2.绘图&np&pd

Matplotlib

1.基础操作

1.绘制基本图像

import matplotlib.pyplot as plt
import numpy as np

x = np.linspace(-3,3,50)
y1 = 2*x+1
y2 = x**2

plt.figure()
plt.plot(x,y1)

plt.figure(num=3,figsize=(8,5))
plt.plot(x,y2)
plt.plot(x,y1,color='red',linewidth=1.0,linestyle='--' ) # 设置y1曲线的颜色、宽度、形状 
plt.show()

2.设置坐标轴范围、标签及原点位置

plt.xlim((-1,2))    # 设置x轴坐标范围
plt.xlim((-2,3))    # 设置y轴坐标范围
plt.xlabel('xxx')   # 设置x轴标题
plt.ylabel('yyy')   # 设置y轴标题

new_ticks = np.linspace(-1,2,5)
plt.xticks(new_ticks)   # 重新设置x轴刻度
plt.yticks([-3,-2,-1,-1.5,3],
           ['really bad',r'$bad$',r'$normal\ \alpha$',r'$good$','$really\ good$'])
                        # 设置y轴刻度为文字，正则表达式设置字体r'$xx$'
                        # \alpha   LaTex语法

# 设置坐标原点
ax = plt.gca()  # 得到边框
ax.spines['right'].set_color('none')    # 取消右边边框
ax.spines['top'].set_color('none')  # 取消顶部边框
ax.xaxis.set_ticks_position('bottom')   # 设置默认x轴
ax.yaxis.set_ticks_position('left')     # 设置默认y轴
ax.spines['bottom'].set_position(('data',0))
ax.spines['left'].set_position(('data',0))  

3.设置图例

l1, = plt.plot(x,y2,label='y2')
l2, = plt.plot(x,y1,color='red',linewidth=1.0,linestyle='--',label='y1')

plt.legend(handles=[l1,l2,],labels=['label1','label2'],loc='best')

4.标注

# 添加文本，设置添加位置、形式
# 方法1
plt.annotate(r'$2x+1=%s$' % y0,xy=(x0,y0),xycoords='data',
             xytext=(+30,-30),textcoords='offset points',
             fontsize=16,arrowprops=dict(arrowstyle='->',connectionstyle='arc3,rad=0.2'))

# 方法2
plt.text(0.5,-3,r'$This\ is\ the\ text.\mu\ \sigma_i\ \alpha_t$',
         fontdict={'size':16,'color':'r'})

5.散点图

n = 1024
X = np.random.normal(0,1,n)
Y = np.random.normal(0,1,n)
T = np.arctan2(Y,X)

plt.scatter(X,Y,s=75,c=T,alpha=0.5)

plt.xlim((-1.5,1.5))
plt.ylim((-1.5,1.5))

plt.show()

6.绘制子图

plt.figure()
plt.subplot(2,2,1)  # 2行2列，开始绘制第1个图
plt.plot([0,1],[0,1])
plt.subplot(2,2,2)  # 2行2列，开始绘制第1个图
plt.plot([0,1],[0,1])
plt.subplot(2,2,3)  # 2行2列，开始绘制第1个图
plt.plot([0,1],[0,1])
plt.subplot(2,2,4)  # 2行2列，开始绘制第1个图
plt.plot([0,1],[0,1])
plt.show()

分格显示

# 方法1:plt.subplot2grid()
ax1 = plt.subplot2grid((3,3),(0,0),colspan=3,rowspan=1)
ax1.plot([1,2],[1,2])
ax1.set_title('ax1_title')
ax2 = plt.subplot2grid((3,3),(1,0),colspan=2)
ax3 = plt.subplot2grid((3,3),(1,2),rowspan=2)
ax4 = plt.subplot2grid((3,3),(2,0))
ax5 = plt.subplot2grid((3,3),(2,1))

import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec
# 方法2:gridspec
gs = gridspec.GridSpec(3,3)
ax1 = plt.subplot(gs[0,:])
ax2 = plt.subplot(gs[1,:2])
ax3 = plt.subplot(gs[1:,2])
ax4 = plt.subplot(gs[-1,0])
ax5 = plt.subplot(gs[-1,-2])

# 方法3
import matplotlib.pyplot as plt
f,((ax11,ax12),(ax21,ax22)) = plt.subplots(2,2,sharex=True,sharey=True)
ax11.scatter([1,2],[1,2])
plt.tight_layout()
plt.show()

Numpy

1.基础操作

• 生成array

  import numpy as np
  # 生成array
  a1 = np.array([2,3,4])
  a2 = np.array([[2,3,4],
                 [5,6,7]])		# 2*3 矩阵
  a3 = np.zeros((3,4))			# 3*4 全0矩阵
  a4 = np.ones((3,4))				# 3*4 全1矩阵
  a5 = np.arange(1,10,2)			# 类似range()
  a6 = np.arange(12).reshape((3,4))# 改变形状
  

