python 词频统计

"""Count words."""

def count_words(s, n):

    """Return the n most frequently occuring words in s."""

    from operator import itemgetter, attrgetter 

    # TODO: Count the number of occurences of each word in s

    strl_ist = s.replace('\n', ' ').lower().split(' ')

    count_dict = {}

    for str in strl_ist:

        if str in count_dict.keys():

            count_dict[str] = count_dict[str] + 1

        else:

            count_dict[str] = 1

    # TODO: Sort the occurences in descending order (alphabetically in case of ties)

    count_list=sorted(count_dict.iteritems(),key=itemgetter(1),reverse=True)

    top_n = count_list[0:n]

    # TODO: Return the top n most frequent words.

    return top_n

def test_run():

    """Test count_words() with some inputs."""

    print count_words("cat bat mat cat bat cat", 3)

    print count_words("betty bought a bit of butter but the butter was bitter", 3)

if __name__ == '__main__':

    test_run()