R ifelse 在 mutate 中引用第二个 data.frame

Question

I am experimenting with building a simple model predicting sex from height. I know that the following is quite naive, but I am stuck with a R syntax problem.. First I create a dataframe with a probability for female for each height group. Then I want to look up this model to assign sex for the test data according to a certain limit

library(tidyverse)
library(dslabs)
library(caret)
# rm(list=ls())
ind <- createDataPartition(heights$height,1, p=0.5, list = FALSE) 
test_height <- heights %>% slice(ind)
train_height <- heights %>% slice(-ind)
head(train_height)

model <- train_height %>% mutate(height=round(height)) %>% group_by(height) %>% summarise(pf = mean(sex=="Female"))

y_hat <- test_height %>% mutate(height=round(height))  %>% mutate(pred = ifelse(model[model$height == height, ]$pf > 0.5, "Female", "Male"))

I get

Error in `mutate()`:
! Problem while computing `pred = ifelse(model[model$height == height, ]$pf > 0.5,
  "Female", "Male")`.
Caused by error:
! Must subset rows with a valid subscript vector.
i Logical subscripts must match the size of the indexed input.
x Input has size 28 but subscript `model$height == height` has size 526.

How can I lookup the right model row in my ifelse expression?

Thanks,

Hans

Answer 1

I would left_join(test_height, model, by=height) to match the model$pf values to test_height according to the right height, and then mutate on ifelse(pf > 0.5,...) . It seems simpler to me.