Question

请帮助。有数据框：

ArtNo   Description     Price

AAA     Lore Ipsum      10
BBB     Lore Ipsum      9
CCC     Lore Ipsum      8
DDD     AAA Lore Ipsum  0
EEE     BBB Lore Ipsum  0
FFF     CCC Lore Ipsum  0
GGG     ZZZ Lore Ipsum  0
HHH     AAA Lore Ipsum  0

我需要在ArtNo列中的Description列中的文章中添加价格乘以而不是零的列。如果不匹配（ArtNo列与Description的第一个单词之间的空格），则将同一行中Price列中的同一图应用于“乘法列”：

ArtNo   Description     Price   Price (multiplied) ???

AAA     Lore Ipsum      10      10  
BBB     Lore Ipsum      9       9   
CCC     Lore Ipsum      8       8   
DDD     AAA Lore Ipsum  0       10  
EEE     BBB Lore Ipsum  0       9   
FFF     CCC Lore Ipsum  0       8
GGG     ZZZ Lore Ipsum  0       0
HHH     AAA Lore Ipsum  0       10

在excel中，它的工作方式如下：

IF (Price != 0; Price multiplied = Price;
    IF(ISERROR(VLOOKUP(MID(Description;1;FIND(
        ' ';Description;1));TABLE;3;0));Price multiplied = Price;
    ESLE: Price multiplied = VLOOKUP(MID(Description;1;FIND(
        ' ';Description;1));TABLE;3;0)
    )
)

谢谢。

Answer 1

您可以构建序列映射并将其应用于Description的第一个单词。

zeros = df['Price'].eq(0)
art_price_map = df[~zeros].set_index('ArtNo')['Price']
first_word = df['Description'].str.split(n=1).str[0]

df['Price (multiplied)'] = df['Price'].mask(zeros, first_word.map(art_price_map))\
                                      .fillna(0).astype(int)

print(df)

  ArtNo     Description  Price  Price (multiplied)
0   AAA      Lore Ipsum     10                  10
1   BBB      Lore Ipsum      9                   9
2   CCC      Lore Ipsum      8                   8
3   DDD  AAA Lore Ipsum      0                  10
4   EEE  BBB Lore Ipsum      0                   9
5   FFF  CCC Lore Ipsum      0                   8
6   GGG  ZZZ Lore Ipsum      0                   0

Answer 2

您可以使用pd.merge来做到这一点，

 #create new dataframe with ArtNo created from part of the Description
 df2 = df.copy()[['Description']]
 df2.columns = ['ArtNo']
 df2['ArtNo'] = df2['ArtNo'].str.split(n=1).str[0]

 #merge price from the first dataframe
 df2 = pd.merge(df2, df[['ArtNo', 'Price']], how='left', on='ArtNo')

 #create a new column 'Price (multiplied)' and fill NANs from original 'Price' column
 df['Price (multiplied)'] = df2['Price'].values
 df['Price (multiplied)'] = df['Price (multiplied)'].fillna(df['Price']).astype(int)

Python 3中的条件vlookup

2 个答案: