[合集]R語言繪圖（ggplot2，ggpubr）從入門到精通
[合集]R語言繪圖（ggplot2，ggpubr）從入門到精通
# * 第一章：快速探索數(shù)據(jù) ------------------------------------------------------------
###################
## 散點圖
# 最簡單的函數(shù):plot(x, y)
plot(mtcars$wt,mtcars$mpg)
# 使用qplot
library(ggplot2)
qplot(mtcars$wt,mtcars$mpg)?
qplot(wt, mpg, data = mtcars)
# 使用ggplot2
ggplot(mtcars, aes(x = wt, y = mpg)) +?
?geom_point()
###################
## 折線圖、曲線圖
# 同樣可以用plot來畫:
# plot中type參數(shù)可以指定繪圖的類型:如:"7"就是指折、曲線圖
??
plot(pressure$temperature, pressure$pressure, type = "l")?
points (pressure$temperature, pressure$pressure)
lines(pressure$temperature, pressure$pressure/2, col ="red")?
points(pressure$temperature, pressure$pressure/2, col = "red")
# 使用qplot
library (ggplot2)
qplot (pressure$temperature, pressure$pressure, geom = "line")
qplot(temperature, pressure,data = pressure, geom = "line") #如果變量x和y 都來自于同一個數(shù)據(jù)框,還可以這樣寫；
# 使用ggplot2
ggplot(pressure, aes(x = temperature, y = pressure)) +?
?geom_line() +
?geom_point()
###################
## 柱狀圖
#使用barplot
BOD<- BOD
mtcars <- mtcars
barplot(BOD$demand,names.arg = BOD$Time)
# 有時“條形圖”指的是一個圖表，其中的條形圖代表列每個類型的案例數(shù)
# 這類似與直方圖，但是x軸是離散的，而不是連續(xù)的，這個時候要用table函數(shù)生成每個類別的計數(shù)
barplot(mtcars$cyl)
barplot(table(mtcars$cyl))
# 使用qplot()
library(ggplot2)
qplot(mtcars$cyl) #連續(xù)變量
qplot(factor(mtcars$cyl)) #分類變量
# ggplot
ggplot(mtcars,aes(cyl))+geom_bar()
# stat = "count"(默認)：表示一個x對應落到該x的樣本數(shù)
# stat = "identity": 表示一個x對應一個y
# 說白了就是，identity提取橫坐標x對應的y值，count提取橫坐標的頻數(shù)
ggplot(BOD,aes(Time,demand))?+?geom_bar(stat = "identity")
ggplot(BOD,aes(factor(Time),demand))?+?geom_bar(stat = "identity")
###################
## 直方圖
# x是連續(xù)變量
#基礎方法：hist函數(shù)
hist(mtcars$mpg)
hist(mtcars$mpg, breaks = 10)
#qplot()
qplot(mpg,data = mtcars, binwidth = 1 ) #binwidth參數(shù)指定組距
#等價于
ggplot(data = mtcars, aes(x = mpg)) + geom_histogram( binwidth = 2)
###################
## 箱線圖
#使用plot（）函數(shù)，當x為因子變量（與數(shù)值變量對應時），默認繪制箱線圖
ToothGrowth <- ToothGrowth
plot(ToothGrowth$supp, ToothGrowth$len)
#使用公式語法
boxplot(len ~ supp, data = ToothGrowth)
#在x軸引入兩變量的交互，可以繪制分組箱線圖
boxplot(len ~ supp + dose, data = ToothGrowth)
#qplot()繪制箱線圖
qplot(supp,len,data = ToothGrowth, geom = "boxplot")
#使用三個獨立的向量參數(shù)
qplot(interaction(supp,dose),len, data = ToothGrowth,geom = "boxplot")
#等價與
ggplot(data = ToothGrowth, aes(x = interaction(supp,dose), y = len)) + geom_boxplot()
###################
## 繪制函數(shù)圖像
#使用curve()函數(shù)繪制，傳入一個關于變量x的表達式
curve(x^3 - 5*x, from = -4, to =4)
#自定義函數(shù)圖像
my_fun <- function(xvar){
?1/(1 + exp(-xvar + 10))
}
curve(my_fun(x), from =?0, to =20)
#原有的基礎上添加一條線
curve(1 - my_fun(x),add = T, col = "red")
