如何在qplot（）抖動圖表中的2個不同列中拆分數據？

您好我有這樣一組數據：如何在qplot（）抖動圖表中的2個不同列中拆分數據？

my.data=read.table(text="   values tag class 
1 100.0000000 X3 CS 
2 100.0000000 X5 CS 
3 54.2370601 X7 CS 
4 63.4813701 X7 CS 
5 51.0402580 X7 CS 
6 60.1430236 X7 CS 
7 70.3939619 X7 CS 
8 56.5981165 X7 CS 
9 75.4148038 X7 CS 
10 88.2687052 X7 CS 
11 70.9697583 X7 CS 
12 54.2014007 X7 CS 
13 63.4325244 X7 CS 
14 51.0086771 X7 CS 
15 60.1034812 X7 CS 
16 70.3397973 X7 CS 
17 56.5630966 X7 CS 
18 75.3652206 X7 CS 
19 88.2007869 X7 CS 
20 70.9258461 X7 CS 
21 38.6325912 X9 CS 
22 27.7455102 X9 CS 
23 21.1778773 X9 CS 
24 100.0000000 X9 CS 
25 73.0803007 X9 CS 
26 55.7814809 X9 CS 
27 85.8614803 X9 CS 
28 38.5666461 X9 CS 
29 27.7114796 X9 CS 
30 21.1580450 X9 CS 
31 72.9906659 X9 CS 
32 55.7292434 X9 CS 
33 85.7810739 X9 CS 
34 0.0000000 XAS CS 
35 49.0701633 XPW CS 
36 40.0228822 XPW CS 
37 23.8702273 XPW CS 
38 100.0000000 XPW CS 
39 89.3022370 XPW CS 
40 53.2611491 XPW CS 
41 69.9892886 XPW CS 
42 0.0000000 XPW CS 
43 14.8840750 X3 CD 
44 17.7316138 X3 CD 
45 6.1164435 X3 CD 
46 0.0000000 X3 CD 
47 1.1435141 X3 CD 
48 14.8904265 X3 CD 
49 17.7375474 X3 CD 
50 6.1241709 X3 CD 
51 1.1506441 X3 CD 
52 14.6751282 X3 CD 
53 17.5364297 X3 CD 
54 5.8621689 X3 CD 
55 0.9089743 X3 CD 
56 74.4165986 X5 CD 
57 76.7441666 X5 CD 
58 80.9582780 X5 CD 
59 58.5811936 X5 CD 
60 62.3494607 X5 CD 
61 69.1719883 X5 CD 
62 57.2509995 X5 CD 
63 61.1402872 X5 CD 
64 68.1819250 X5 CD 
65 74.3887236 X5 CD 
66 76.7211353 X5 CD 
67 80.9428400 X5 CD 
68 58.5360647 X5 CD 
69 62.3121736 X5 CD 
70 69.1469947 X5 CD 
71 57.2044213 X5 CD 
72 61.1018026 X5 CD 
73 68.1561287 X5 CD 
74 74.3425809 X5 CD 
75 76.6830205 X5 CD 
76 80.9173038 X5 CD 
77 58.4613610 X5 CD 
78 62.2504669 X5 CD 
79 69.1056522 X5 CD 
80 57.1273184 X5 CD 
81 61.0381141 X5 CD 
82 68.1134585 X5 CD 
83 66.3076784 X7 CD 
84 60.5650688 X7 CD 
85 68.2935472 X7 CD 
86 49.3767792 X7 CD 
87 40.7484217 X7 CD 
88 52.3605769 X7 CD 
89 36.4202616 X7 CD 
90 25.5835605 X7 CD 
91 40.1677332 X7 CD 
92 66.3298303 X7 CD 
93 60.5954120 X7 CD 
94 68.3131654 X7 CD 
95 49.4100626 X7 CD 
96 40.7940127 X7 CD 
97 52.3900536 X7 CD 
98 36.4620636 X7 CD 
99 25.6408202 X7 CD 
100 40.2047541 X7 CD 
101 66.1477122 X9 CD 
102 75.6876522 X9 CD 
103 81.4426220 X9 CD 
104 55.0173821 X9 CD 
105 67.6939692 X9 CD 
106 75.3411217 X9 CD 
107 51.6125069 X9 CD 
108 65.2486246 X9 CD 
109 73.4746140 X9 CD 
110 66.2054974 X9 CD 
111 75.7174719 X9 CD 
112 81.4600003 X9 CD 
113 55.0941665 X9 CD 
114 67.7335933 X9 CD 
115 75.3642139 X9 CD 
116 51.6951034 X9 CD 
117 65.2912480 X9 CD 
118 73.4994541 X9 CD 
119 25.6270138 XAS CD 
120 45.2920118 XAS CD 
121 44.5101287 XAS CD 
122 0.0000000 XAS CD 
123 17.9916840 XAS CD 
124 16.8196263 XAS CD 
125 7.3447585 XPW CD 
126 24.4280115 XPW CD 
127 54.9277702 XPW CD 
128 0.0000000 XPW CD 
129 4.6378241 XPW CD 
130 43.1246155 XPW CD 
131 39.7525599 XPW CD 
132 100.0000000 XPW CD")

