problem with replacing summarise_all() with summarise(across()) in sparklyr

tidyverse
1

I've tried to change the summarise_all() with summarise(across()) equivalent in the following code, but I've got an error.

library(sparklyr)
library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union

sc <- spark_connect('local', version = '2.4')
data <- copy_to(sc, mtcars)

# 1st query
data %>%
  mutate(transmission = ifelse(am == 0, "automatic", "manual")) %>%
  group_by(transmission) %>%
  summarise_all(mean)
#> Warning: Missing values are always removed in SQL.
#> Use `mean(x, na.rm = TRUE)` to silence this warning
#> This warning is displayed only once per session.
#> # Source: spark<?> [?? x 12]
#>   transmission   mpg   cyl  disp    hp  drat    wt  qsec    vs    am  gear  carb
#>   <chr>        <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 manual        24.4  5.08  144.  127.  4.05  2.41  17.4 0.538     1  4.38  2.92
#> 2 automatic     17.1  6.95  290.  160.  3.29  3.77  18.2 0.368     0  3.21  2.74

# 2nd query
data %>%
  mutate(transmission = ifelse(am == 0, "automatic", "manual")) %>%
  group_by(transmission) %>%
  summarise(across(.fns = mean))
#> Error: Can't rename variables in this context.

Created on 2021-02-05 by the reprex package (v0.3.0)

2

I can't run the following line for some reason:
sc <- spark_connect('local', version = '2.4')

In any case, the problem might be that you need summarise(across(everything(), mean)) to specify that you want to summarize across all columns:

mtcars %>%
  mutate(transmission = ifelse(am == 0, "automatic", "manual")) %>%
  group_by(transmission) %>%
  summarise(across(everything(), mean))

## `summarise()` ungrouping output (override with `.groups` argument)
## # A tibble: 2 x 12
##   transmission   mpg   cyl  disp    hp  drat    wt  qsec    vs    am  gear  carb
##   <chr>        <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 automatic     17.1  6.95  290.  160.  3.29  3.77  18.2 0.368     0  3.21  2.74
## 2 manual        24.4  5.08  144.  127.  4.05  2.41  17.4 0.538     1  4.38  2.92
3

my code will work in R as you said, but when I use it on a data form spark ( data) it makes an error.

are you sure that you installed the sparklyr and your spark version is 2.4?

4

I used install.packages("sparklyr")....

5

now use sparklyr::spark_installed_versions()and replace 2.4 with your spark version and then run the code.

6

i found the answer to this problem.
the support for summarise(across(...) is added to the new version of sparklyr.
you sould install the new version of sparklyr from github.