根据条件替换“熊猫DF”列中的值

时间:2019-06-19 21:32:36

标签: python-3.x jupyter-notebook

我是Python的新手,所以我为此很抱歉的代码提前道歉。我正在尝试完成一个Web抓取项目,当前有一个带有价格列(当前为字符串)的数据框。我的困境是我想遍历每一行,如果价格显示为每周(包含pw),那么我想将价格更新为每月;也就是说,将其乘以4。对于已经是每月价格的行,我什么也不想做。

mydf = pd.DataFrame({"prices":["350pw", "1000pm", "600pw", "1000pm", "1000pm"], "Column2":["H", "E", "L", "P", "!"]})

它将生成:

    prices Column2
0    350pw       H
1   1000pm       E
2    600pw       L
3   1000pm       P
4   1000pm       !

我能够找到行并仅提取数字。从那里,我转换为int并乘以4,但无法将replace函数与int一起使用。

for x in mydf[mydf['prices'].str.contains('pw')]['prices']:
    weekly_price = int(x[0:3])
    monthly_price_int = weekly_price * 4

不确定从这里要去哪里。...

最终结果将是:

    prices Column2
0   1400pw       H
1   1000pm       E
2   2400pw       L
3   1000pm       P
4   1000pm       !

2 个答案:

答案 0 :(得分:0)

import pandas as pd
def preprocess(x):
    if(x['position']>=0):
        x['prices']=str(int(x['prices'][:x['position']])*4)+"pm"
        return x
    return x

mydf = pd.DataFrame({"prices":["350pw", "1000pm", "600pw", "1000pm", "1000pm"], "Column2":["H", "E", "L", "P", "!"]})
mydf["position"]=mydf.prices.str.find('pw')
mydf=mydf.apply(preprocess, axis=1)
mydf.drop(['position'],axis=1,inplace=True)

print(mydf)

enter image description here

答案 1 :(得分:0)

这更多是一个熊猫问题,但这是您可能应该执行的操作:

import pandas as pd

mydf = [your df above]

#define a function to convert from weekly to monthly
def make_monthly(cell):
    if 'pw' in cell:
        weekly_price = int(cell[0:3])
        monthly_price_int = weekly_price * 4
        new_cell = str(monthly_price_int)+'pm' #you need to update the period designation as well
        return new_cell
    else:
        return cell

最后,在必要时修改“价格”行中的值:

mydf['prices'] = mydf['prices'].map(make_monthly)

输出:

   prices   Column2
0   1400pm  H
1   1000pm  E
2   2400pm  L
3   1000pm  P
4   1000pm  !