关联数据图-R/python

关联数据大致可以分为以下几个图形来表示

• 散点图（两变量，多变量）
• 抖动图
• 计数图
• 气泡图
• 边际直方图/箱线图

数据前处理

R数据

library(ggplot2)
library(plotrix)
library(ggExtra)

## 全局主题设置
options(scipen=999)  # 关掉像 1e+48 这样的科学符号
# 颜色设置（灰色系列）
cbp1 <- c("#999999", "#E69F00", "#56B4E9", "#009E73",
          "#F0E442", "#0072B2", "#D55E00", "#CC79A7")

# 颜色设置（黑色系列）
cbp2 <- c("#000000", "#E69F00", "#56B4E9", "#009E73",
          "#F0E442", "#0072B2", "#D55E00", "#CC79A7")


ggplot <- function(...) ggplot2::ggplot(...) + 
  scale_color_manual(values = cbp1) +
  scale_fill_manual(values = cbp1) + # 注意: 使用连续色阶时需要重写
  theme_bw()

python数据

import numpy as np
import pandas as pd
import matplotlib as mpl
import matplotlib.pyplot as plt
import seaborn as sns
import warnings
#warnings.filterwarnings(action='once')

# 主题设置
plt.style.use('seaborn-whitegrid')
sns.set_style("whitegrid")
#print(mpl.__version__)# 3.5.1
#print(sns.__version__)# 0.12.0

散点图

该图展示两个变量间的关系。

• R

data("midwest", package = "ggplot2") #加载数据集

ggplot(midwest, aes(x=area, y=poptotal)) + 
  geom_point(aes(col=state, size=popdensity)) + 
  geom_smooth(method="loess", se=F) + 
  xlim(c(0, 0.1)) + 
  ylim(c(0, 500000)) + 
  labs(subtitle="Area Vs Population", 
       y="Population", 
       x="Area", 
       title="Scatterplot", 
       caption = "Source: midwest")

• python

midwest = pd.read_csv("/Volumes/Flower0501/研究学习目录/图形及其代码/data/midwest.csv")

sns.set(style="ticks",font_scale=1.5)#设置绘图风格
ax = sns.relplot(x="area",y="poptotal",data=midwest,
            height=7,
            hue='state',#添加分类变量
            #**dict(s=20),#传入词典，调用Matplotlib中的参数,
            size='popdensity',
            legend='brief',#图例设置,"brief","full"
            #col="state",#设置分面的行坐标
            #row="popdensity" # 设置分面的纵坐标
           )

ax.set_axis_labels("Area","Population").set(xlim=(0,0.1),ylim=(0, 500000))
plt.text(x=0.045, y=520000, s='Scatterplot', fontsize=16, weight='bold')
plt.text(x=0.045, y=500050, s='Area Vs Population', fontsize=10, alpha=0.75)
plt.show()

python的散点图只能对每一个分组添加趋势线，而没有办法对总的添加（至少目前我办不到）

sns.set(style="ticks",font_scale=1.5)#设置绘图风格
ax = sns.lmplot(x="area",y="poptotal",data=midwest,
            height=7,
            hue='state',#添加分类变量
            #**dict(s=20),#传入词典，调用Matplotlib中的参数,
            legend='brief',#图例设置,"brief","full"
            #col="state",#设置分面的行坐标
            #row="popdensity" # 设置分面的纵坐标
            ci=None,# 置信区间显示范围
           )

ax.set_axis_labels("Area","Population").set(xlim=(0,0.1),ylim=(0, 500000))
plt.text(x=0.045, y=520000, s='Scatterplot', fontsize=16, weight='bold')
plt.text(x=0.045, y=500050, s='Area Vs Population', fontsize=10, alpha=0.75)
plt.show()

抖动图

多个点具有完全相同的X和Y值, 为避免多个点相互绘制并隐藏，可稍微抖动点，以便直观地看到它们。

• R

ggplot(mpg, aes(cty, hwy))+ 
  geom_jitter(width = .5, size=1) +
  labs(subtitle="mpg: city vs highway mileage", 
       y="hwy", 
       x="cty", 
       title="Jittered Points")

• python

mpg = pd.read_csv("./datasets/mpg_ggplot2.csv")

fig, ax = plt.subplots(figsize=(10, 6), dpi=80)
sns.stripplot(x=mpg.cty,
              y=mpg.hwy,
              jitter=0.25,
              size=8,
              ax=ax,
              linewidth=.5,
              palette='Set1')

sns.set(style="whitegrid", font_scale=1.1)
plt.title('Use jittered plots to avoid overlapping of points')
plt.show()

计数图

区别于抖动图，多个点相互绘制并隐藏时，可使用点的大小区分重叠的程度，点的大小越大，周围的点的集中度就越大，重叠的越多。

• R

ggplot(mpg, aes(cty, hwy))+ 
 geom_count(col="tomato3", show.legend=F) +
  labs(subtitle="mpg: city vs highway mileage", 
       y="hwy", 
       x="cty", 
       title="Counts Plot")

