Das deutsche Python-Forum

Code: Alles auswählen

dat = """The AT
Fulton NP
County NN
an AT
an AT
an AT
an NN
an IN
investigation NN
investigation XXX
investigation NN
of IN
Atlanta's NP$
recent JJ
primary NN
election NN
produced VBD """

lines = dat.splitlines()  # ersetzen durch Lesen aus Datei
lines = map(str.split, lines)

lexicon = {}
for word, tag in lines:
    if not lexicon.has_key(word):
        lexicon[word] = {tag:1}
    else:
        try:
            lexicon[word][tag] += 1
        except:
            lexicon[word][tag] = 1

Code: Alles auswählen

>>> lexicon
{'produced': {'VBD': 1}, 'of': {'IN': 1}, "Atlanta's": {'NP$': 1}, 'primary': {'NN': 1}, 'an': {'AT': 3, 'NN': 1, 'IN': 1}, 'County': {'NN': 1}, 'investigation': {'XXX': 1, 'NN': 2}, 'election': {'NN': 1}, 'The': {'AT': 1}, 'Fulton': {'NP': 1}, 'recent': {'JJ': 1}}

>>> for a, b in zip(lexicon.keys(), lexicon.values()):
        print a, b

        
produced {'VBD': 1}
of {'IN': 1}
Atlanta's {'NP$': 1}
primary {'NN': 1}
an {'AT': 3, 'NN': 1, 'IN': 1}
County {'NN': 1}
investigation {'XXX': 1, 'NN': 2}
election {'NN': 1}
The {'AT': 1}
Fulton {'NP': 1}
recent {'JJ': 1}

Code: Alles auswählen

from __future__ import division, with_statement
from collections import defaultdict
from functools import partial
from pprint import pprint


def main():
    lexicon = defaultdict(partial(defaultdict, int))
    with open('test.txt') as lines:
        for line in lines:
            word, tag = line.rstrip().split()
            lexicon[word][tag] += 1
    
    for word, tags in sorted(lexicon.iteritems()):
        print '%s: %s' % (word, sorted(tags.iteritems()))


if __name__ == "__main__":
    main()

Code: Alles auswählen

from collections import defaultdict
from functools import partial