unlink("man/figures/*") knitr::opts_chunk$set( collapse = TRUE, comment = "#>", fig.path = "man/figures/README-", out.width = "50%" )
Graphics functions which provide a simple, easy-to-learn interface, allowing beginners to explore and visualize data easily.
Easily install the latest release version from CRAN:
install.packages("iNZightRegression")
For the latest development version:
# install.packages("devtools") devtools::install_github("iNZightVIT/iNZightTools") devtools::install_github("iNZightVIT/iNZightMR") devtools::install_github("iNZightVIT/iNZightPlots")
iNZightPlots
provides a simple interface to graphics, which are most easily produced using the following function:
library(iNZightPlots) inzplot(~Sepal.Length, data = iris)
library(iNZightPlots) t <- file.path("man", "figures", "README-demo-1.png") png(t) inzplot(~Sepal.Length, data = iris) dev.off()
r knitr::include_graphics(t)
From there, formula notation can be used to explore relationships between variables:
inzplot(Sepal.Length ~ Species, data = iris) inzplot(Sepal.Length ~ Sepal.Width, data = iris) inzplot(Sepal.Length ~ Sepal.Width | Species, data = iris)
t <- file.path("man", "figures", "README-demo-usage-%d.png") png(t) inzplot(~Sepal.Length, data = iris) inzplot(Sepal.Length ~ Sepal.Width, data = iris) inzplot(Sepal.Length ~ Sepal.Width | Species, data = iris) dev.off()
These functions all produce different graphs (dot plot and scatter plot), which means the focus is on exploring data, not on which type of graph to use. Subsetting works as expected, and can include up to two variables (e.g., y ~ x | g1 + g2
).
There are additionally two companion functions for summary and inference information, which are
inzsummary(~Sepal.Length, data = iris) inzinference(Sepal.Length ~ Species, data = iris)
These use the same notation as the plot, and are intended to be used alongside.
There are many, many features available in iNZightPlots
, all of which are added simply through additional arguments to inzplot()
. For a full list, check ?iNZightPlot
(this is the base function for which inzplot()
is a convenience wrapper) and ?inzpar
.
Probably the most common features to add to a graph, we have the sizeby
and colby
arguments which map an additional variable to size and colour, respectively:
inzplot(Sepal.Length ~ Sepal.Width, data = iris, colby = Species)
Usually you want to know if what you see has any statistical significance, and iNZightPlots
makes this easy. We also provide comparison intervals which can be, very simply, interpreted as "overlapping comparison intervals indicate no significant difference between groups".
# Add confidence intervals inzplot(Sepal.Length ~ Species, data = iris, inference.type = c("conf", "comp"), # red and black, respectively inference.par = "mean" )
t <- file.path("man", "figures", "README-demo-inf.png") png(t) inzplot(Sepal.Length ~ Species, data = iris, inference.type = c("conf", "comp"), # red and black, respectively inference.par = "mean" ) dev.off()
r knitr::include_graphics(t)
For scatter plots, inference uses linear trends. You can add a bootstrap sample of trend estimates using the bs.inference
argument:
inzplot(Sepal.Length ~ Sepal.Width, data = iris, trend = "linear", bs.inference = TRUE )
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.