R: mapped_discrete` objects can only be created from numeric vectors-CodePudding

The

Thanks!

CodePudding user response：

It looks like you were wanting to produce this plot for missing data on each row rather than for each variable (though I've provided both here). The main issue is that levels is not provided, so we can create that here, then provide as a factor to scale_x_discrete.

library(tidyverse) output <- df %>% mutate(id = row_number()) %>% pivot_longer(-id, names_to = "key", values_to = "val") %>% select(-key) %>% group_by(id) %>% mutate(isna = is.na(val), total = n()) %>% group_by(id, total, isna) %>% summarise(num.isna = n()) %>% mutate(pct = num.isna / total * 100) levels <- output %>% filter(isna == T) %>% arrange(desc(pct)) %>% pull(id) row.plot <- output %>% ggplot() geom_bar(aes( x = reorder(id, desc(pct)), y = pct, fill = isna ), stat = 'identity', alpha = 0.8) scale_x_discrete(limits = factor(levels)) scale_fill_manual( name = "", values = c('steelblue', 'tomato3'), labels = c("Present", "Missing") ) coord_flip() labs(title = "Percentage of missing values", x = 'Row Number', y = "% of missing values")

Output

Or if you want to do it by variable, then:

output <- df %>% pivot_longer(everything(), names_to = "key", values_to = "val") %>% group_by(key) %>% mutate(isna = is.na(val), total = n()) %>% group_by(key, total, isna) %>% summarise(num.isna = n()) %>% mutate(pct = num.isna / total * 100) levels <- output %>% filter(isna == T) %>% arrange(desc(pct)) %>% pull(key) row.plot <- output %>% ggplot() geom_bar(aes( x = reorder(key, desc(pct)), y = pct, fill = isna ), stat = 'identity', alpha = 0.8) scale_x_discrete(limits = levels) scale_fill_manual( name = "", values = c('steelblue', 'tomato3'), labels = c("Present", "Missing") ) coord_flip() labs(title = "Percentage of missing values", x = 'Variable', y = "% of missing values")

Output

CodePudding user response：

The error is caused by scale_x_discrete.
You don't need it because in your example id is numeric and doesn't have levels as a factor would:

df %>% mutate(id = row_number()) %>% gather(-id, key = "key", value = "val") %>% mutate(isna = is.na(val)) %>% ggplot(aes(key, id, fill = isna)) geom_raster(alpha=0.8) scale_fill_manual(name = "", values = c('steelblue', 'tomato3'), labels = c("Present", "Missing")) #scale_x_discrete(limits = levels) labs(x = "Variable", y = "Row Number", title = "Missing values in rows") coord_flip()