Python：Pandas学习

import pandas as pd
import numpy as np
s = pd.Series([1, 3, 6, np.nan, 44, 1])

df= pd.DataFrame(np.random.random((4,5)))

# data frame 常用属性
df.dtypes
df.index
df.columns
df.values

# data frame 常用方法
df.describe()
df.T
df.sort_index(axis = 1, ascending = False)
df.sort_values(by = 4)

# 选择数据
dates = pd.date_range('20160101', periods = 6)
df = pd.DataFrame(np.arange(24).reshape((6,4)), index = dates,
                  columns = ['A', 'B', 'C', 'D'])

'''row or column''' # 行不可隔着选择
print(df[0:3])
print(df[['A', 'D']])

'''select by label:loc''' # 行不可隔着选择
print(df.loc['20160101', :])
print(df.loc[:,['A', 'B']])

'''select by position:iloc'''
print(df.iloc[[0, 2], [0, 3]])

'''mixed selection:ix'''
print(df.ix[[0, 2], ['A', 'D']])

'''Boolean indexing'''
print(df[df.B > 5])

# 设置数据
df.iloc[2, 2] = 111
df.loc['20160101', 'D'] = 222
df.B[df.A > 5] = 0
print(df)

df['F'] = np.nan
df['E'] = range(6)
print(df)

# 处理缺失数据
df.iloc[0, 1] = np.nan
df.iloc[1, 2] = np.nan
print(df)
print(df.dropna(axis = 0, how = 'all')) # how = {'any', 'all'}
print(df.fillna(value = 0))
print(np.any(df.isnull()))

# data frame 合并
'''concatenating'''
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'])

res = pd.concat([df1, df2, df3], axis = 0, ignore_index = True)
res1 = pd.concat([df1, df2, df3], axis = 1)

'''join参数'''
df1 = pd.DataFrame(np.ones((3,4))*0, columns = ['a', 'b', 'c', 'd'], index = [1, 2, 3])
df2 = pd.DataFrame(np.ones((3,4))*1, columns = ['b', 'c', 'd', 'e'], index = [2, 3, 4])

res = pd.concat([df1, df2], join = 'outer', ignore_index = True)
res = pd.concat([df1, df2], join = 'inner', ignore_index = True)
print(res)

'''join_axes'''
res = pd.concat([df1, df2], axis = 1, join = 'inner')
res = pd.concat([df1, df2], axis = 1, join_axes = [df1.index])

# append
df1 = pd.DataFrame(np.ones((3,4))*0, columns = ['a', 'b', 'c', 'd'], index = [1, 2, 3])
df2 = pd.DataFrame(np.ones((3,4))*1, columns = ['b', 'c', 'd', 'e'], index = [2, 3, 4])
df3 = pd.DataFrame(np.ones((3,4))*1, columns = ['b', 'c', 'd', 'e'], index = [2, 3, 4])

res = df1.append([df2, df3], ignore_index = True)
res1 = pd.concat([df1, df2, df3])
print(res)
print(res1)

# data frame merge
'''merge one key'''
left = pd.DataFrame({'key':['K1','K2','K3'],
                     'A':[1,2,3],
                     'B':[4,5,6]})

right = pd.DataFrame({'key':['K0','K1','K3'],
                     'A':[11,43,53],
                     'D':[12,-1,0]})
res = pd.merge(left, right, on = 'key', how = 'outer')
print(res)

'''merge two or more keys'''
left = pd.DataFrame({'key0':['K1','K2','K3'],
                     'key1':['X0','X2','X3'],
                     'A':[1,2,3],
                     'B':[4,5,6]})

right = pd.DataFrame({'key0':['K0','K1','K3'],
                      'key1':['X1','X0','K3'],
                     'A':[11,43,53],
                     'D':[12,-1,0]})
res = pd.merge(left, right, on = ['key0', 'key1'], how = 'outer')
print(res)

'''merge index'''
left = pd.DataFrame({'A':[1,2,3],
                     'B':[4,5,6]},
                    index = ['K0', 'K1', 'K2'])

right = pd.DataFrame({'A':[11,43,53],
                     'D':[12,-1,0]},
                    index = ['K1', 'K2', 'K3'])
res = pd.merge(left, right, left_index = True,
               right_index = True)
print(res)

'''handle overlapping columns'''
left = pd.DataFrame({'key':['K1','K2','K3'],
                     'A':[1,2,3],
                     'B':[4,5,6]})

right = pd.DataFrame({'key':['K0','K1','K3'],
                     'A':[11,43,53],
                     'B':[12,-1,0]})
res = pd.merge(left, right, on = 'key',
               suffixes = ['_left', '_right'] , how = 'outer')
print(res)

# 作图
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

'''plot data'''
'''Series'''
data = pd.Series(np.random.randn(1000), index = np.arange(1000))
data = data.cumsum()
data.plot()
print(data)

'''Data Frame'''
data = pd.DataFrame(np.random.randn(1000, 4), 
                    index = np.arange(1000),
                    columns = list("ABCD"))
print(data.head())
data = data.cumsum()
data.plot()
ax = data.plot.scatter(x = 'A', y = 'C',
                       color = 'Red',
                       label = 'Class 2')
data.plot.scatter(x = 'A', y = 'B',
                  color = 'DarkGreen',
                  label = 'Class 2',
                  ax = ax)