• 运算

  import numpy as np
  a = np.array([1,2,3,4])
  b = np.arange(4)		# a与b进行+/-/*，都是逐个对应运算
  # np.sin(),np.cos(),...
  
  c = np.array([[2,3],
                 [5,6]])
  d = np.arange(4).reshape((2,2))
  result1 = c*d			# 对应元素相乘
  result2 = np.dot(c,d)	# 矩阵乘法
  result3 = c.dot(d)		# 矩阵乘法
  

• 索引

  np.sum(),	np.min(),	np.max()
  np.transpose(c)	# 转置，c.T
  c[1,1],c[1][1],c[1,1:3]	# 索引
  c.flatten()	# 转为1行
  np.argmin(c)	# 最小值索引
  np.sort()	# 排序	
  np.clip()
  np.mean(c,axis=0) # 计算平均值,axis=0表示列，1表示行
  

• 矩阵合并

  A = np.array[1,1,1]
  B = np.array[2,2,2]
  C = np.vstack((A,B))	# vertical stack，C：2*3
  D = np.hstack((A,B))	# horizontal stack，D：1*6
  

• 矩阵分割

  A = np.arange(12).reshape((3,4))
  np.split(A,2,axia=1)
  

• deep copy

  b = a.copy()	# 对b操作不会影响a
  

Pandas

1.基础操作

• 选择数据

  import pandas as pd
  import numpy as np
  s = pd.Series([1,2,3])
  a1 = pd.array((1,2,3,4,5,6))
  data1 = pd.DataFrame(np.arange(12).reshape((3,4)))
  
  range1 = pd.date_range('20210101',periods=6)
  data2 = pd.DataFrame(np.arange(24).reshape((6,4)),
                       index=range1,columns=['A','B','C','D'])
  print(data2['A'],data2.A)
  print(data2[0:3],data['20210101':'20210103'])
  # 用标签及索引筛选
  print(data2.loc['20210101':'20210103'])
  print(data2.loc[:,['A','B']])
  print(data2.loc['20210102',['A','B']])
  
  
  # 用索引筛选
  print(data2.iloc[3,1])
  print(data2.iloc[3:5,1:3])	 # print(data2.iloc[[1,3,4],1:3])
  
  print(data2[data2.A>8])
  
  

• 设置值

  data2.iloc[2,2] = 111
  data2.loc['20210102','B'] = 222
  data2[data2.B>17] = 1
  data2.C[data2.A>4] = 0
  data2['F'] = np.nan
  data2['E'] = pd.Series([1,2,3,4,5,6],
                         index=pd.date_range('20210101',periods=6))
  print(data2)
  

• 处理NaN数据

  print(data2.dropna(axis=0,how='any'))	# 隐藏NaN所在的行
  data2.fillna(value=0)			# 将NaN设置为0
  data2.isnull()			# 判断是否有NaN，返回True/False
  

• 数据的导入与导出

  data = pd.read_csv('xxx.csv')		# 导入数据
  data.to_pickle('xxx.pickle')		# 导出数据
  

• 合并DataFrame

  df1 = pd.DataFrame(np.ones((3,4))*0,columns=['a','b','c','d'])
  df2 = pd.DataFrame(np.ones((3,4))*1,columns=['a','b','c','d'])
  df3 = pd.DataFrame(np.ones((3,4))*2,columns=['a','b','c','d'])
  # columns相同的数据，上下合并
  res1 = pd.concat([df1,df2,df3],axis=0)	# 0竖向，1横向
  res2 = pd.concat([df1,df2,df3],axis=0,ignore_index=True)
  
  df4 = pd.DataFrame(np.ones((3,4))*0,columns=['a','b','c','d'])
  df5 = pd.DataFrame(np.ones((3,4))*1,columns=['b','c','d','e'])
  res3 = pd.concat([df4,df5],join='inner',ignore_index=True)
  res4 = df4.append(df5,ignore_index=True)
  

  # merge
  df1 = pd.DataFrame({'key':['k0','k1','k2','k3'],
                      'A':['a0','a1','a2','a3'],
                      'B':['b0','b1','b2','b3']})
  df2 = pd.DataFrame({'key':['k0','k1','k2','k3'],
                      'C':['c0','c1','c2','c3'],
                      'D':['d0','d1','d2','d3']})
  res1 = pd.merge(df1,df2,on='key')
  print(res1)
  res2 = pd.merge(df1,df2,on=['key1','key2']) # 默认how='inner'
  

• 绘图