根据列值

Question

我如何提取列与从excel文件创建的Dataframe中的特定值匹配的行？

以下是Dataframe中的几行：

    Food            Men     Women
0   Total fruit     86.20   88.26
1   Apples, Total   89.01   89.66
2   Apples as fruit 89.18   90.42
3   Apple juice     88.78   88.42
4   Bananas         95.42   94.18
5   Berries         84.21   81.73
6   Grapes          88.79   88.13

这是我用来读取excel文件的代码，选择我需要的列并适当地重命名它们：

data1= pd.read_excel('USFoodCommodity.xls', sheetname='94-98 FAH', skiprows=76,skip_footer=142, parse_cols='A, H, K')
data1.columns = ['Food', 'Men', 'Women']

# Try 1: data1 = data1[data1['Food'].isin(['Total fruit']) == True] works
# Try 2: data1 = data1[data1['Food'].isin(['Apple, Total']) == True] doesn't work
# Try 3: data1 = data1.iloc[[1]] returns Apples, Total but not appropriate to use integer index
# Try 4: data1[data1['Food'] == 'Berries'] doesn't work

到目前为止，基于this，this或here等答案，我只能返回Food =“Total fruit”的第一个索引。当我尝试上面的其他方法时，我只获得列名，如：

Food    Men Women

我是大熊猫的新手，无法看到我哪里出错了。为什么我可以提取食物= =总水果而不是其他任何东西的第一行？

Answer 1

对我来说它很好用，也许有些空格问题 - 用strip删除它们：

print (data1.Food.tolist())
['Total fruit', 'Apples, Total ', 'Apples as fruit', 
'Apple juice', 'Bananas', ' Berries', 'Grapes']

data1['Food'] = data1['Food'].str.strip()

print (data1.Food.tolist())
['Total fruit', 'Apples, Total', 'Apples as fruit', 
'Apple juice', 'Bananas', 'Berries', 'Grapes']

data2 = data1[data1['Food'].isin(['Total fruit'])]
print (data2)
          Food   Men  Women
0  Total fruit  86.2  88.26

data3 = data1[data1['Food'].isin(['Apples, Total'])]
print (data3)
            Food    Men  Women
1  Apples, Total  89.01  89.66

data3 = data1[data1['Food'].isin(['Berries'])]
print (data3)
      Food    Men  Women
5  Berries  84.21  81.73

Answer 2

使用此代码

data1= pd.read_excel('USFoodCommodity.xls', sheetname='94-98 FAH', skiprows=76,skip_footer=142, parse_cols='A, H, K')
list_of_strings_to_match = ['Total fruit', 'Berries', 'Grape']
for index, row in data1.iterrows():
   if row['Food'] in list_of_strings_to_match:
      print row

Answer 3

这个问题可能很古老，但这是一种更简单直观的方法。

注意：此解决方案仅适用于pandas >= 0.13。

您现在可以使用.query()方法从数据框中选择列。

这很简单：

df.query('column == value') # The comparison operator can be anything.

例如，在您的情况下，您可以这样查询：

data1.query('Food == "Total Fruit"')

或

data1.query('Food == Berries')

要访问变量，请使用@。

fruit = "berries"
data1.query('Food == @fruit')

您甚至可以使用&来满足多种条件。

data1.query('condition1 == value1 & condition2 == value2')

希望有帮助。