{ 中的错误：任务 2 失败 -“`mutate()` 输入 `Model` 的问题。x 0（非 NA）案例ℹ 输入 `Model` 是 `map(...)`。”

Question

这是一个代码块，用于复制我遇到错误消息的一些数据：

# Set up
library(tidyverse)
library(lubridate)
library(foreach)

# Create data
mydf <- data.frame(
  cohort = seq(ymd('2019-01-01'), ymd('2019-12-31'), by = '1 days'),
  n = rnorm(365, 1000, 50) %>% round,
  cohort_cost = rnorm(365, 800, 50)
) %>% 
  crossing(tenure_days = 0:365) %>% 
  mutate(activity_date = cohort + days(tenure_days)) %>% 
  mutate(daily_revenue = rnorm(nrow(.), 20, 1)) %>% 
  group_by(cohort) %>% 
  arrange(activity_date) %>% 
  mutate(cumulative_revenue = cumsum(daily_revenue)) %>% 
  arrange(cohort, activity_date) %>% 
  mutate(payback_velocity = round(cumulative_revenue / cohort_cost, 2)) %>% 
  select(cohort, n, cohort_cost, activity_date, tenure_days, everything())

## wider data
mydf_wide <- mydf %>% 
  select(cohort, n, cohort_cost, tenure_days, payback_velocity) %>% 
  group_by(cohort, n, cohort_cost) %>% 
  pivot_wider(names_from = tenure_days, values_from = payback_velocity, names_prefix = 'velocity_day_') %>% 
  mutate(Category = rep(LETTERS[1:3], length.out = n()))

models <- data.frame(
  from = mydf$tenure_days %>% unique,
  to = mydf$tenure_days %>% unique
) %>% 
  expand.grid %>% 
  filter(to > from) %>% 
  filter(from > 0) %>% 
  arrange(from) %>% 
  mutate(mod_formula = paste0('velocity_day_', to, ' ~ velocity_day_', from)) %>% 
  mutate(Category = rep(LETTERS[1:3], length.out = n()))

model_splits <- models %>% split(.$Category)

我有一个数据框，每一行都包含一个模型规范，我想在其中将模型拟合为变异字段。

运行上面的块后，我正在使用的结果数据如下所示：

model_splits$A %>% glimpse
Rows: 22,144
Columns: 4
$ from        <int> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1…
$ to          <int> 2, 5, 8, 11, 14, 17, 20, 23, 26, 29, 32, 35, 38, 41, 44, 47, 50, 53, 56, 59, 62, 65, 68, 71, 74, 77, 80, 83, 86, 89, 92, 95, 98, 101, 104, 107, 110,…
$ mod_formula <chr> "velocity_day_2 ~ velocity_day_1", "velocity_day_5 ~ velocity_day_1", "velocity_day_8 ~ velocity_day_1", "velocity_day_11 ~ velocity_day_1", "veloci…
$ Category    <chr> "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A", "A"…

即包含模型规范的数据框。我也有如下所示的数据框 mydf_wide：

mydf_wide %>% head()
# A tibble: 6 x 370
# Groups:   cohort, n, cohort_cost [6]
  cohort         n cohort_cost velocity_day_0 velocity_day_1 velocity_day_2 velocity_day_3 velocity_day_4 velocity_day_5 velocity_day_6 velocity_day_7 velocity_day_8
  <date>     <dbl>       <dbl>          <dbl>          <dbl>          <dbl>          <dbl>          <dbl>          <dbl>          <dbl>          <dbl>          <dbl>
1 2019-01-01   884        723.           0.03           0.05           0.08           0.11           0.14           0.17           0.19           0.22           0.25
2 2019-01-02  1026        698.           0.03           0.06           0.09           0.12           0.15           0.17           0.2            0.23           0.26
3 2019-01-03   911        906.           0.02           0.04           0.07           0.09           0.11           0.13           0.15           0.18           0.2 
4 2019-01-04   893        828.           0.02           0.05           0.07           0.1            0.12           0.15           0.17           0.2            0.22
5 2019-01-05   924        821.           0.02           0.05           0.07           0.1            0.12           0.15           0.17           0.2            0.22
6 2019-01-06  1032        797.           0.02           0.05           0.08           0.1            0.13           0.15           0.18           0.2            0.23

在一个循环中，我想遍历model_splits，并且在每种情况下，使用 map 来拟合模型：

# fit some models in a loop
foreach::foreach(c = model_splits %>% names, .combine='c') %do% {
  df <- model_splits[[c]] %>%
     sample_n(3) %>% 
    mutate(Model = map(.x = mod_formula, ~lm(.x, data = mydf_wide %>% filter(Category == c))))
}
Error in { : task 2 failed - "Problem with `mutate()` input `Model`.
x 0 (non-NA) cases
ℹ Input `Model` is `map(...)`."

我做了一些谷歌搜索，建议我寻找 NA 值，但上面的 DF 都没有丢失数据。

期望的结果是在每一行数据框模型上变异的新模型。

Answer 1

问题是因为“mydf_wide”中只有一个“类别”导致lm 调用出现问题

unique(mydf_wide$Category)
#[1] "A"

如果我们改变（只是为了演示）

mydf_wide$Category[91:201] <- "B"
mydf_wide$Category[202:361] <- "C"

代码有效

foreach::foreach(c = model_splits %>% names, .combine='c') %do% {
   df <- model_splits[[c]] %>%
      sample_n(3) %>% 
     mutate(Model = map(.x = mod_formula, ~lm(.x, data = mydf_wide %>% filter(Category == c))))
 }
#$from
#[1]  15 144   4

#$to
#[1] 330 237 137

#$mod_formula
#[1] "velocity_day_330 ~ velocity_day_15"  "velocity_day_237 ~ velocity_day_144" "velocity_day_137 ~ velocity_day_4"  

#$Category
#[1] "A" "A" "A"

#$Model
#$Model[[1]]

#Call:
#lm(formula = .x, data = mydf_wide %>% filter(Category == c))

#Coefficients:
#    (Intercept)  velocity_day_15  
#         0.4575          19.5458  
# 

#$Model[[2]]

#Call:
#lm(formula = .x, data = mydf_wide %>% filter(Category == c))

#Coefficients:
#     (Intercept)  velocity_day_144  
#        0.007502          1.639855  


#$Model[[3]]

#Call:
#lm(formula = .x, data = mydf_wide %>% filter(Category == c))

#Coefficients:
#   (Intercept)  velocity_day_4  
#       0.7735         21.4259  
# ...

---

或者也可以选择使用tryCatch 或possibly 进行包装，以保持函数运行，以防遇到错误