matplotlib的使用

基本知识

1. matplotlib中的图都是画在Figure上的，而每个图像都是一个子图Axes，在一个Figure上可以有多个子图Axes

创建子图python

# 导入模块
import matplotlib.pyplot as plt

# subplots返回一个figure对象和一个Axes对象数组
# 单独返回一个Axes对象
fig, ax = plt.subplots()

# 返回多个Axes对象
fig, axs = plt.subplots(2, 2)

# 使用元组的拆解给每一个ax都赋值一个Axes对象
fig, (ax1, ax2) = plt.subplots(1, 2)
fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2, 2)

绘制基本图形

直方图

1. 普通直方图

# 画直方图
import matplotlib.pyplot as plt
from matplotlib import rcParams
import random

# 让中文正确显示
rcParams['font.family'] = rcParams['font.sans-serif'] = 'SimHei'

fig, ax = plt.subplots()
# cordx 是直方图中直方柱的中心点
cordx = range(10)
# cordy 是直方图的直方柱的高度，这里之所以转成list，是因为我想用random.shuffle，而random.shuffle只能对可变序列进行原地修改，返回的是None
cordy = list(range(0, 10))
random.shuffle(cordy)
ax.bar(x=cordx, height=cordy, color="blue", width=1,edgecolor="white")
ax.set_title('直方图')
plt.show()

2. 堆叠直方图

# 画堆叠直方图的关键在于堆在上面的直方的bottom应该是下面的直方的高度
import matplotlib.pyplot as plt
from matplotlib import rcParams
import random

# 让中文正确显示
rcParams['font.family'] = rcParams['font.sans-serif'] = 'SimHei'

fig, ax = plt.subplots()
month = range(1, 13)
profit1 = list(range(10, 130, 10))
random.shuffle(profit1)
profit2 = list(range(10, 130, 10))
random.shuffle(profit2)

# 画图
ax.bar(x=month, height=profit1, color="red", label='Mihoyo', width=0.5)
# 关键就在于这个bottom的设置
ax.bar(x=month, height=profit2, color='green', label='Timi', width=0.5, bottom=profit1)

# 设置刻度
xlabel = ['Jan', 'Feb','Mar','Apr','May','Jun','Jul', 'Aug', 'Sep', 'Oct','Nov', 'Dec']
ax.set_xticks(month)
ax.set_xticklabels(xlabel)
ax.set_ylabel('Profit')
ax.set_xlabel('2020年')

# 设置说明图，根据画图时的label属性，会自动找空白处放置说明图
ax.legend()

# 设置标题
ax.set_title('堆叠直方图')

plt.show()

3. 画多个直方图

# 画多个直方图
import matplotlib.pyplot as plt
from matplotlib import rcParams
import numpy  as np
import random

# 让中文正常显示
rcParams['font.family'] = rcParams['font.sans-serif'] = 'SimHei'

# 设置窗口大小横10, 纵5
fig, ax = plt.subplots(figsize=(10, 5))

# 设置横纵坐标
profit ={
    'jindong': [random.randint(10, 100) for i in range(12)],
    'taobao': [random.randint(10, 100) for i in range(12)],
    'tianmao': [random.randint(10,100) for i in range(12)]
}
xlabel = ['Jan', 'Feb', 'Mar','Apr','May','Jun','Jul','Aug','Sep','Oct','Nov','Dec']
# 每一个x轴标签的位置，也是直方柱的中心
cordx = np.arange(1, 13)
# 为了让cordx就是直方柱的中心，需要将起始点向左偏移刻度间距的一半
start = cordx - 0.5
width = 1 / (len(profit.keys()) + 1)
multiplier = 1

for company, cprofit in profit.items():
    offset = width * multiplier
    ax.bar(x = start+offset, height=cprofit, label=company, width=width)
    multiplier += 1
ax.set_xticks(cordx)
ax.set_xticklabels(xlabel)
ax.set_ylabel('Profit')
ax.set_xlabel('2020年')
ax.set_title('2020年各电商收入情况')
ax.legend(title='company')
plt.show()

散点图

import matplotlib.pyplot as plt
import numpy as np

fig, ax = plt.subplots()

x = np.array([10, 20, 30, 40, 30, 50])
y = np.array([100,200,300,400,500,600])

# marker属性指定形状， c 是 color 的缩写，r 是 red 的缩写
ax.scatter(x, y, marker='x', c='r')
plt.show()

折线图

import matplotlib.pyplot as plt
import numpy as np

# 曲线其实就是利用折线图来画的，为了实现平滑的视觉效果，要用很多个密集的点
x = np.linspace(0,10,100)
y = 4 + np.sin(x)

fig, ax = plt.subplots()
ax.plot(x, y, linewidth=2)

ax.set(xlim=(0,8), xticks=np.arange(0,11),
       ylim=(0,8),yticks=np.arange(1,8))

plt.show()

饼图

import matplotlib.pyplot as plt
from matplotlib import rcParams

rcParams['font.family'] = rcParams['font.sans-serif'] = 'SimHei'

ax = plt.figure().add_subplot()


def drawPie(ax):
    sectors = [12.5 , 20, 45, 22.5]
    labels = ['A', 'B', 'C', 'D']
    expl = [0, 0, 0.1, 0]
    colors=['red','skyblue','green','orange']
    ax.pie(x=sectors, labels=labels, startangle=90,autopct='%.2f', labeldistance=1.1, pctdistance=0.6, explode=expl, shadow=True, colors=colors)
    ax.set_title('饼图')

热力图

import numpy as np
import matplotlib.pyplot as plt
from matplotlib import rcParams

rcParams['font.family'] = rcParams['font.sans-serif'] = 'SimHei'

data = np.random.randint(0, 100, 30).reshape((5,6))
xlabels = ['Beijing','Tianjing','Shanghai','Guangzhou','Chongqin','Sichuan']
ylabels = ['2016', '2017', '2018', '2019', '2020']

ax = plt.figure().add_subplot()
ax.set_xticks(range(len(xlabels)))
ax.set_xticklabels(xlabels)
ax.set_yticks(range(len(ylabels)))
ax.set_yticklabels(ylabels)
heatmap = ax.imshow(data, cmap=plt.cm.hot, aspect='equal', vmin=0, vmax=255)
for i in range(len(xlabels)):
    for j in range(len(ylabels)):
        ax.text(i, j, data[j][i], ha='center', va='center', color='blue', size=16)
plt.colorbar(heatmap)
plt.xticks(rotation=45, ha='right')
ax.set_title('热力图')
plt.show()

雷达图

import matplotlib.pyplot as plt
from matplotlib import rcParams
import numpy as np
from math import pi

rcParams['font.family'] = rcParams['font.sans-serif'] = 'SimHei'

# 绘制雷达图
def drawRadar(ax):
    labels = np.array(['EQ', 'IQ', '人缘', '魅力', '财富', '体力'])
    attrnum = len(labels)
    names = ['张三', '李四','王五']
    data = [
        [0.4, 0.32, 0.35], [0.85, 0.35, 0.3],
        [0.40,0.32,0.35], [0.40, 0.82, 0.75],
        [0.14, 0.12, 0.35],[0.80, 0.92, 0.35]
    ]
    angles = [2 * pi * i /attrnum for i in range(attrnum)]
    angles2 = [angle * 180 / pi  for angle in angles]
    ax.set_thetagrids(angles2, labels=labels,fontproperties ='SimHei')
    ax.fill(angles, data, alpha=0.25)
    ax.legend(names, loc=(0.95, 0.9))

# 这里一定要加projection='polar',这样生成的子图对象中才是以极坐标开始的，并且也只有这样，这个Axes对象中才有set_thetagrids方法
fig, ax = plt.subplots(subplot_kw=dict(projection='polar'))
# ----------------------------- #
# 或者以下面的方式书写也可以
# fig = plt.figure(figsize=(10,5))
# ax = fig.add_subplot(projection='polar')
# ----------------------------- #
drawRadar(ax)
plt.show()

绘制多个图

# 以上面的雷达图为例子
fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2,2, subplot_kw=dict(projection='polar'))

drawRadar(ax1)
drawRadar(ax2)
drawRadar(ax3)
drawRadar(ax4)
plt.show()