Plot barplot for all non zero columns only

Question

I have a data.frame with a single record and a N numeric columns. I would like to graph in R as a barplot of all columns having non-zero value on the same graph in ggplot.

For example:

df <- data.frame(a=1, b=0, c=10, d=20)

I transposed the data.frame but didn't manage to name both columns.

Note: Python implements the same graph though sns:

sns.barplot(x = 'Name', y = 'count', data*) 

Answer 1

For doing bar plot, you can use tidyr to pivot your data into a longer format compatible with ggplot and filter to remove zero values:

library(tidyverse)
library(ggplot2)
df=data.frame(a=1, b=0, c=10, d=20)
df %>% pivot_longer(everything(), names_to = "Variable", values_to = "Values") %>% filter(., Values != 0)

# A tibble: 3 x 2
  Variable Values
  <chr>     <dbl>
1 a             1
2 c            10
3 d            20

And if you are combining it with ggplot, it gives you:

df %>% pivot_longer(everything(), names_to = "Variable", values_to = "Values") %>% filter(., Values != 0) %>%
  ggplot(., aes(x = Variable, y = Values, fill= Variable))+
  geom_bar(stat = "identity")

Answer 2

In base R this is as easy as

barplot(unlist(df[1, df[1, ] != 0]))

If you prefer ggplot2 graphics,

library(tidyverse)

df %>%
  gather(key, value) %>%
  filter(value != 0) %>%
  ggplot(aes(x = key, y = value)) +
  geom_col()