使用R中的Pivot_wider函数时出现错误？

huangapple
go评论228阅读模式
英文:

Error when using Pivot_wider function in R?

问题

我正在使用pivot_wider函数处理以下示例数据，但似乎我漏掉了一些东西，因为输出产生了NA，并没有正确将数据放入其相应的列中。有什么建议吗？

  1. library(tidyverse)
  2. set.seed(123)
  3. DF <- data.frame(Date = seq(as.Date("1991-01-01"), 
  4.                             to = as.Date("2000-12-31"), 
  5.                             by = "year"), 
  6.                  Parameter = rep(c("A", "B"), times = 10),
  7.                  Value = runif(10, 1, 50)) %>%
  8.   pivot_wider(names_from = Parameter, values_from = Value)

输出

  1. > print(DF)
  2. # A tibble: 10 × 3
  3.    Date       A         B        
  4.    <date>     <list>    <list>   
  5.  1 1991-01-01 <dbl [2]> <NULL>   
  6.  2 1992-01-01 <NULL>    <dbl [2]>
  7.  3 1993-01-01 <dbl [2]> <NULL>   
  8.  4 1994-01-01 <NULL>    <dbl [2]>
  9.  5 1995-01-01 <dbl [2]> <NULL>   
  10.  6 1996-01-01 <NULL>    <dbl [2]>
  11.  7 1997-01-01 <dbl [2]> <NULL>   
  12.  8 1998-01-01 <NULL>    <dbl [2]>
  13.  9 1999-01-01 <dbl [2]> <NULL>   
  14. 10 2000-01-01 <NULL>    <dbl [2]>

请注意，代码部分没有进行翻译。

英文:

I am using pivot_wider function on the following sample data but look like I am missing something here as the output are producing NA&#39;s with not rightly placing data into its respective column. Any suggestion, please?

  1. library(tidyverse)
  2. set.seed(123)
  3. DF &lt;- data.frame(Date = seq(as.Date(&quot;1991-01-01&quot;), 
  4.                         to = as.Date(&quot;2000-12-31&quot;), 
  5.                         by = &quot;year&quot;), 
  6.                         Parameter = rep(c(&quot;A&quot;,&quot;B&quot;), times = 10),
  7.                         Value = runif(10,1,50)) %&gt;% 
  8.                         pivot_wider(names_from = Parameter, values_from = Value)

Output

  1. &gt; print(DF)
  2. # A tibble: 10 &#215; 3
  3.    Date       A         B        
  4.    &lt;date&gt;     &lt;list&gt;    &lt;list&gt;   
  5.  1 1991-01-01 &lt;dbl [2]&gt; &lt;NULL&gt;   
  6.  2 1992-01-01 &lt;NULL&gt;    &lt;dbl [2]&gt;
  7.  3 1993-01-01 &lt;dbl [2]&gt; &lt;NULL&gt;   
  8.  4 1994-01-01 &lt;NULL&gt;    &lt;dbl [2]&gt;
  9.  5 1995-01-01 &lt;dbl [2]&gt; &lt;NULL&gt;   
  10.  6 1996-01-01 &lt;NULL&gt;    &lt;dbl [2]&gt;
  11.  7 1997-01-01 &lt;dbl [2]&gt; &lt;NULL&gt;   
  12.  8 1998-01-01 &lt;NULL&gt;    &lt;dbl [2]&gt;
  13.  9 1999-01-01 &lt;dbl [2]&gt; &lt;NULL&gt;   
  14. 10 2000-01-01 &lt;NULL&gt;    &lt;dbl [2]&gt;

答案1

得分: 2

我认为你在生成样本数据的方式上出现了错误。如果这不是你要找的解决方案，我为此道歉，但以下是我认为你可以获得所需输出的方法：

编辑： @s_pike在我之前大约一分钟提供了相同的解决方案，并解释了我们都做了什么更改。抱歉没有提及并感谢他们，现在包括在这里以确保清晰。

问题出在你对rep()函数的调用上。你当前正在重复A和B各十次，总共有20个值在Parameter列中。因为你提供了一个包含10个日期的向量，它将被循环使用以匹配Parameter的长度。由于A和B是依次复制的，每个日期在Date循环时都具有相同的Parameter值，这是重复的原因。

如果你改为使用以下调用：Parameter = c(rep("A", 10), rep("B", 10))，那么Date中的每个值都会获得A和B的Parameter，因此不会出现重复。请参见以下代码：

