R语言进行EDA ——ggplot 图形分面、直方图 和频率多边形

分面通常使用绘图方法+

①facet_wrap(~varible)/facet_wrap(formula)  较适用于单个变量

②facet_grid(vertical ~ horizion)/facet_grid(formula)  较适用于多个变量

详细讲解可参考 http://www.cookbook-r.com/Graphs/Facets_(ggplot2)/

其他图形调整

1、转换数据

### Transforming Data
Notes: 数据转换
```{r}
library(gridExtra) 
library(ggplot2)
qplot(x=friend_count, data = pf)
summary(pf$friend_count)
summary(log10(pf$friend_count + 1))
summary(sqrt(pf$friend_count))

friend_count <- ggplot(aes(x = friend_count), data = pf) +
  geom_histogram(binwidth = 10, color = I("black"), fill = I("#F49045")) +
  scale_x_continuous(limits = c(0,2000))
friend_count_log <- ggplot(aes(x = log10(pf$friend_count+1)), data = pf) +
  geom_histogram(binwidth = 10, color = I("black"), fill = I("#C49045")) +
  scale_x_continuous(limits = c(0,5)) +
  xlab("log friend count")
friend_count_sqrt <- ggplot(aes(x = sqrt(pf$friend_count)), data = pf) +
  geom_histogram(binwidth = 10, color = I("black"), fill = I("#C49045")) +
  scale_x_continuous(limits = c(0,40)) +
  xlab("sqrt friend count")
grid.arrange(friend_count, friend_count_log, friend_count_sqrt, ncol = 2)


library(ggplot2)
p1 <- ggplot(aes(x = friend_count), data = pf) + geom_histogram()
p2 <- p1 + scale_x_log10() #log转换，qplot也可以这样做 ————标度层法
p3 <- p1 + scale_x_sqrt() # sqrt 转换
grid.arrange(p1,p2,p3,ncol=3)
```

在一个图像中输出多个图形方法：

首先下载程序包：

1	install.packages (