#等價于
ggplot(data.frame(x = c(0,20)),aes(x =x)) + stat_function(fun = my_fun, geom ="line")
my_fun2 <- function(xvar){
?1- 1/(1 + exp(-xvar + 10))
}
ggplot(data.frame(x = c(0,20)),aes(x =x)) + stat_function(fun = my_fun, geom ="line") + stat_function(fun = my_fun2, geom ="line",color="red")
# * 第二章：柱狀圖深入研究 ------------------------------------------------------------
library(gcookbook)
pg_mean <- pg_mean
# 修改柱狀圖的填充（fill）和描邊（color）：
ggplot(pg_mean,aes(group, weight)) +
?geom_bar(stat = "identity", fill = "lightblue", color = "black")
# 也可通過分組變量設置顏色：
cabbage_exp <- cabbage_exp
ggplot(cabbage_exp, aes(Date, fill = Cultivar)) +
?geom_bar(position =?"dodge")
?#拓展：postition參數(shù)： 主要是指對圖像的微調，最常見的應用是在分組的柱形圖（bar）中，因為分組的柱形圖會產(chǎn)生祖內堆積和不堆積兩種主要的效果
?# - position常用參數(shù)值：其中stack和dodge最為常用。
?#?- "identity": 不調整，組內前后重疊；
?#?- "stack": 堆積， 默認；
?#?- "fill": 按比例堆積；
?#?- "dodge": 分散??
?# - ColorBrewer 配色，使用的是scale_colour_brewer() 、scale_fill_brewer()。 想要了解所有的調色板，可以使用RColorBrewer::display.brewer.all()查看
# 通過scale_fill_brewer()修改顏色模式：
ggplot(cabbage_exp, aes(Date, Weight, fill = Cultivar)) +
?geom_bar(position =?"dodge", stat = "identity", color = "black") +
?scale_fill_brewer(palette = "Pastel1")
# 可以使用以下代碼查看調色板
RColorBrewer::display.brewer.all()
###################
## 分組柱狀圖
library(gcookbook)
cabbage_exp <- cabbage_exp
ggplot(cabbage_exp, aes(Date, Weight, fill = Cultivar)) +
?geom_bar(position = "dodge", stat = "identity", color = "black")
ggplot(cabbage_exp, aes(Date, Weight, fill = Cultivar)) +?
?geom_bar(stat = "identity") # 默認的position = "stack";
# 此外，除了fill可以指定分組變量，color, linestyle也可以指定；
ggplot(cabbage_exp, aes(Date, Weight, color = Cultivar)) +?
?geom_bar(position = "dodge", stat = "identity")
# 請注意，如果類別變量的組合有任何缺失，則該欄將缺失，相鄰的欄將擴展
# 以填充該空間。
ce =?cabbage_exp[1:5,]
ggplot(ce, aes(Date, Weight, fill = Cultivar)) +
?geom_bar(position = "dodge", stat = "identity", color = "black") +
?scale_fill_brewer(palette = "Pastel1")
# 實際情況下確實存在有一種類別沒有對應的y值，此時可以使用NA或者0代替
ce_NA <- cabbage_exp
ce_NA$Weight[6] <- 0
ggplot(ce_NA, aes(Date, Weight, fill = Cultivar)) +
?geom_bar(position = "dodge", stat = "identity", color = "black") +
?scale_fill_brewer(palette = "Pastel1")
###################
## 修改顏色的技巧
library(gcookbook)
upc <- subset(uspopchange, rank(Change) > 40)
upc
ggplot(upc, aes(x = Abb, y = Change, fill = Region)) +?
?geom_bar(stat = "identity")
# 此示例還使用reorder()函數(shù)，將條形按其高度進行排序：
ggplot(upc, aes(x = reorder(Abb, change), y = Change, fill = Region)) +??
?geom_bar(stat = "identity", color = "black") +
?scale_fill_manual(values = c("#669933", "#FFCC66")) +?
?xlab("State")
??#拓展：reorder()函數(shù)：?
