Display multiple plots with ggplot (from loop function)

[Rugada]

Hello,

I have 6 dataframes called "January","February","March","April","May","June" (they show the reflectance of trees). I made a list of these dataframes and then a loop function to generate multiple plots.

long_list <- c("January","February","March","April","May","June")

myplot <- function(data, title){
  ggplot(data, aes(x = bands, y = reflectance, color = class, group = class)) +
    geom_point(color="grey") +
    geom_line () +
    labs(title = title)
    }
for(i in long_list){
  print(myplot(get(i), i))
}

It worked; I am happy with the result.

But(!) I am not able to display all this plots in one window. I tried it with
par(mfrow=c(3,2))
and
myplot(long_list, col=3)
and
cowplot::plot_grid()
and
grid.arrange(myplot(long_list), nrow = 2, ncol = 3)
but was`t successful

How can I display these 6 generated plots in one window (col 3 and row 2)

Thanks!

[mattwarkentin]

Hi @Rugada,

ggplot2 has built in support for spreading data across multiple "plots" using facets. This makes it easier to achieve what I think you want. You can avoid the for loop entirely. You could be best combining all the data into a single data frame and letting ggplot handle creating the multiple plots.

Try this code:

library(tidyverse)

months <- c("January", "February", "March", "April", "May", "June")
  
dd <- tibble(bands = rnorm(600), reflectance = rnorm(600),
             class = rep(months, each = 100)) %>% 
  mutate(class = factor(class, levels = months))
  
ggplot(dd, aes(bands, reflectance, color = class)) +
  geom_point() +
  facet_wrap(~class)

^{Created on 2020-06-07 by the reprex package (v0.3.0)}

[Rugada]

First: Thank you for your answer.

I don`t know why, but I have problems with the installation of tidyverse; maybe because of the version of RStudio I use.

But I think I was not precise enough, cause my classes are tree species and not the dates. The dataframes I used to make the list are in long format. The classes are tree species (in the reduced example its class oak and class spruce) and the sensor bands range from B1 to B5. Each dataframe represents a specific month. Data looks like:

head(January)

  class     band reflectance
1  oak      B1     182.04762
2  spruce   B1     80.63158
...
3  oak      B2     408.61905
4  spruce   B2     183.84211
...
5  oak      B3     279.85714
6  spruce   B3     144.52632
... until...B5

So the ggplot function would look like:

 ggplot(Janaury, aes(x = bands, y = reflectance, color = class, group = class)) +
  geom_point() +

Before I convert my data into long format they looked like this in wide format:

head(January_wide)

     class        B1       B2       B3       B4       B5
1   oak       182.04762  408.6190  279.8571 993.7619 2490.286
...
2   spruce    80.63158   183.8421  144.5263 584.8421 1489.316
...

[mattwarkentin]

Hi @Rugada,

I don't know your specific data or what exactly you want the final output to look like. You were using ggplot2 already, so presumably you have it installed. The entire tidyverse is not necessary. In almost all cases, long data format are preferred, especially for plotting with ggplot.

If you can share the data in a usable format, I am happy to try and reproduce your error and help find a solution.