Why it keep saying non-numeric argument to binary operator

[Hades]

I need to find the difference between '2016' and '2018' but it doesn't work. Because I cannot reach the website so I downloaded it. You may work with it as I put the original data at the top.

knitr::opts_chunk$set(echo = TRUE)
library(tidyverse)
energy_types <- read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-08-04/energy_types.csv")
#> Error in open.connection(structure(4L, class = c("curl", "connection"), conn_id = <pointer: 0x000000000000024b>), : Could not resolve host: raw.githubusercontent.com
country_totals <- read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-08-04/country_totals.csv")
#> Error in open.connection(structure(5L, class = c("curl", "connection"), conn_id = <pointer: 0x000000000000024d>), : Could not resolve host: raw.githubusercontent.com

energy_types_filter <- energy_types %>%
  filter(type == "Conventional thermal") 
#> Error in filter(., type == "Conventional thermal"): 找不到对象'energy_types'
energy_types_filter %>%
  mutate(change = diff('2016', '2018'))
#> Error in mutate(., change = diff("2016", "2018")): 找不到对象'energy_types_filter'
arrange(energy_types_filter, desc(change))
#> Error in arrange(energy_types_filter, desc(change)): 找不到对象'energy_types_filter'
mutate(energy_types_filter, country_name = fct_inorder(country_name))
#> Error in mutate(energy_types_filter, country_name = fct_inorder(country_name)): 找不到对象'energy_types_filter'

gg_energy_types_filter <- ggplot(
  energy_types_filter,
  aes(x = 0, xend = change, y = country_name, yend = country_name), color = "blue"
) +
  geom_segment(arrow = grid::arrow(length = unit(1, "mm"), type = "closed"), color = 0.9) +
  theme(axis.title.y = element_blank()) +
  geom_text(color = c("firebrick", "steelblue"))
#> Error in ggplot(energy_types_filter, aes(x = 0, xend = change, y = country_name, : 找不到对象'energy_types_filter'
  theme_minimal_vgrid() +
  theme(legend = element_blank())
#> Error in theme_minimal_vgrid(): 没有"theme_minimal_vgrid"这个函数

[pe2ju]

How about this;

library(tidyverse)
energy_types <- read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-08-04/energy_types.csv")
country_totals <- read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-08-04/country_totals.csv")

typeof(energy_types) ; typeof(country_totals)
#> [1] "list"
#> [1] "list"

Data is in a list format, Convert them to tibble format

energy_types <- as_tibble(energy_types)
country_totals <- as_tibble(country_totals)

Manipulate data

energy_types_filter <- energy_types %>%
filter(type == "Conventional thermal",) %>%
mutate( 'change' = 2018 - 2016,
country_name = fct_inorder(country_name) ) %>%
arrange(desc(change) )

Go to plot

gg_energy_types_filter <- ggplot(
energy_types_filter,
aes( x = 0, xend = change, y = country_name, yend = country_name), color = "blue"
) +
geom_segment(arrow = grid::arrow(length = unit(1, "mm"), type = "closed"), color = 0.9) +
#geom_text( color = c("firebrick", "steelblue") ) + Add the label argument
cowplot::theme_minimal_vgrid() +
theme( axis.title.y = element_blank() )
gg_energy_types_filter

[Hades]

Well, in this case, I will only get "2" for the "change" variable. I need to subtract the 2016 column and the 2018 column from the dataset. I tried to add " in 2016 and 2018 but will only result

Error: Problem with `mutate()` column `change`.
i `change = "2018" - "2016"`.
x non-numeric argument to binary operator

[Sanjmeh]

The answer is in the error message. It is clearly shown that 2018 and 2016 are stored as character variables. Convert them to numeric and the arithmetic operator will work.

[pe2ju]

Try these codes below;

library(tidyverse)
energy_types <- read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-08-04/energy_types.csv")
country_totals <- read_csv("https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-08-04/country_totals.csv")

# Check data format
typeof(energy_types) ; typeof(country_totals)
#> [1] "list"
#> [1] "list"

# Data is in a list format, Convert them to tibble format
energy_types <- as_tibble(energy_types)
country_totals <- as_tibble(country_totals)

# Manipulate data
energy_types_filter <- energy_types %>%
  filter(type == "Conventional thermal",) %>%
  mutate( change = `2018` - `2016`,
          country_name = fct_inorder(country_name) ) %>%
  arrange(desc(change) )

# Go to plot
gg_energy_types_filter <- ggplot(
  energy_types_filter,
  aes( x = 0, xend = change, y = country_name, yend = country_name), color = "blue"
) +
  geom_segment(arrow = grid::arrow(length = unit(1, "mm"), type = "closed"), color = 0.9) +
  #geom_text( color = c("firebrick", "steelblue") ) + Add the label argument
  cowplot::theme_minimal_vgrid() +
  theme( axis.title.y = element_blank() )
gg_energy_types_filter