??# reorder(x, X, FUN = mean, ..., order = is.ordered(x), decreasing = FALSE)
??# x: x是要排序的數(shù)據(jù)，排序后的結果作用于x上
??# X: 要根據(jù)X進行排序
??#拓展：scale_fill_manual()函數(shù):
??#自定義顏色：輸入的變量長度與分組變量的長度一致
??#scale_fill_manual(..., values, aesthetics = "fill", breaks = waiver(), na.value = "grey50")
###################
## 正負兩極不同的顏色
csub <- subset(climate, Source == "Berkeley" & Year >= 1900)
# 思路：顏色根據(jù)正負數(shù)來填充？怎么識別正負數(shù)呢？
# 只能新建一個字段；該字段描述來正和負；
# 指定fill為該字段；
csub$pos <- csub$Anomaly10y >= 0?
csub
ggplot(csub, aes(x = Year, y = Anomaly10y, fill = pos)) +
?geom_bar(stat = "identity", position = "identity")
# 使用scale_fill_manual()修改顏色；guide = FALSE參數(shù)去掉圖例
ggplot(csub, aes(x = Year, y = Anomaly10y, fill = pos)) +?
?geom_bar(stat = "identity", position = "identity", color = "black", size = 0.25) +
?scale_fill_manual(values = c("#CCEEFF", "#FFDDDD"), guide = FALSE)
??
###################
## 調整條形的寬度和間距（width參數(shù)）：
library(gcookbook)
# width默認是0.9
ggplot(pg_mean, aes(x = group, y = weight)) +
?geom_bar(stat = "identity")?
ggplot(pg_mean, aes(x = group, y = weight)) +?
?geom_bar(stat = "identity", width = 0.5)
# width最大值只能設置為1:
ggplot(pg_mean, aes(x = group, y = weight)) +?
?geom_bar(stat = "identity", width = 1)
## 調節(jié)分組條形圖之間的間距：
# 默認的同一分組之間的條形是沒有間距的：
ggplot(cabbage_exp, aes(x = Date, y = Weight, fill = Cultivar)) +
?geom_bar(stat = "identity", width = 0.5, position = "dodge")
# 只需要將position_dodge參數(shù)設置的比width參數(shù)大一些就好了！
ggplot(cabbage_exp, aes(x = Date, y = Weight, fill = Cultivar)) +
?geom_bar(stat = "identity", width = 0.5, position = position_dodge(0.7))
ggplot(cabbage_exp, aes(x = Date, y = Weight, fill = Cultivar)) +
?geom_bar(stat = "identity", width = 0.5, position = position_dodge(0.3))
ggplot(cabbage_exp, aes(x = Date, y = Weight, fill = Cultivar)) +
?geom_bar(stat = "identity", width = 0.5, position = position_dodge(0))
# 思考：position_dodge有什么含義？為什么比width大就會有間隙嗎？
?# position_dodge是從一個柱子的右邊到另一個柱子的右邊的距離
?# 因為默認的position_dodge()里的內容一定是和width相等的
?
###################
## 堆積柱狀圖：
# position的默認值為stack；
# 即如果不設置position，并且設置了分組變量，就是畫堆積圖；
library(gcookbook)
ggplot(cabbage_exp, aes(x = Date, y = Weight, fill = Cultivar)) +
?geom_bar(stat = "identity")
# 修改圖例堆積的順序：guides()
ggplot(cabbage_exp, aes(x = Date, y = Weight, fill = Cultivar)) +
?geom_bar(stat = "identity") +
?guides(fill = guide_legend(reverse = T))
# 修改圖形堆積順序：修改因子水平
cabbage_exp$Cultivar <- factor(cabbage_exp$Cultivar, levels = c("c52", "c39"))
ggplot(cabbage_exp, aes(x = Date, y = Weight, fill = Cultivar)) +
?geom_bar(stat = "identity")
ggplot(cabbage_exp, aes(x = Date, y = Weight, fill = Cultivar)) +
?geom_bar(stat = "identity", color = "black") +?
?guides(fill = guide_legend(reverse = T)) +
?scale_fill_brewer(palette = "Pastel1")
###################
## 修改標簽：
library(ggplot2)
library(gcookbook)
cabbage_exp <- cabbage_exp
# 標簽位置的設定，vjust參數(shù)：
ggplot(cabbage_exp, aes(x = interaction(Date, Cultivar), y = Weight)) +
?geom_bar(stat = "identity") +
?geom_text(aes(label = Weight), vjust = 1.5, colour = "white")
ggplot(cabbage_exp, aes(x = interaction(Date, Cultivar), y = Weight)) +?
?geom_bar(stat = "identity") +
?geom_text(aes(label = Weight), vjust = -0.2, colour = "white")
# 為了防止標簽跑出圖形，可以調整y軸的范圍：
# 方法一：ylim()函數(shù)：
ggplot(cabbage_exp, aes(x = interaction(Date, Cultivar), y = Weight)) +
?geom_bar(stat = "identity") +
?geom_text(aes(label = Weight), vjust = -0.2) +
?ylim(0, max(cabbage_exp$Weight) * 1.05)
# 方法二：以weight為基準，調節(jié)y值，圖形高度會自動適配：
ggplot(cabbage_exp, aes(x = interaction(Date, Cultivar), y = Weight)) +?
?geom_bar(stat = "identity") +
?geom_text(aes(y = Weight + 0.1, label = Weight))
# 分組柱狀圖加標簽：需要設定position_dodge(),以調整字體適合位置
ggplot(cabbage_exp, aes(x = interaction(Date, Cultivar), y = Weight, fill = Cultivar)) +?
?geom_bar(stat = "identity", position = "dodge") +?
?geom_text(aes(label = Weight), vjust = 1.5, color = "white", position = position_dodge(0.9), size = 3)?
?# width參數(shù)默認是0.9，position_dodge(0.9)，這樣子可以讓標簽位于柱子中間
library(plyr)
# 堆積柱狀圖添加label
ce <- arrange(cabbage_exp, Date, Cultivar)
ce <- ddply(ce, "Date", transform, label_y = cumsum(Weight))
ce$Cultivar <- factor(ce$Cultivar, levels = c("c52", "c39"))
ggplot(ce, aes(x = Date, y = Weight, fill = Cultivar)) +?
?geom_bar(stat = "identity") +?
?geom_text(aes(y = label_y, label = Weight), vjust = 1.5, colour = "white")
# 修改標簽至中央
ce <- arrange(cabbage_exp, Date, Cultivar)
ce <- ddply(ce, "Date", transform, label_y = cumsum(Weight) - 0.5 * Weight)
ce$Cultivar <- factor(ce$Cultivar, levels =?c("c52", "c39"))
ggplot(ce, aes(x = Date, y = Weight, fill = Cultivar)) +?
?geom_bar(stat = "identity") +?
?geom_text(aes(y = label_y, label = Weight), vjust = 1.5, colour = "white")
# 添加單位，并修改顏色模式
ggplot(ce, aes(x = Date, y = Weight, fill = Cultivar)) +
?geom_bar(stat = "identity", colour = "black") +
?geom_text(aes(y = label_y, label = paste(format(Weight,nsmall = 2), "kg")), size = 4) +
?scale_fill_brewer(palette = "Pastel1")
###################
## 柱狀圖的拓展：克里夫蘭點圖繪制
library(gcookbook)
tophit <- tophitters2001[1:25,]
# 從最基本的散點圖出發(fā)
ggplot(tophit, aes(x = avg, y = name)) +
?geom_point()
# reorder排序一定要熟練：前面柱狀圖排序講過
ggplot(tophit, aes(x = avg, y = reorder(name, avg))) +?
?geom_point(size = 3) + #修改點的大小
?theme_bw() + #修改背景
?theme(panel.grid.major.x = element_blank(), #設置縱向網(wǎng)格線為空
????panel.grid.minor.x = element_blank(),?#設置橫向網(wǎng)格線為虛線（dashed）
????panel.grid.major.y = element_line(colour = "grey60", linetype = "dashed"))
# 顛倒圖形的x軸和y軸
ggplot(tophit, aes(x = reorder(name, avg), y =?avg))+?
?geom_point(size = 3) + #修改點的大小
?theme_bw() + #修改背景
?theme(panel.grid.major.y = element_blank(),?
????panel.grid.minor.y = element_blank(),?