  1. > set.seed(123)
  2. > DF <- data.frame(Date = seq(as.Date("1991-01-01"),
  3.                             to = as.Date("2000-12-31"), 
  4.                             by = "year"), 
  5.                  Parameter = c(rep("A",10),rep("B",10)),
  6.                  Value = runif(10,1,50))
  8. > DF
  9.          Date Parameter     Value
  10. 1  1991-01-01         A 15.091298
  11. 2  1992-01-01         A 39.626952
  12. 3  1993-01-01         A 21.039869
  13. 4  1994-01-01         A 44.267853
  14. 5  1995-01-01         A 47.082897
  15. 6  1996-01-01         A  3.232268
  16. 7  1997-01-01         A 26.877169
  17. 8  1998-01-01         A 44.728533
  18. 9  1999-01-01         A 28.020316
  19. 10 2000-01-01         A 23.374122
  20. 11 1991-01-01         B 15.091298
  21. 12 1992-01-01         B 39.626952
  22. 13 1993-01-01         B 21.039869
  23. 14 1994-01-01         B 44.267853
  24. 15 1995-01-01         B 47.082897
  25. 16 1996-01-01         B  3.232268
  26. 17 1997-01-01         B 26.877169
  27. 18 1998-01-01         B 44.728533
  28. 19 1999-01-01         B 28.020316
  29. 20 2000-01-01         B 23.374122

这应该符合你的要求，现在你的pivot_wider应该正常工作：

  1. > DF %>%
  2. +     pivot_wider(names_from = Parameter, values_from = Value)
  3. # A tibble: 10 × 3
  4.        Date     A     B
  5.      <date> <dbl> <dbl>
  6. 1  1991-01-01  15.1  15.1 
  7. 2  1992-01-01  39.6  39.6 
  8. 3  1993-01-01  21.0  21.0 
  9. 4  1994-01-01  44.3  44.3 
  10. 5  1995-01-01  47.1  47.1 
  11. 6  1996-01-01   3.23   3.23
  12. 7  1997-01-01  26.9  26.9 
  13. 8  1998-01-01  44.7  44.7 
  14. 9  1999-01-01  28.0  28.0 
  15. 10 2000-01-01  23.4  23.4
英文:

I think you've got an error in how you're generating your sample data. Apologies if this isn't what you're looking for, but here's how I think you can get your desired output:

EDIT: @s_pike provided the same solution below, about a minute before me, with an explanation of what we both changed. Sorry for omission and thanks to them for that- included now for clarity.

The problem is in your call to rep(). You're currently repeating A and B ten times, for a total of 20 values in the Parameter column. Because you're providing a vector of 10 dates, it will be recycled to match the length of Parameter. Because A and B are replicated one after the other, each date has the same Parameter value when Date is recycled: the cause of the duplication.

If instead you change the call to: Parameter = c(rep(&quot;A&quot;,10),rep(&quot;B&quot;,10)) each value in Date gets a Parameter of both A and B, so there are no duplications. See below:

  1. &gt; set.seed(123)
  2. &gt; DF &lt;- data.frame(Date = seq(as.Date(&quot;1991-01-01&quot;),
  3.                             to = as.Date(&quot;2000-12-31&quot;), 
  4.                             by = &quot;year&quot;), 
  5.                  Parameter = c(rep(&quot;A&quot;,10),rep(&quot;B&quot;,10)),
  6.                  Value = runif(10,1,50))
  8. &gt; DF
  9.          Date Parameter     Value
  10. 1  1991-01-01         A 15.091298
  11. 2  1992-01-01         A 39.626952
  12. 3  1993-01-01         A 21.039869
  13. 4  1994-01-01         A 44.267853
  14. 5  1995-01-01         A 47.082897
  15. 6  1996-01-01         A  3.232268
  16. 7  1997-01-01         A 26.877169
  17. 8  1998-01-01         A 44.728533
  18. 9  1999-01-01         A 28.020316
  19. 10 2000-01-01         A 23.374122
  20. 11 1991-01-01         B 15.091298
  21. 12 1992-01-01         B 39.626952
  22. 13 1993-01-01         B 21.039869
  23. 14 1994-01-01         B 44.267853
  24. 15 1995-01-01         B 47.082897
  25. 16 1996-01-01         B  3.232268
  26. 17 1997-01-01         B 26.877169
  27. 18 1998-01-01         B 44.728533
  28. 19 1999-01-01         B 28.020316
  29. 20 2000-01-01         B 23.374122

This should do what you want and your pivot_wider should work now:

  1. &gt; DF %&gt;% 
  2. +     pivot_wider(names_from = Parameter, values_from = Value)
  3. # A tibble: 10 &#215; 3
  4.    Date           A     B
  5.    &lt;date&gt;     &lt;dbl&gt; &lt;dbl&gt;
  6.  1 1991-01-01 15.1  15.1 
  7.  2 1992-01-01 39.6  39.6 
  8.  3 1993-01-01 21.0  21.0 
  9.  4 1994-01-01 44.3  44.3 
  10.  5 1995-01-01 47.1  47.1 
  11.  6 1996-01-01  3.23  3.23
  12.  7 1997-01-01 26.9  26.9 
  13.  8 1998-01-01 44.7  44.7 
  14.  9 1999-01-01 28.0  28.0 
  15. 10 2000-01-01 23.4  23.4

答案2

得分: 1

尝试更改参数中的重复次数为 c(rep("A", times = 10), rep("B", times=10))，假设您的意图是每年有一个 "A" 和一个 "B"。