繪製與qplot這些數據（）給了我下面的結果：

qplot(tag,values,colour=factor(class),data=my.data,geom="jitter")+stat_summary(fun.y='mean', geom='errorbarh', aes(xmin=as.integer(tag)-0.3, xmax=as.integer(tag)+0.3), height=0)

jitter plot

我想分裂藍紅色從相同的列分成2個獨立的列，右對齊的列給予類似的東西，但我的抖動點：

example

因爲我們在它，你知道如何減少列的寬度（減少從給定的x給予點的寬度空間），使列變窄嗎？另外如何增加列之間的空間，甚至是一個水平條，以便我們清楚地看到不同的列？

來源

2013-06-24 Wicelo

看看'facet_grid' –

'facet_grid（。〜class）'將點分成2個方面，有沒有辦法像我實際銷售的例子那樣做一個集羣圖？ – Wicelo

現在還沒有時間去挖掘，但這兩個問題可能會有所幫助：http://stackoverflow.com/questions/10493084/ggplot2-jitter-and-position-dodge-together和http：// stackoverflow .com/questions/8510003/can-one-offset-jitter-points-in-ggplot-boxplot - 本質上，你需要同時抖動和閃避你的觀點，或者找出如何做特定組的偏移量;後一種選擇可能意味着將它們作爲兩個獨立的geom，因此我會首先使用一些Google gogplot抖動和閃避。 –

這是一個解決方案。它的靈感來自於this answer的一般想法。

# create an interaction of both factors 
int <- with(my.data, interaction(class, tag)) 
# create a numeric variable (tag2 is used instead of tag) 
my.data <- transform(my.data, tag2 = as.numeric(int)) 
# calculate positions of x-axis breaks 
breaks <- seq_along(levels(int))[c(TRUE, FALSE)] + .5 

library(ggplot2) 
ggplot(data = my.data, aes(x = tag2, y = values, color = as.factor(class))) + 
    geom_jitter(position = position_jitter(width = .4)) + 
    stat_summary(fun.y = 'mean', geom = 'errorbarh', 
       aes(xmin = tag2 - 0.4, xmax = tag2 + 0.4), height = 0) + 
    scale_x_continuous(breaks = breaks, labels = levels(my.data$tag))

enter image description here

來源

2013-06-24 16:40:43

好的工作，這就是我想要的。太糟糕了，我們不能在ggplot2中做到這一點，或者用藍色和紅色列做更好的聚類。信息流程與分面非常相似，所以我不知道他們爲什麼沒有實現它。 – Wicelo

看看這會工作。。。

ggplot(data=my.data, aes(x = factor(tag), fill = factor(class), y = values)) + 
    geom_dotplot(binaxis = "y", stackdir="center", position = "dodge", size = 1, 
       binwidth = 4, stackratio = .5) + 
    stat_summary(fun.y='mean', geom='errorbarh', aes(
    xmin=as.integer(tag)-0.3, xmax=as.integer(tag)+0.3), height=0) + 
    scale_fill_brewer(palette="Set1")

jitter points

我發現這個挖通過on-line documentation。

來源

2013-06-24 16:55:09

如何在qplot（）抖動圖表中的2個不同列中拆分數據？

回答

相關問題