????panel.grid.major.x = element_line(colour = "grey60", linetype = "dashed"))
# 按照lg和avg對name進行排序，先按lg排，再按avg排
nameorder <- tophit$name[order(tophit$lg, tophit$avg)]
# 將name變量轉化為因子，因子水平設定位nameorder：
tophit$name <- factor(tophit$name, levels = nameorder)
## 繪制彩色克利夫蘭點圖：
ggplot(tophit, aes(x = avg, y = name)) +
?geom_segment(aes(yend = name), xend = 0, colour = "grey50") +
?geom_point(size = 3, aes(color = lg)) + # 設置分組變量
?# limits限定顏色先后順序：
?scale_colour_brewer(palette = "Set1", limits = c("NL", "AL")) +
?theme_bw() +?
?theme(panel.grid.major.y = element_blank(), #去除橫線網(wǎng)格線
????legend.position = c(1, 0.55), # 設置圖例的位置：這里的1指的是與x軸的比例；
????#legend.justification = c(1, 0.5) #表示圖例右邊緣中點；
????# (1, 0) 表示右下角，(0, 1) 表示左上角，以此類推；
????legend.justification = c(1, 0.5))
?
# 分面繪制：facet_grid()函數(shù)：
?ggplot(tophit, aes(x = avg, y = name)) +
??geom_segment(aes(yend = name), xend = 0, colour = "grey50") +
??geom_point(size = 3, aes(color = lg)) + # 設置分組變量
?# guides去除圖例：
?scale_colour_brewer(palette = "Set1", limits = c("NL", "AL"), guide = F) +
??theme_bw() +
??theme(panel.grid.major.y = element_blank()) +
??# 一列多行：lg~ 行數(shù)等于lg的種類數(shù)目；
??# scales設置每個分塊的單位寬度；space設置每個分塊的寬度；
??facet_grid(lg~.,scales = "free_y", space = "free_y")
??
??
??
# * 第三章：各式各樣的餅圖 ------------------------------------------------------------
library(ggplot2)
# 技巧篇--ggplot繪制各種餅圖：
mpg <- mpg
ggplot(mpg, aes(class)) +
?geom_bar()
# 把y軸方向扭曲來，柱子都邊成了彎的：
ggplot(mpg, aes(class)) +
?geom_bar() +
?coord_polar(theta = "y")
# 把x軸方向扭曲來，柱子都從一個中心出發(fā)：
ggplot(mpg, aes(class)) +
?geom_bar() +
?coord_polar(theta = "x")
# 加上顏色分組：
ggplot(mpg, aes(class)) +
?geom_bar(aes(fill = drv)) +?
?coord_polar(theta = "y")
# 加上顏色分組：
ggplot(mpg, aes(class)) +
?geom_bar(aes(fill = drv)) +?
?coord_polar(theta = "x")
# 如何繪制正常的餅圖？
ggplot(mpg, aes(1, fill = class)) +
?geom_bar(width = 0.5)?
ggplot(mpg, aes(1, fill = class)) +
?geom_bar(width = 0.5) +?
?coord_polar(theta = "y")
# 加上標簽：
ggplot(mpg, aes(1, fill = class)) +
?geom_bar(width = 0.5) +?
?coord_polar(theta = "y") +
?geom_text(stat = "count", aes(label = scales::percent(..count../100)),
??????size = 3, position = position_stack(vjust = 0.5))
# 課后作業(yè)：如何使用position_stack()修改標簽位置呢？
ce <- arrange(cabbage_exp, Date, Cultivar)
ce <- ddply(ce, "Date", transform, label_y = cumsum(Weight) - 0.5 * Weight)
ce$Cultivar <- factor(ce$Cultivar, levels =?c("c52", "c39"))
ggplot(ce, aes(x = Date, y = Weight, fill = Cultivar)) +?
?geom_bar(stat = "identity") +?
?geom_text(aes(y = label_y, label = Weight), vjust = 1.5, colour = "white")
# 使用position_stack()可以大大節(jié)約我們的代碼：
ggplot(cabbage_exp, aes(x = Date, y = Weight, fill = Cultivar)) +
?geom_bar(stat = "identity") +?
?geom_text(aes(label = Weight), position = position_stack(vjust = 0.5), colour = "white")?