• python

PS:也不知道为什么，把size改成sizes就可以运行了，否则就要报错：s must be a scalar, or float array-like with the same size as x and y

mpg = pd.read_csv("/Volumes/Flower0501/研究学习目录/图形及其代码/data/mpg_ggplot.csv")
df_counts = mpg.groupby(['hwy', 'cty']).size().reset_index(name='counts')
# Draw Stripplot
fig, ax = plt.subplots(figsize=(10, 6), dpi=80)
sns.stripplot(x=df_counts.cty,
              y=df_counts.hwy,
              sizes=df_counts.counts * 20,
              ax=ax,
              palette='Set1')

# Decorations
sns.set(style="whitegrid", font_scale=1.1)
plt.title('Counts Plot - Size of circle is bigger as more points overlap')
plt.show()

边际直方图/箱线图

如果您想在同一个图中显示变量的关系和分布，可以使用边际直方图。

边际直方图

• R

library(ggExtra)


mpg_select <- mpg[mpg$hwy >= 35 & mpg$cty > 27, ]
g <- ggplot(mpg, aes(cty, hwy)) + 
  geom_count() + 
  geom_smooth(method="lm", se=F)

ggMarginal(g, type = "histogram", fill="transparent")

• python-matplotlib

mpg = pd.read_csv("/Volumes/Flower0501/研究学习目录/图形及其代码/data/mpg_ggplot.csv")


fig = plt.figure(figsize=(10, 6), dpi=100)
grid = plt.GridSpec(4, 4, hspace=0.5, wspace=0.2)


ax_main = fig.add_subplot(grid[:-1, :-1])
ax_right = fig.add_subplot(grid[:-1, -1], xticklabels=[], yticklabels=[])
ax_bottom = fig.add_subplot(grid[-1, 0:-1], xticklabels=[], yticklabels=[])

## 散点主图
ax_main.scatter('cty',
                'hwy',
                s=mpg.cty * 4,
                c=mpg.manufacturer.astype('category').cat.codes,
                alpha=.9,
                data=mpg,
                cmap="Set1",
                edgecolors='gray',
                linewidths=.5)

## 底部直方图
ax_bottom.hist(mpg.cty,
               40,
               histtype='stepfilled',
               orientation='vertical',
               color='#098154')
ax_bottom.invert_yaxis()

## 左侧直方图
ax_right.hist(mpg.hwy,
              40,
              histtype='stepfilled',
              orientation='horizontal',
              color='#098154')

# 修饰图形
ax_main.set(title='Scatterplot with Histograms \n cty vs hwy',
            xlabel='cty',
            ylabel='hwy')
ax_main.title.set_fontsize(10)
for item in ([ax_main.xaxis.label, ax_main.yaxis.label] +
             ax_main.get_xticklabels() + ax_main.get_yticklabels()):
    item.set_fontsize(10)

xlabels = ax_main.get_xticks().tolist()
ax_main.set_xticklabels(xlabels)
plt.show()

• python-seaborn

mpg = pd.read_csv("/Volumes/Flower0501/研究学习目录/图形及其代码/data/mpg_ggplot.csv")

sns.set(style="whitegrid", font_scale=1.5)  #设置主题，文本大小
g = sns.jointplot(
    x='cty',
    y='hwy',
    data=mpg,  #输入两个绘图变量
    color='#098154',  #修改颜色
)
g.fig.set_size_inches(10, 8)  #设置图尺寸

边际箱线图

• R

g <- ggplot(mpg, aes(cty, hwy)) + 
  geom_count() + 
  geom_smooth(method="lm", se=F)+
  labs(title = "Scatterplot with boxplot",subtitle = "cty vs hwy",x="cty",y="hwy")

ggMarginal(g, type = "boxplot", fill="transparent")

• python

mpg = pd.read_csv("/Volumes/Flower0501/研究学习目录/图形及其代码/data/mpg_ggplot.csv")

# 创建绘图区域并设置王哥
fig = plt.figure(figsize=(10, 8), dpi=100)
grid = plt.GridSpec(
    4, 4, hspace=0.5, wspace=0.2
)  

# 定义坐标轴
ax_main = fig.add_subplot(grid[:-1, :-1])
ax_right = fig.add_subplot(grid[:-1, -1], xticklabels=[], yticklabels=[])
ax_bottom = fig.add_subplot(grid[-1, 0:-1], xticklabels=[], yticklabels=[])

# 主图散点图
ax_main.scatter(x='cty',
                y='hwy',
                s=mpg.cty * 5,
                c=mpg.manufacturer.astype('category').cat.codes,
                alpha=.9,
                data=df,
                cmap="Set1",
                edgecolors='black',
                linewidths=.5)

# 为每个轴添加轴线图
sns.boxplot(mpg.hwy, ax=ax_right, orient="v", linewidth=1, palette='Set1')
sns.boxplot(mpg.cty, ax=ax_bottom, orient="h", linewidth=1, palette='Set1')

# 修饰主图
# 移除x/y轴坐标轴
ax_bottom.set(xlabel='')
ax_right.set(ylabel='')

# 设置标题与坐标轴
ax_main.set(title='Scatterplot with Histograms \n cty vs hwy',
            xlabel='cty',
            ylabel='hwy')

# 设置主题参数
ax_main.title.set_fontsize(12)
for item in ([ax_main.xaxis.label, ax_main.yaxis.label] +
             ax_main.get_xticklabels() + ax_main.get_yticklabels()):
    item.set_fontsize(11)

plt.show()