一种简洁的方法是
df[1:max(which(df$a>0)),]
# A tibble: 7 x 1
# a
# <dbl>
# 1 1
# 2 2
# 3 3
# 4 0
# 5 5
# 6 0
# 7 7
阿尔索
df[1:which.max(cumsum(df$a)),]
head(df,1-which.max(rev(df$a)>0))
df[rev(cumsum(rev(df$a>0)))>0,]
df$a
比较所有方法:
df <- data.frame(a = rbinom(5000, 2, 0.2) - 1)
microbenchmark(
df[1:max(which(df$a>0)),],
df[1:which.max(cumsum(df$a)),],
head(df,1-which.max(rev(df$a)>0)),
df[rev(cumsum(rev(df$a>0)))>0,],
df[1:tail(which(sign(df$a) == 1), 1),],
times = 10000
)
# Unit: microseconds
# expr min lq mean median uq max neval cld
# df[1:max(which(df$a > 0)), ] 52.817 58.5800 102.80519 62.2160 71.5910 17108.65 10000 a
# df[1:which.max(cumsum(df$a)), ] 36.190 40.7620 65.68274 43.0785 49.7835 18827.08 10000 a
# head(df, 1 - which.max(rev(df$a) > 0)) 214.812 230.7590 355.37321 249.1085 297.4340 18158.22 10000 c
# df[rev(cumsum(rev(df$a > 0))) > 0, ] 106.391 114.6345 192.44990 124.4690 141.5650 14473.12 10000 b
# df[1:tail(which(sign(df$a) == 1), 1), ] 106.152 116.8985 207.69863 125.6520 150.3425 195384.36 10000 b
