R函数用于计算某个值的频率低于特定值的频率

Question

这对R脚本中的零经验证明是一个怪物。我有57列，数据的30行的数据帧

这里就是我想要做：

1）到各列：

2）计数的时间2号/连续3/4/5/6/7/8/9值小于-1

3）打印结果作为文本文件

4）重复步骤2和3的第二列，因此on

我环顾四周，也于r计算器

check number of times consecutive value appear based on a certain criteria

这是我的一列数据：

data<-c(-0.996,-1.111,-0.638,0.047,0.694,1.901,2.863,2.611,2.56,2.016,0.929,-0.153,-0.617,-0.143 
0.199,0.556,0.353,-0.638,0.347,0.045,-0.829,-0.882,-1.143,-0.869,0.619,0.923,-0.474,0.227 
0.394,0.789,1.962,1.132,0.1,-0.278,-0.303,-0.606,-0.705,-0.858,-0.723,-0.081,1.206,2.329 
1.863,2.1,1.547,2.026,0.015,-0.441,-0.371,-0.304,-0.668,-0.953,-1.256,-1.185,-0.891,-0.569 
0.485,0.421,-0.004,0.024,-0.39,-0.58,-1.178,-1.101,-0.882,0.01,0.052,-0.166,-1.703,-1.048 
-0.718,-0.036,-0.561,-0.08,0.272,-0.041,-0.811,-0.929,-0.853,-1.047,0.431,0.576,0.642,1.62 
2.324,1.251,1.384,0.195,-0.081,-0.335,-0.176,1.089,-0.602,-1.134,-1.356,-1.203,-0.795,-0.752 
-0.692,-0.813,-1.172,-0.387,-0.079,-0.374,-0.157,0.263,0.313,0.975,2.298,1.71,0.229,-0.313 
-0.779,-1.12,-1.102,-1.01,-0.86,-1.118,-1.211,-1.081,-1.156,-0.972) 

当我运行下面的代码：

for (col in 1:ncol(data)) { 
    runs <- rle(data[,col]) 
    print(runs$lengths[which(runs$values < -1)]) 
} 

它给我这个：

[1] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

它已经计算了数值< -1但未运行。这是我在这里错了吗？

Answer 1

（大量编辑）

固定数据载体（缺少逗号）：

data <- c(-0.996,-1.111,-0.638,0.047,0.694,1.901,2.863,2.611,2.56,2.016,0.929,-0.153,-0.617,-0.143, 
      0.199,0.556,0.353,-0.638,0.347,0.045,-0.829,-0.882,-1.143,-0.869,0.619,0.923,-0.474,0.227, 
      0.394,0.789,1.962,1.132,0.1,-0.278,-0.303,-0.606,-0.705,-0.858,-0.723,-0.081,1.206,2.329, 
      1.863,2.1,1.547,2.026,0.015,-0.441,-0.371,-0.304,-0.668,-0.953,-1.256,-1.185,-0.891,-0.569, 
      0.485,0.421,-0.004,0.024,-0.39,-0.58,-1.178,-1.101,-0.882,0.01,0.052,-0.166,-1.703,-1.048, 
      -0.718,-0.036,-0.561,-0.08,0.272,-0.041,-0.811,-0.929,-0.853,-1.047,0.431,0.576,0.642,1.62, 
      2.324,1.251,1.384,0.195,-0.081,-0.335,-0.176,1.089,-0.602,-1.134,-1.356,-1.203,-0.795,-0.752, 
      -0.692,-0.813,-1.172,-0.387,-0.079,-0.374,-0.157,0.263,0.313,0.975,2.298,1.71,0.229,-0.313, 
      -0.779,-1.12,-1.102,-1.01,-0.86,-1.118,-1.211,-1.081,-1.156,-0.972) 

做data < -1为您提供了一个合乎逻辑的载体，我们可以指望的TRUE & FALSE运行：

runs <- rle(data < -1) 

print(runs) 

## Run Length Encoding 
## lengths: int [1:21] 1 1 20 1 29 2 8 2 4 2 ... 
## values : logi [1:21] FALSE TRUE FALSE TRUE FALSE TRUE ... 

然后提取长度只有TRUE运行：

print(runs$lengths[which(runs$values)]) 
## [1] 1 1 2 2 2 1 3 1 3 4 

和如前所示，迭代的数据帧的列：

# make a data frame from sampled versions of data 

set.seed(1492) # repeatable 

df <- data.frame(V1=data, 
       V2=sample(data, length(data), replace=TRUE), 
       V3=sample(data, length(data), replace=TRUE), 
       V4=sample(data, length(data), replace=TRUE)) 

# do the extraction 

for (col in 1:ncol(df)) { 
    runs <- rle(df[, col] < -1) 
    print(runs$lengths[which(runs$values)]) 
} 

## [1] 1 1 2 2 2 1 3 1 3 4 
## [1] 1 1 1 1 2 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 2 1 1 1 1 1 1 
## [1] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1