简体   繁体   English

具有2D马赛克的labeling_cells函数

[英]labeling_cells function with 2d mosaic

There seems to be something (many things?) I don't understand about how to use mosaic in package vcd and labeling_cells because I keep getting an error. 似乎有些事情(很多事情?)我不了解如何在软件包vcdlabeling_cells使用mosaic ,因为我不断遇到错误。

Setup: 设定:

I have a 2d array of values (row and column names changed to protect the innocent): 我有一个二维数组(更改行和列名以保护无辜):

tmparray
       C1    C2   C3    C4    C5    C6
P1  65241 15534 5754 12747 57902 48367
P2    205 28581 4819  7617  5504 34656
P3     27  3503 1473   757  3308 15363
P4 417580  5967 5353   636   514  9758
P5  31482   915  249   914   328  1301
P6  75501  4307  294   176    27  8376
P7  43589  2864  465   179    61 12631
P8  64095  1008  335   288     3  7438

mosaic(tmparray) gets me the expected plot (rep too low to post image). mosaic(tmparray)我提供了预期的绘图(rep太低而无法发布图像)。

I would like to add text labels for each cell showing only the values in tmparray > 1000 . 我想为每个单元格添加文本标签,仅显示tmparray > 1000的值。 Following the documentation for labeling_cells , I try the following: 按照labeling_cells的文档,我尝试以下操作:

celltxt <- ifelse(tmparray < 1000, NA, tmparray/1000)
celltxt
        C1     C2    C3     C4     C5     C6
P1  65.241 15.534 5.754 12.747 57.902 48.367
P2      NA 28.581 4.819  7.617  5.504 34.656
P3      NA  3.503 1.473     NA  3.308 15.363
P4 417.580  5.967 5.353     NA     NA  9.758
P5  31.482     NA    NA     NA     NA  1.301
P6  75.501  4.307    NA     NA     NA  8.376
P7  43.589  2.864    NA     NA     NA 12.631
P8  64.095  1.008    NA     NA     NA  7.438

labeling_cells(text = celltxt)(tmparray)

Which results in the following error: 导致以下错误:

# Error in ifelse(abbreviate_varnames, sapply(seq_along(dn), function(i) abbreviate(dn[i],  : replacement has length zero

Can anyone please point out how to get the labeling to work? 谁能指出如何使标签生效?

OK, so I'm assuming class(tmparray)=="matrix" . 好的,所以我假设使用class(tmparray)=="matrix" That seems to be fine for plotting, but it looks like if you want to use labeling_cells you're going to want to convert that to a table, being sure to add named dimensions. 这似乎可以很好地进行绘制,但是如果您要使用labeling_cells ,则需要将其转换为表格,并确保添加命名维度。 You can do that with 你可以这样做

tmparray <- as.table(tmparray)
names(dimnames(tmparray))<-c("P","C")

So when you create celltext now with 所以,当你创建celltext现在

celltxt <- ifelse(tmparray < 1000, NA, tmparray/1000)

It will also be a table with the same dimnames . 它也将是一个具有相同dimnames的表。

The other important thing you must do is pass pop=F to mosaic . 您必须做的另一件事是将pop=F传递给mosaic This will leave the grid viewport structure in place so that the labeling_cells function can find the correct locations for the labels. 这样会将网格视口结构保留在适当的位置,以便labeling_cells函数可以找到labeling_cells的正确位置。 So you can create the plot with 所以你可以用

mosaic(tmparray , pop=F)
labeling_cells(text = celltxt, margin=0)(tmparray)

Which results in 导致 带标签的马赛克图

(I did round the values to one decimal place to try to make things a bit neater) (我确实将这些值四舍五入到小数点后一位,以使内容更整洁)

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM