_32"),
                   color = c("prob_rate_12" = "green",
                             "prob_rate_13" = "red",
                             "prob_rate_31"= "blue",
                             "prob_rate_32" = "orange"))
plots[[1]]

# remove file
# WARNING: only run for example dataset!
# otherwise you might delete your data!
file.remove(dest_path_rj)




RevGadgets documentation built on May 29, 2024, 10:03 a.m.