Ошибка при использовании Beautiful Soup в Python

Question

Мой код отлично работает. Однако с некоторыми данными мой код дает ошибку. проблематичные данные: T 10 в этом месяце. В честь юбилея и предстоящего T @ 10 Issue, эта серия оглядывается на некоторые из самых запоминающихся историй из журналов в первом десятилетии.

сообщенная проблема является

Traceback (most recent call last): File "/Users/mas/Documents/workspace/DeepLearning/BagOfWords.py", line 41, in clean_train_reviews.append(" ".join(KaggleWord2VecUtility.review_to_wordlist(train["Snippet"][i], True))) File "/Users/mas/Documents/workspace/DeepLearning/KaggleWord2VecUtility.py", line 22, in review_to_wordlist review_text = BeautifulSoup(review).get_text() File "/Library/Python/2.7/site-packages/bs4/init.py", line 162, in init elif len(markup) <= 256: TypeError: object of type 'float' has no len()

код:

def deprecated_argument(old_name, new_name): 
     if old_name in kwargs: 
      warnings.warn(
       'The "%s" argument to the BeautifulSoup constructor ' 
       'has been renamed to "%s."' % (old_name, new_name)) 
      value = kwargs[old_name] 
      del kwargs[old_name] 
      return value 
     return None 

    parse_only = parse_only or deprecated_argument(
     "parseOnlyThese", "parse_only") 

    from_encoding = from_encoding or deprecated_argument(
     "fromEncoding", "from_encoding") 

    if len(kwargs) > 0: 
     arg = kwargs.keys().pop() 
     raise TypeError(
      "__init__() got an unexpected keyword argument '%s'" % arg) 

    if builder is None: 
     if isinstance(features, basestring): 
      features = [features] 
     if features is None or len(features) == 0: 
      features = self.DEFAULT_BUILDER_FEATURES 
     builder_class = builder_registry.lookup(*features) 
     if builder_class is None: 
      raise FeatureNotFound(
       "Couldn't find a tree builder with the features you " 
       "requested: %s. Do you need to install a parser library?" 
       % ",".join(features)) 
     builder = builder_class() 
    self.builder = builder 
    self.is_xml = builder.is_xml 
    self.builder.soup = self 

    self.parse_only = parse_only 

    if hasattr(markup, 'read'):  # It's a file-type object. 
     markup = markup.read() 
    elif len(markup) <= 256: 
     # Print out warnings for a couple beginner problems 
     # involving passing non-markup to Beautiful Soup. 
     # Beautiful Soup will still parse the input as markup, 
     # just in case that's what the user really wants. 
     if (isinstance(markup, unicode) 
      and not os.path.supports_unicode_filenames): 
      possible_filename = markup.encode("utf8") 
     else: 
      possible_filename = markup 
     is_file = False 
     try: 
      is_file = os.path.exists(possible_filename) 
     except Exception, e: 
      # This is almost certainly a problem involving 
      # characters not valid in filenames on this 
      # system. Just let it go. 
      pass 
     if is_file: 
      warnings.warn(
       '"%s" looks like a filename, not markup. You should probably open this file and pass the filehandle into Beautiful Soup.' % markup) 
     if markup[:5] == "http:" or markup[:6] == "https:": 
      # TODO: This is ugly but I couldn't get it to work in 
      # Python 3 otherwise. 
      if ((isinstance(markup, bytes) and not b' ' in markup) 
       or (isinstance(markup, unicode) and not u' ' in markup)): 
       warnings.warn(
        '"%s" looks like a URL. Beautiful Soup is not an HTTP client. You should probably use an HTTP client to get the document behind the URL, and feed that document to Beautiful Soup.' % markup) 

    for (self.markup, self.original_encoding, self.declared_html_encoding, 
    self.contains_replacement_characters) in (
     self.builder.prepare_markup(markup, from_encoding)): 
     self.reset() 
     try: 
      self._feed() 
      break 
     except ParserRejectedMarkup: 
      pass 

    # Clear out the markup and remove the builder's circular 
    # reference to this object. 
    self.markup = None 
    self.builder.soup = None 

это мой основной код:

import os 
from sklearn.feature_extraction.text import CountVectorizer 
from sklearn.ensemble import RandomForestClassifier 
from KaggleWord2VecUtility import KaggleWord2VecUtility 
import pandas as pd 
import numpy as np 

if __name__ == '__main__': 
    train = pd.read_csv(os.path.join(os.path.dirname(__file__), 'data', 'NYTimesBlogTrain.csv'), header=0) 
    test = pd.read_csv(os.path.join(os.path.dirname(__file__), 'data', 'NYTimesBlogTest.csv'), header=0) 

    print 'A sample Abstract is:' 
    print train["Abstract"][2838] 

    print 'A sample Snippet is:' 
    print train["Snippet"][2838] 
    #raw_input("Press Enter to continue...") 


    #print 'Download text data sets. If you already have NLTK datasets downloaded, just close the Python download window...' 
    #nltk.download() # Download text data sets, including stop words 

    # Initialize an empty list to hold the clean reviews 
    clean_train_reviews = [] 

    # Loop over each review; create an index i that goes from 0 to the length 
    # of the movie review list 
    print len(train["Snippet"]) 
    print "Cleaning and parsing the training set abstracts...\n" 
    for i in xrange(0, 3000): 
     clean_train_reviews.append(" ".join(KaggleWord2VecUtility.review_to_wordlist(train["Snippet"][i], True))) 
     if not train["Snippet"][i]: 
      print i 
# 

Answer 1

Не видя полный контекст (например, значение review что передается в конструктор) возможно ли, что ваш метод KaggleWord2VecUtility разбивается на @ sy mbol и/или цифру, так что токен передается как float вместо объекта string/unicode? Исключение указывает на то, что markup является неожиданным поплавок при инициализации ожидает строки или Юникода объект

def __init__(self, markup="", features=None, builder=None, 
      parse_only=None, from_encoding=None, **kwargs): 

Answer 2

Для меня эта проблема была вызвана тем, что некоторые образцы, не содержащие никаких данных вообще в функции обзора. Вы можете изменить это и установить образцы без каких-либо исключений, используя следующие данные:

train = train.fillna(" ")