02_pandas

02_pandas

分列:split

                        start  ...        date
0  2021-12-13 08:37:29.971975  ...  2021-12-13
1  2021-12-13 08:39:12.057289  ...  2021-12-13
2  2021-12-13 08:41:33.620852  ...  2021-12-13
3  2021-12-13 08:43:59.893395  ...  2021-12-13
4  2021-12-13 08:46:08.058491  ...  2021-12-13
5  2021-12-13 08:48:16.099560  ...  2021-12-13

a = data['start'].str.split(' ', expand=True)
data['start_date'] = a[0]
a[1] = a[1].str[:4]
data['start_hour'] = a[1]

删除列

# data.drop('col_name', 1, inplace=True)
data.drop('start', 1, inplace=True)

数据类型转换

dataFrame = dataFrame.astype(numpy.str)

groupby

data = {"fruit": ["apple", "pear"], "price": [3, 4]}
dataFrame = pd.DataFrame(data=data)

dataFrame.groupby("fruit")['price'].sum()
"""
fruit
apple    3
pear     4
Name: price, dtype: int64
"""
dataFrame.groupby("fruit").sum()
"""
       price
fruit       
apple      3
pear       4
"""

for group in dataFrame.groupby("fruit"):
	groupName, groupData = group