info方法查看数据类型

info()方法可以用来查看数据的类型以及不为null的行数

import pandas as pd

df = pd.read_excel(r"C:\Users\matrix\Desktop\jd-ads-discovery_2022-09-08_2022-09-08.xls")
df.info()

dtypes查看某列的数据类型

import pandas as pd

df = pd.read_excel(r"C:\Users\matrix\Desktop\jd-ads-discovery_2022-09-08_2022-09-08.xls")
for column in df.columns:
    print(df[column].dtypes)

输出符合指定类型的列

import pandas as pd

df = pd.read_excel(r"C:\Users\matrix\Desktop\lazada-ads-discovery_2022-09-08_2022-09-08.xls")
for column in df.columns:
    if df[column].dtypes == 'int64':
        print(column)

查看空值的列

import pandas as pd

df = pd.read_excel(r"C:\Users\matrix\Desktop\lazada-ads-discovery_2022-09-08_2022-09-08.xls")
# 输出包含空值的列
df.loc[:,(df.isnull()).any()].columns
# 输出全为空值的列
df.loc[:,(df.isnull()).all()].columns

求某列的sum和

df[column].sum()
【注】包含非数字字符的列会按照字符串拼接处理

import pandas as pd

df = pd.read_excel(r"C:\Users\matrix\Desktop\lazada-ads-discovery_2022-09-08_2022-09-08.xls")
for column in df.columns:
    dtype = df[column].dtype
    if dtype == 'int64' or dtype == 'float64':
        print(df[column].sum())

python正则匹配

import re

str= '0.00%'
## 0.00%
print(re.search(r'0+\.*0*%$', str).group())

str= '0.00%'
## 0.00%
print(re.fullmatch(r'(0+\.*0*%$)', str) != None)

某列匹配正则

# array([ True,  True,  True, False])
df['Direct CVR'].apply(lambda s: re.fullmatch(r'0+\.*0*%$', s) != None).values

判断是否全为空或者全为0

import pandas as pd
import re
from pandas import Series,DataFrame
import numpy as np

df = pd.read_excel(r"C:\Users\Sun Tian Sheng\Desktop\lazada-ads-discovery_2022-09-08_2022-09-081.xls")
flag = False
for column in df.columns:
    if dtype == 'int64' or dtype == 'float64':
        sum = df[column].sum()
        if sum != 0:
            flag = True
    else:
        if df[column].isnull().all() == False:
            flag = True
        
        f = lambda s: re.fullmatch(r'0+\.*0*%$', s) != None
        blank_value_rows = np.sum(df[column].apply(f).values == True)
        column_rows = len(df[column])
        if blank_value_rows != column_rows:
            flag = True
print(flag)

参考

【1】python 利用all(),any() 查找存在0、全是0、含缺失值的列
【2】python pandas对多列求和
【3】pandas中查看数据类型的几种方式 （含python、pandas、numpy类型对照表）
【4】Python正则表达式，这一篇就够了！
【5】pandas（12）：数据清洗（缺失值）
【6】pandas.DataFrame.apply