CODEWITHSUNDEEP
rggplot2heatmap
user1700890
user1700890

Reputation: 7730

Heat map in ggplot - no color variation

I would like to plot simple heatmap using ggplot

library(ggplot2)

my_df <- data.frame(var_1 = sample(c('a','b', 'c'), 1000, replace = TRUE),
                    var_2 = sample(c('foo', 'bar', 'foo_bar'), 1000, replace = TRUE),
                    var_3 = sample(c(1,0), 1000, replace = TRUE))

ggplot(data = my_df, aes(x = var_1, y = var_2, fill = var_3)) +
  geom_tile()

Here is what I get: Heat map

I cannot understand where there is no color variation. I changed number of rows in sample function, but always received only two colors in heat map. Any suggestions?

Upvotes: 1

Views: 964

Answers (3)

Brian
Brian

Reputation: 8275

Replace geom_tile with a layer that computes the statistical transformation you desire:

ggplot(data = my_df, aes(x = var_1, y = var_2)) + 
  stat_summary_2d(fun = "mean", geom = "tile", 
                  aes(z = var_3, fill = ..value..))

enter image description here

Upvotes: 2

dww
dww

Reputation: 31452

To plot the proportion of 1 and 0 by factor, you will need to calculate the mean of var_3 for each combination of var_1 and var_2. Here I show how to do that using data.table, although you could also use dplyr.summarise or aggregate from base R.

library(data.table)
setDT(my_df)
ggplot(data = my_df[, var_3 := mean(var_3), by = .(var_1,var_2)], 
       aes(x = var_1, y = var_2, fill = var_3)) +
  geom_tile()

Upvotes: 2

sweetmusicality
sweetmusicality

Reputation: 937

that's because the only values you are choosing to input are either 0 or 1 from your vector in your var_3 = sample(c(1,0), 1000, replace = TRUE). If instead you want to choose values between 0 and 1, use this code instead:

var_3 = sample(0:10, 1000, replace = TRUE)/10

Upvotes: 3

Related Questions