2

我在 R 中有下表：

  ExperimentID clients connections msgSize     Mean Deviation     Error
1             4      50          50      10 73.19379 21.313889 0.2263254
2             5      50          10      10 73.25170 21.457139 0.2265276
3             6      50         100      10 73.20642 21.396485 0.2261432
4             7      50          50    1999 53.75247 11.863616 0.1695395
5             8      50          10    1999 53.88464 12.778998 0.2234775
6             9      50         100    1999 53.99422 11.947930 0.2085102
7            10      10          50    1999 49.74034  9.296995 0.3855425
8            11      10          10    1999 49.77624  8.639379 0.3566724
9            12      10         100    1999 50.30912 10.800443 0.4442306
10           13      10          50      10 68.80108 19.674006 0.5892552
11           14      10          10      10 69.41143 19.671618 0.5845524
12           15      10         100      10 69.09130 19.821473 0.5894541
13           16     100          10    1999 56.32045 16.370877 0.1940681
14           17     100          50    1999 55.93405 14.007772 0.2272496

现在，我想将“平均值”列绘制为以下因素的函数：客户端、连接和 msgSize。我认为这样做的合理方法是准备两个直方图，一个用于 msgSize = 10，另一个用于 msgSize = 1999。在这两个直方图中，我可以在 x 轴上显示客户端数量， y轴上的平均值。对于 x 轴上的每个刻度（即，对于特定值的客户端），我可以有 3 个条形图（每个对应于 10、50 或 100 个连接的平均值）。我怎样才能在 R 中实现这一点？

PS：我是R的新手，所以有点详细的答案会有所帮助。如果 ggplot 有更好的答案，那也很好。

4

1 回答 1

2 
data = read.csv(file="home",sep=",")

df = data.frame(data)
df

   ExperimentID clients connections msgSize     Mean Deviation     Error
1             4      50          50      10 73.19379 21.313889 0.2263254
2             5      50          10      10 73.25170 21.457139 0.2265276
3             6      50         100      10 73.20642 21.396485 0.2261432
4             7      50          50    1999 53.75247 11.863616 0.1695395
5             8      50          10    1999 53.88464 12.778998 0.2234775
6             9      50         100    1999 53.99422 11.947930 0.2085102
7            10      10          50    1999 49.74034  9.296995 0.3855425
8            11      10          10    1999 49.77624  8.639379 0.3566724
9            12      10         100    1999 50.30912 10.800443 0.4442306
10           13      10          50      10 68.80108 19.674006 0.5892552
11           14      10          10      10 69.41143 19.671618 0.5845524
12           15      10         100      10 69.09130 19.821473 0.5894541
13           16     100          10    1999 56.32045 16.370877 0.1940681
14           17     100          50    1999 55.93405 14.007772 0.2272496

par(mfrow=c(1,2)) #plot two graphs (Message Size = 10 and Message Size = 1999) side by side.

msg_10 = subset(df, data_df$msgSize == 10)
msg_10

   ExperimentID clients connections msgSize     Mean Deviation     Error
1             4      50          50      10 73.19379  21.31389 0.2263254
2             5      50          10      10 73.25170  21.45714 0.2265276
3             6      50         100      10 73.20642  21.39648 0.2261432
10           13      10          50      10 68.80108  19.67401 0.5892552
11           14      10          10      10 69.41143  19.67162 0.5845524
12           15      10         100      10 69.09130  19.82147 0.5894541


plot(msg_10$Mean ~ msg_10$clients, col=as.factor(msg_10$connections), pch=19,  xlab="Clients", ylab="Mean", xlim=c(0,60), ylim=c(68,74), main="Message Size = 10",cex.main=0.85)
legend("bottomright", legend=unique(msg_10$connections), col=as.factor(msg_10$connections),pch=19,title="connections")

msg_1999 = subset(df, data_df$msgSize == 1999)
msg_1999

   ExperimentID clients connections msgSize     Mean Deviation     Error
4             7      50          50    1999 53.75247 11.863616 0.1695395
5             8      50          10    1999 53.88464 12.778998 0.2234775
6             9      50         100    1999 53.99422 11.947930 0.2085102
7            10      10          50    1999 49.74034  9.296995 0.3855425
8            11      10          10    1999 49.77624  8.639379 0.3566724
9            12      10         100    1999 50.30912 10.800443 0.4442306
13           16     100          10    1999 56.32045 16.370877 0.1940681
14           17     100          50    1999 55.93405 14.007772 0.2272496


plot(msg_1999$Mean ~ msg_1999$clients, col=as.factor(msg_1999$connections), pch=19, xlab="Clients", ylab="Mean", xlim=c(0,100), ylim=c(48,58), main="Message Size = 1999",cex.main=0.85)
legend("bottomright", legend=unique(msg_1999$connections), col=as.factor(msg_1999$connections),pch=19,title="connections")


这是输出：

在此处输入图像描述

编辑. 我忽略了标签。所以这是另一种选择：

库（ggplot2）

#Message Size = 10
ggplot(data=msg_10,aes(clients,Mean)) + geom_point(data=msg_10,aes(color=as.factor(connections)),size=5) + theme_bw() + labs(title="Message Size = 10", color="Connections")

#Message Size = 1999
ggplot(data=msg_1999,aes(clients,Mean)) + geom_point(data=msg_1999,aes(color=as.factor(connections)),size=5) + theme_bw() + labs(title="Message Size = 1999", color="Connections")

第二个图表（消息大小 = 1999）如下所示：

在此处输入图像描述

于 2013-11-08T16:06:03.240 回答