我想把一个 data.frame 到 Spark DataFrame :

library(sparklyr)

conf <- list()

conf$`sparklyr.cores.local` <- 4
conf$`sparklyr.shell.driver-memory` <- "16G"
conf$spark.memory.fraction <- 0.9

sc <- spark_connect(master = "local", 
                    config = conf)

load("./data/dfm_training.RData")
load("./data/training.RData")

df <- training[,c(1,2,3,18)]
df$TextLength <- nchar(df$txt)
df <- df[df$TextLength > 0,]

df.tokens.df <- cbind(nchar = df$TextLength, label = df$training2_CFT,
                  convert(dfm_training, to = "data.frame"))

## to spark

sdf <- copy_to(sc, df.tokens.df)

但是，我得到以下错误： Error: C stack usage 133776899 is too close to the limit 我试过切换内核的数量和使用的内存的数量，但没有改变。