与您的原始代码进行比较：

  1. library(tidyverse)
  3. set.seed(123)
  4. DF <- data.frame(Date = seq(as.Date("1991-01-01"), 
  5.                             to = as.Date("2000-12-31"), 
  6.                             by = "year"), 
  7.                  Parameter = rep(c("A","B"), times = 10),
  8.                  Value = runif(10,1,50)) %>%
  9.   pivot_wider(names_from = Parameter, values_from = Value)
  11. DF

使用以下代码：

  1. DF <- data.frame(Date = seq(as.Date("1991-01-01"), 
  2.                             to = as.Date("2000-12-31"), 
  3.                             by = "year"), 
  4.                  Parameter = c(rep("A", times = 10), rep("B", times=10)),
  5.                  Value = runif(10,1,50)) %>%
  6.   pivot_wider(names_from = Parameter, values_from = Value)
  8. DF

<sup>创建于2023年08月10日，使用 reprex v2.0.2</sup>

英文:

Try changing the repeat in the Parameter to c(rep(&quot;A&quot;, times = 10), rep(&quot;B&quot;, times=10)), assuming your intention is to have one &quot;A&quot; per year, and one &quot;B&quot; per year.

Compare your original:

  1. library(tidyverse)
  3. set.seed(123)
  4. DF &lt;- data.frame(Date = seq(as.Date(&quot;1991-01-01&quot;), 
  5.                             to = as.Date(&quot;2000-12-31&quot;), 
  6.                             by = &quot;year&quot;), 
  7.                  Parameter = rep(c(&quot;A&quot;,&quot;B&quot;), times = 10),
  8.                  Value = runif(10,1,50)) %&gt;% 
  9.   pivot_wider(names_from = Parameter, values_from = Value)
  11. DF
  12. #&gt; # A tibble: 10 &#215; 3
  13. #&gt;    Date       A         B        
  14. #&gt;    &lt;date&gt;     &lt;list&gt;    &lt;list&gt;   
  15. #&gt;  1 1991-01-01 &lt;dbl [2]&gt; &lt;NULL&gt;   
  16. #&gt;  2 1992-01-01 &lt;NULL&gt;    &lt;dbl [2]&gt;
  17. #&gt;  3 1993-01-01 &lt;dbl [2]&gt; &lt;NULL&gt;   
  18. #&gt;  4 1994-01-01 &lt;NULL&gt;    &lt;dbl [2]&gt;
  19. #&gt;  5 1995-01-01 &lt;dbl [2]&gt; &lt;NULL&gt;   
  20. #&gt;  6 1996-01-01 &lt;NULL&gt;    &lt;dbl [2]&gt;
  21. #&gt;  7 1997-01-01 &lt;dbl [2]&gt; &lt;NULL&gt;   
  22. #&gt;  8 1998-01-01 &lt;NULL&gt;    &lt;dbl [2]&gt;
  23. #&gt;  9 1999-01-01 &lt;dbl [2]&gt; &lt;NULL&gt;   
  24. #&gt; 10 2000-01-01 &lt;NULL&gt;    &lt;dbl [2]&gt;

With:

  1. DF &lt;- data.frame(Date = seq(as.Date(&quot;1991-01-01&quot;), 
  2.                             to = as.Date(&quot;2000-12-31&quot;), 
  3.                             by = &quot;year&quot;), 
  4.                  Parameter = c(rep(&quot;A&quot;, times = 10), rep(&quot;B&quot;, times=10)),
  5.                  Value = runif(10,1,50)) %&gt;% 
  6.   pivot_wider(names_from = Parameter, values_from = Value)
  8. DF
  9. #&gt; # A tibble: 10 &#215; 3
  10. #&gt;    Date           A     B
  11. #&gt;    &lt;date&gt;     &lt;dbl&gt; &lt;dbl&gt;
  12. #&gt;  1 1991-01-01 47.9  47.9 
  13. #&gt;  2 1992-01-01 23.2  23.2 
  14. #&gt;  3 1993-01-01 34.2  34.2 
  15. #&gt;  4 1994-01-01 29.1  29.1 
  16. #&gt;  5 1995-01-01  6.04  6.04
  17. #&gt;  6 1996-01-01 45.1  45.1 
  18. #&gt;  7 1997-01-01 13.1  13.1 
  19. #&gt;  8 1998-01-01  3.06  3.06
  20. #&gt;  9 1999-01-01 17.1  17.1 
  21. #&gt; 10 2000-01-01 47.8  47.8

<sup>Created on 2023-08-10 with reprex v2.0.2</sup>

huangapple
  • 本文由 发表于 2023年8月10日 23:40:18
  • 转载请务必保留本文链接：https://go.coder-hub.com/76877305.html
  • dataframe
  • dplyr
  • pivot
  • r
  • tidyverse
匿名

发表评论

匿名网友

:?: :razz: :sad: :evil: :!: :smile: :oops: :grin: :eek: :shock: :???: :cool: :lol: :mad: :twisted: :roll: :wink: :idea: :arrow: :neutral: :cry: :mrgreen:

确定