在python pandas中创建新列时出现KeyError

Question

I am trying to create a new column in python pandas, and I keep getting an (unsteady) reoccurring KeyError. The section of the script is very straightforward so I am not sure what could be causing the error since none of the columns in the dataset have the same name.

My goal is to created a new column and append it to the dataframe that contains the new translations of the content of the column ticket_contents. Here is a sample of the data;

25483   0   outstanding 0   Los-Angeles e-payment   delayed Ticket  1/7/19 7:54
39363   0   outstanding 0   Los-Angeles e-payment   delayed Ticket  1/7/19 7:54
83584   0   outstanding 6   Los-Angeles e-payment   delayed Ticket  1/7/19 7:54
34537   0   outstanding 7   Los-Angeles e-payment   lost    Ticket  1/7/19 7:53



colnames = ['id', 'ln_id', 'status', 
'number_outstanding', 'country', 'subject', 'ticket_contents', 'subtopic', 
'date']
test_data = pandas.read_csv(test_data, names = colnames, encoding 
= 'utf-8')
test_data = pandas.DataFrame(test_data)

translated_description = []

from_lang = 'tl'
to_lang = 'en-us'

def test_translation(contents):
    translator = Translator(from_lang = from_lang, to_lang = to_lang)
    translation = translator.translate(contents)
    translated_description.append(translation)
    #print(translated_description)


for contents, row in test_data.iterrows():
    contents = test_data.ticket_contents.iloc[contents -1]
    test_translation(contents)

test_data['translated_descriptions'].copy = translated_description

Here is the error output:

KeyError Traceback (most recent call last)
<ipython-input-70-55e39cf5e328> in <module>()
     16     test_translation(contents)
     17 
---> 18 test_data['translated_descriptions'].copy = translated_description
     19 

/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/pandas/core/frame.pyc in __getitem__(self, key)
   1962             return self._getitem_multilevel(key)
   1963         else:
-> 1964             return self._getitem_column(key)
   1965 
   1966     def _getitem_column(self, key):

/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/pandas/core/frame.pyc in _getitem_column(self, key)
   1969         # get column
   1970         if self.columns.is_unique:
-> 1971             return self._get_item_cache(key)
   1972 
   1973         # duplicate columns & possible reduce dimensionality

/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/pandas/core/generic.pyc in _get_item_cache(self, item)
   1643         res = cache.get(item)
   1644         if res is None:
-> 1645             values = self._data.get(item)
   1646             res = self._box_item_values(item, values)
   1647             cache[item] = res

/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/pandas/core/internals.pyc in get(self, item, fastpath)
   3588 
   3589             if not isnull(item):
-> 3590                 loc = self.items.get_loc(item)
   3591             else:
   3592                 indexer = np.arange(len(self.items))[isnull(self.items)]

/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/pandas/core/indexes/base.pyc in get_loc(self, key, method, tolerance)
   2442                 return self._engine.get_loc(key)
   2443             except KeyError:
-> 2444                 return self._engine.get_loc(self._maybe_cast_indexer(key))
   2445 
   2446         indexer = self.get_indexer([key], method=method, tolerance=tolerance)

pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc (pandas/_libs/index.c:5280)()

pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc (pandas/_libs/index.c:5126)()

pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item (pandas/_libs/hashtable.c:20523)()

pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item (pandas/_libs/hashtable.c:20477)()

KeyError: u'translated_descriptions'

Answer 1

I agree with the comments that you shouldn't be iterating through the dataframe. You should compute all of the values into a list, array, or Series, and assign them all at once.

However your error comes from this line:

test_data['translated_descriptions'].copy = translated_description

What it's doing is overwriting the the copy attribute/method of the test_data['translated_descriptions'] series. Since that series doesn't exist yet, you get an error.

To create a new column with your sequence of values, I would do the following:

test_data = test_data.assign(translated_descriptions=translated_description_values)

Answer 2

The error occurs at:

test_data['translated_descriptions'].copy = translated_description

What does it actually contain:

• test_data['translated_descriptions'].copy - is a reference to copy method of yet not existing column.
• ... = translated_description - you attempt to substitute a list to this reference.

If you want to create a new column, write just:

test_data['translated_descriptions'] = translated_description

Edit

If you want to get rid of the error mentioned in comment, then:

• Start from copying the Dataframe: df2 = test_data.copy() (invoke copy method of the whole DataFrame, not its column).
• Then use df2 - the new DataFrame.

And a couple of hints how to improve your program:

Define translator outside of the translating function:

translator = Translator(from_lang = from_lang, to_lang = to_lang)

Then define the translating function as:

def test_translation(contents):
    return translator.translate(contents)

And then the new colun can be created as simply as:

test_data['translated_descriptions'] = \
    test_data.ticket_contents.apply(test_translation)

without any intermediate list.

Look also at the following fragment of your program:

test_data = pandas.read_csv(test_data, names = colnames,
    encoding = 'utf-8')
test_data = pandas.DataFrame(test_data)

Note that:

• The first instruction reads the DataFrame from CSV file and saves it under test_data variable.
• Then you create a next DataFrame (actually a view of the existing DataFrame), and assign it to the same variable.

The result is that:

• The previous DataFrame exists somewhere, but is now unreachable.
• You have access only to the view, created with the second instruction.
• And this is why you get the mentioned error.

Conclusion: Drop the second instruction. It is enough to have one DataFrame.