如何从R中的数据框中提取高数值

Question

I am presenting a small data frame here that is from model output file and I extracted the required parameters time and WatBlar and converted it into data frame. Complete code starts from here.

library(stringr)

x <- readLines("G:/Rlearning/Mohsin-FM/Balance.out")

a <- grep("[T]", x, value = T)
b <- grep("Time", a , value = T)

c <-  b[-c(1,2)]
d <- grep("WatBalR", x, value = T)

The data is like that

data <- data.frame(time =c, watbalr = d)

> data


                         time                          watbalr
1  Time       [T]        3.0000  WatBalR  [%]              0.040
2  Time       [T]        6.0000  WatBalR  [%]              0.024
3  Time       [T]        9.0000  WatBalR  [%]              0.044
4  Time       [T]       30.0000  WatBalR  [%]              0.034

I checked the data class it is data frame that is shown below.

> c
[1] " Time       [T]        3.0000" " Time       [T]        6.0000"
[3] " Time       [T]        9.0000" " Time       [T]       30.0000"

> class(c)
[1] "character" 



> d
[1] " WatBalR  [%]              0.040" " WatBalR  [%]              0.024"
[3] " WatBalR  [%]              0.044" " WatBalR  [%]              0.034"

> class(d)
[1] "character"

> class(data)
[1] "data.frame"

The code to extract the required values is written as shown below. But it is just assigning the value of the time 0 to 9, any value above than 9 it just start it again 0 to 9.

times   <- sub("^.+?(\\d)", "\\1", c)
WatBlaR <- sub("^.+?(\\d)", "\\1", d)

times   <- as.numeric(times)
WatBlaR <- as.numeric(WatBlaR)

# plot 
plot(x = times, y = WatBlaR)

The results for 4 values as mentioned above in data frame are shown below.

> times
[1] 3 6 9 0

But the required results for time are

3, 6, 9, 30

When I want to extract the model data from the daily basis data it present the values as

> times    
0,1,2,3,4,5,6,7,8,9, 0,1,2,3,4,5,6,7,8,9, 0,1,2,3,4,5,6,7,8,9

It just followed the sequence of all the time available 0 to 9, the required out should be like that

> times
1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30

Answer 1

So you are trying to extract values from a character vector in R. The base string functions are not as rich as might be desired for these kinds of situations. Consider adding stringr which is very handy for this kind of work.

library(stringr)

# I will create a toy df
df <- data.frame(A=c(1,2,16,5), B=c(0.1, 0.4, 0.6, 0.8), C=c('3.0000  WatBalR', '3.0000  WatBalR', '12.0000  WatBalR', '6.0000  WatBalR'),
            stringsAsFactors = FALSE)

# now I can extract with a simple regex pattern
times <- as.numeric(str_extract(df$C, '^[0-9]+'))

Here we use str_extract to extract with a signature of (data on which to operate, regex pattern) . We are also using $ to name the column in the data.frame and make it more legible, so we can pass what we need.

I can also get the doubles easily:

watblar <- as.double(str_extract(df$B, '^[.0-9]+'))

and types are correct

> str(times)
 num [1:4] 3 3 12 6
> str(watblar)
 num [1:4] 0.1 0.4 0.6 0.8

Answer 2

You can extract the numbers using sub from the base package where you look for a pattern of the form

• any number of digits followed by
• a dot (optional) followed by
• any number of digits (optional)

This is how you could do it:

library(magrittr)   ## For pipe %>%

# Some sample data
data <- data.frame(time = c(" Time       [T]        3.0000", 
                " Time       [T]        6.0000",
                " Time       [T]        9.0000", 
                " Time       [T]       30.0000"),
        watbalr = c(" WatBalR  [%]              0.040", 
                " WatBalR  [%]              0.024", 
                " WatBalR  [%]              0.044", 
                " WatBalR  [%]              0.034"),    stringsAsFactors = FALSE)

## Extract pattern and convert to numeric:
times <- sub("[^[:digit:]]*(\\d+\\.?\\d*).*", "\\1", data$time) %>%
        as.numeric
WatBalR  <- sub("[^[:digit:]]*(\\d+\\.?\\d*).*", "\\1", data$watbalr) %>%
        as.numeric

> times
# [1]  3  6  9 30
> WatBalR
# [1] 0.040 0.024 0.044 0.034