熊猫数据框索引在索引数据框的子集时引起问题。 如何删除索引，或防止发生错误？

Question

I have a dataframe x1 . I made a subset of the dataframe, x1_sub , where I need to use a for loop to index its items. But because the subset retains the indexing of the original pandas dataframe, it has its rows like so:

x1_sub['words']

1         investment
2               fund
4            company
7              claim
9           customer
20              easy
...              ...

So, when I do something like this to index the rows of x1_sub serially:

for i in range(len(x1)):
    for j in range(len(x1_sub)):
        if (x1['word'][i]==x1_sub['word'][j]):
            print(i, j)

it gives the following error:

KeyError                                  Traceback (most recent call last)
<ipython-input-48-e3c9806732a6> in <module>()
      3 for i in range(len(x1)):
      4     for j in range(len(x1_sub)):
----> 5         if (x1['word'][i]==x1_sub['word'][j]):
      6             print(i, j)
      7 

c:\users\h473\appdata\local\programs\python\python35\lib\site-packages\pandas\core\series.py in __getitem__(self, key)
    621         key = com._apply_if_callable(key, self)
    622         try:
--> 623             result = self.index.get_value(self, key)
    624 
    625             if not is_scalar(result):

c:\users\h473\appdata\local\programs\python\python35\lib\site-packages\pandas\core\indexes\base.py in get_value(self, series, key)
   2558         try:
   2559             return self._engine.get_value(s, k,
-> 2560                                           tz=getattr(series.dtype, 'tz', None))
   2561         except KeyError as e1:
   2562             if len(self) > 0 and self.inferred_type in ['integer', 'boolean']:

pandas\_libs\index.pyx in pandas._libs.index.IndexEngine.get_value()

pandas\_libs\index.pyx in pandas._libs.index.IndexEngine.get_value()

pandas\_libs\index.pyx in pandas._libs.index.IndexEngine.get_loc()

pandas\_libs\hashtable_class_helper.pxi in pandas._libs.hashtable.Int64HashTable.get_item()

pandas\_libs\hashtable_class_helper.pxi in pandas._libs.hashtable.Int64HashTable.get_item()

KeyError: 0

---

EDIT: Some example data:

The following data is saved in a csv file named example.csv :

word    score
service 1
customer    4
agent   3
product 6
easy    2
claim   2
fast    1
financial   5
information 1
benefit 4
company 3
helpful 6
time    2
future  2
policy  1
health  5
life    1
fund    4
complicated 3
investment  6
join    2
payment 2
premium 1
excellent   5
experience  1
family  4
nice    3
proces  6
satisfactory    2

And the code is this:

import pandas as pd

x1 = pd.read_csv(r'C:\Users\h473\Documents\Indonesia_verbatims W1 2018\Insurance Data X3\example.csv')

x1_sub = x1[x1['score']<=2]

for i in range(len(x1)):
    for j in range(len(x1_sub)):
        if (x1['word'][i]==x1_sub['word'][j]):
            print(i, j)

And this is the output:

0 0
---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
<ipython-input-63-08d55a712c99> in <module>()
      7 for i in range(len(x1)):
      8     for j in range(len(x1_sub)):
----> 9         if (x1['word'][i]==x1_sub['word'][j]):
     10             print(i, j)

c:\users\h473\appdata\local\programs\python\python35\lib\site-packages\pandas\core\series.py in __getitem__(self, key)
    621         key = com._apply_if_callable(key, self)
    622         try:
--> 623             result = self.index.get_value(self, key)
    624 
    625             if not is_scalar(result):

c:\users\h473\appdata\local\programs\python\python35\lib\site-packages\pandas\core\indexes\base.py in get_value(self, series, key)
   2558         try:
   2559             return self._engine.get_value(s, k,
-> 2560                                           tz=getattr(series.dtype, 'tz', None))
   2561         except KeyError as e1:
   2562             if len(self) > 0 and self.inferred_type in ['integer', 'boolean']:

pandas\_libs\index.pyx in pandas._libs.index.IndexEngine.get_value()

pandas\_libs\index.pyx in pandas._libs.index.IndexEngine.get_value()

pandas\_libs\index.pyx in pandas._libs.index.IndexEngine.get_loc()

pandas\_libs\hashtable_class_helper.pxi in pandas._libs.hashtable.Int64HashTable.get_item()

pandas\_libs\hashtable_class_helper.pxi in pandas._libs.hashtable.Int64HashTable.get_item()

KeyError: 1

---

EDIT 2: Also, if x1_sub is a list, then the error is different:

import pandas as pd

x1 = pd.read_csv(r'C:\Users\h473\Documents\Indonesia_verbatims W1 2018\Insurance Data X3\example.csv')

#x1_sub = x1[x1['score']<=2]
x1_sub = ['service', 'claim', 'health', 'fund', 'premium', 'nice', 'process']

for i in range(len(x1)):
    for j in range(len(x1_sub)):
        if (x1['word'][i]==x1_sub['word'][j]):
            print(i, j)

Produces the following output:

TypeError                                 Traceback (most recent call last)
<ipython-input-68-dec8c7e33757> in <module>()
      8 for i in range(len(x1)):
      9     for j in range(len(x1_sub)):
---> 10         if (x1['word'][i]==x1_sub['word'][j]):
     11             print(i, j)

TypeError: list indices must be integers or slices, not str

Answer 1

I think looping is best avoid in pandas, because very slow if exist some vectorized solution:

x1_sub  = ['service', 'claim', 'health', 'fund', 'premium', 'nice', 'process']

x2 = x1[x1['word'].isin(x1_sub)]
print (x2)
       word  score
0   service      1
5     claim      2
15   health      5
17     fund      4
22  premium      1
26     nice      3

Answer 2

尝试使用Dataframe.set_index（keys，inplace = True）分配单独的索引，请参阅此文档https://pandas.pydata.org/pandas-docs/stable/generation/pandas.DataFrame.set_index.html