机器学习数据科学包（十五）——Matplotlib模块：实例

实例1：画函数积分图

步骤

• 画出类似的图像
• 让x坐标和y坐标刻度消失
• x坐标上标有a，b
• 坐标轴外有x，y注释
• 阴影部分是一个不规则的多边形
• 将公式显示

目标

1.画出类似的图像

2.让x坐标和y坐标刻度消失

ax.set_xticks([a,b])
ax.set_yticks([])

3.x坐标上标有a，b

ax.set_xticklabels([str1,str2])

4.坐标轴外有x，y注释

plt.figtext(x,y,str)  # 注:x,y范围是(0,1)

5.阴影部分是一个不规则的多边形

画多边形

6.将公式显示

完整代码

# 1.画基本图形
def fun(x):
    return -(x-2)*(x-8)+40
x = np.linspace(0,10)
y =fun(x)
fig,ax = plt.subplots()
ax.plot(x,y,color='r',linewidth=2)
ax.axis([0,10,0,50])

# 2.坐标消失
a,b =2,8
ax.set_xticks([a,b])   
ax.set_yticks([])

# 3.指定a，b标签
ax.set_xticklabels(['$a$','$b$'])

# 4.注释坐标名称
plt.figtext(0.9,0.05,'$x$')    
plt.figtext(0.1,0.9,'$y$') 
  
# 5.画阴影部分
ix = np.linspace(a,b)        
iy = fun(ix)                
ixy = zip(ix,iy)  
verts = [(a,0),*ixy,(b,0)] 
poly =plt.Polygon(verts, facecolor='0.9', edgecolor='0.5')  
ax.add_patch(poly)    

# 6.显示公式
ax.text((a+b)/2,20,r'$\int_a^b(-(x-2)*(x-8)+40)dx$',horizontalalignment='center',fontsize=13)

 

 

实例2：将散点图xy坐标以直方图显示

步骤

• 分析图组成
• 生成画布
• 画散点图
• 画直方图

目标

分析图组成

生成画布

画散点图

画直方图

完整代码

# 生成画布
plt.style.use('ggplot')
margin_border = 0.1
width = 0.6
height =0.2
margin_between = 0.02
rect_s=[margin_border, margin_border, width, width]
rect_x=[margin_border, margin_border+width+margin_between, width, height]
rect_y=[margin_border+width+margin_between, margin_border, height, width]
fig = plt.figure(1,figsize=(8,8))    
axScatter = plt.axes(rect_s)   
histX = plt.axes(rect_x)        
histY = plt.axes(rect_y)        
histX.set_xticks([])            
histY.set_yticks([])

# 画散点图
x = np.random.randn(200)
y = x + np.random.randn(200) * 0.5
axScatter.scatter(x,y)

# 固定坐标轴，让显示数据对应上
bin_width = 0.25     
xymax = np.max([np.max(np.fabs(x)),np.max(np.fabs(y))]) 
lim = int(xymax/bin_width+1)*bin_width   
axScatter.set_xlim(-lim, lim)   
axScatter.set_ylim(-lim, lim)   
histX.set_xlim(-lim, lim)      
histY.set_ylim(-lim, lim)      

# 画直方图
bins = np.arange(-lim,lim+bin_width,bin_width)  
histX.hist(x,bins=bins)
histY.hist(y,bins=bins,orientation='horizontal')

title = fig.text(0.5,0.95,'Scatter and Hist')

实例三：球员能力图

• 中文字体
• plt.rcParams['font.sans-serif']=['SimHei']
  plt.rcParams['axes.unicode_minus']=False

# 识别中文字体
plt.rcParams['font.sans-serif']=['SimHei']
plt.rcParams['axes.unicode_minus']=False

# 画图参数
plt.style.use('ggplot')
ability_size = 6
ability = ['进攻','防守','盘带','速度','体力','射速']
fig = plt.figure(figsize=(8,8))
ax1 = plt.subplot(221, projection = 'polar')
ax2 = plt.subplot(222, projection = 'polar')
ax3 = plt.subplot(223, projection = 'polar')
ax4 = plt.subplot(224, projection = 'polar')
player = {
    'M':np.random.randint(60,99,size = ability_size),
    'H':np.random.randint(60,99,ability_size),
    'P':np.random.randint(60,99,ability_size),
    'Q':np.random.randint(60,99,ability_size)
}
theta = np.linspace(0,2*np.pi,6,endpoint=False)
theta = np.append(theta, theta[0])
player['M'] = np.append(player['M'], player['M'][0])
player['H'] = np.append(player['H'], player['H'][0])
player['P'] = np.append(player['P'], player['P'][0])
player['Q'] = np.append(player['Q'], player['Q'][0])

# 画球员能力图
ax1.plot(theta, player['M'], 'r')
ax1.fill(theta, player['M'], color='r',alpha=0.5)
ax1.set_xticks(theta)
ax1.set_yticks([20,40,60,80,100])
ax1.set_xticklabels(ability)
ax1.set_title('梅西',size=20,color='r')

ax2.plot(theta, player['H'], 'g')
ax2.fill(theta, player['H'], color='g',alpha=0.5)
ax2.set_xticks(theta)
ax2.set_yticks([20,40,60,80,100])
ax2.set_xticklabels(ability)
ax2.set_title('哈维',size=20,color='g')

ax3.plot(theta, player['P'], 'b')
ax3.fill(theta, player['P'], color='b',alpha=0.5)
ax3.set_xticks(theta)
ax3.set_yticks([20,40,60,80,100])
ax3.set_xticklabels(ability)
ax3.set_title('皮克',position=(0.5,1), size=20,color='b')

ax4.plot(theta, player['Q'], 'y')
ax4.fill(theta, player['Q'], color='y',alpha=0.5)
ax4.set_xticks(theta)
ax4.set_yticks([20,40,60,80,100])
ax4.set_xticklabels(ability)
ax4.set_title('切赫',position=(0.5,1), size=20,color='y')

 

实例四：股票k型图

import pandas_datareader as pdr
import mpl_finance as mpf
plt.style.use('ggplot')
quotes = pdr.get_data_yahoo('INTC', '2019/12/1', '2020/1/1')
quotes
volumns = np.array([])
dates = np.array([])
for record in quotes:
    dates = np.append(dates, record[0])
    volumns = np.append(volumns, record[5])

left, width = 0.1, 0.8
rect_vol = [left,0.1,width,0.3]
rect_main = [left,0.45,width,0.5]
fig = plt.figure(figsize=(8,8))

ax_vol = fig.add_axes(rect_vol)
ax_vol.fill_between(dates, volumns, color='y')
ax_vol.xaxis_date()
plt.setp(ax_vol.get_xticklables(),rotation=30,horizontalalignment='right')

ax_mian = fig.add_axes(rect_main)
mpf.candlestick_ochl(ax_main,quotes,width=0.5,colorup='r',colordown='g')
ax_main.axes.get_xaxis().set_visible(False)

plt.title('Stock INTC Price and Volumn')

课程中使用的matplotlib.finance模块现在已经取消，解决方式见Acegem博客，

《如何解决使用matplotlib.finance获取雅虎财经网站股票数据报错？》