函数 | 含义 |
---|---|
d[a] 或d.b | 选中某一列(只能是列,为啥我也不知道),列名为a(a为数字或者字符串);b必须是字符串,不能为数字 |
d[a:b] | 选取a->b的数据,区间的开闭视情况而定(只能选取整行整列) |
d.loc[] | 可以选取整列、整行,或者某一个区域(select by label) |
d.iloc[] | 同上,但通过下标来选择(select by label) |
d[条件表达式] | 可用来筛选符合条件的 |
d[a]
import pandas as pd
import numpy as np
d = pd.DataFrame(np.linspace(2, 17, 16).reshape(4, 4), ['a', 'b', 'c', 'd'], ['A', 2, 3, 4])
print(type(d)) # <class 'pandas.core.indexes.datetimes.DatetimeIndex'>
print(d)
print()
print(d[2])
print(d['A'])
print(d.A)
d[a:b]
import pandas as pd
import numpy as np
d = pd.DataFrame(np.linspace(2, 17, 16).reshape(4, 4), ['a', 'b', 'c', 'd'], ['A', 2, 3, 4])
print(type(d)) # <class 'pandas.core.indexes.datetimes.DatetimeIndex'>
print(d)
print()
print(d[0:3]) # 左闭右开
print(d['a': 'c']) # 闭区间
d.loc[]
(select by label)
import pandas as pd
import numpy as np
d = pd.DataFrame(np.linspace(2, 17, 16).reshape(4, 4), ['a', 'b', 'c', 'd'], ['A', 2, 3, 4])
print(type(d)) # <class 'pandas.core.indexes.datetimes.DatetimeIndex'>
print(d)
print()
print(d.loc['a']) # 选择a这一行
print(d.loc[:, 2]) # 选择2这一列
print(d.loc[['a','c'],[2,'A']]) # 选择某一区域
d.iloc[]
import pandas as pd
import numpy as np
d = pd.DataFrame(np.linspace(2, 17, 16).reshape(4, 4), ['a', 'b', 'c', 'd'], ['A', 2, 3, 4])
print(type(d)) # <class 'pandas.core.indexes.datetimes.DatetimeIndex'>
print(d)
print()
print(d.iloc[1]) # 取整行
print(d.iloc[:, 2]) # 取整列
print(d.iloc[1:3,2:4]) # 取一部分
d[条件表达式]
import pandas as pd
import numpy as np
d = pd.DataFrame(np.linspace(2, 17, 16).reshape(4, 4), ['a', 'b', 'c', 'd'], ['A', 2, 3, 4])
print(type(d)) # <class 'pandas.core.indexes.datetimes.DatetimeIndex'>
print(d)
print()
print(d[d.A > 5]) # 以列A中的 >5 为条件,筛选整个matrix