# This module is a port of the Textblob Averaged Perceptron Tagger
# Author: Matthew Honnibal <honnibal+gh@gmail.com>,
# Long Duong <longdt219@gmail.com> (NLTK port)
# URL: <https://github.com/sloria/textblob-aptagger>
# <https://www.nltk.org/>
# Copyright 2013 Matthew Honnibal
# NLTK modifications Copyright 2015 The NLTK Project
#
# This module is provided under the terms of the MIT License.
import json
import logging
import random
from collections import defaultdict
from os.path import join as path_join
from tempfile import gettempdir
from nltk import jsontags
from nltk.data import find, load
from nltk.tag.api import TaggerI
try:
import numpy as np
except ImportError:
pass
[docs]
@jsontags.register_tag
class AveragedPerceptron:
"""An averaged perceptron, as implemented by Matthew Honnibal.
See more implementation details here:
https://explosion.ai/blog/part-of-speech-pos-tagger-in-python
"""
json_tag = "nltk.tag.perceptron.AveragedPerceptron"
[docs]
def __init__(self, weights=None):
# Each feature gets its own weight vector, so weights is a dict-of-dicts
self.weights = weights if weights else {}
self.classes = set()
# The accumulated values, for the averaging. These will be keyed by
# feature/clas tuples
self._totals = defaultdict(int)
# The last time the feature was changed, for the averaging. Also
# keyed by feature/clas tuples
# (tstamps is short for timestamps)
self._tstamps = defaultdict(int)
# Number of instances seen
self.i = 0
def _softmax(self, scores):
s = np.fromiter(scores.values(), dtype=float)
exps = np.exp(s)
return exps / np.sum(exps)
[docs]
def predict(self, features, return_conf=False):
"""Dot-product the features and current weights and return the best label."""
scores = defaultdict(float)
for feat, value in features.items():
if feat not in self.weights or value == 0:
continue
weights = self.weights[feat]
for label, weight in weights.items():
scores[label] += value * weight
# Do a secondary alphabetic sort, for stability
best_label = max(self.classes, key=lambda label: (scores[label], label))
# compute the confidence
conf = max(self._softmax(scores)) if return_conf == True else None
return best_label, conf
[docs]
def update(self, truth, guess, features):
"""Update the feature weights."""
def upd_feat(c, f, w, v):
param = (f, c)
self._totals[param] += (self.i - self._tstamps[param]) * w
self._tstamps[param] = self.i
self.weights[f][c] = w + v
self.i += 1
if truth == guess:
return None
for f in features:
weights = self.weights.setdefault(f, {})
upd_feat(truth, f, weights.get(truth, 0.0), 1.0)
upd_feat(guess, f, weights.get(guess, 0.0), -1.0)
[docs]
def average_weights(self):
"""Average weights from all iterations."""
for feat, weights in self.weights.items():
new_feat_weights = {}
for clas, weight in weights.items():
param = (feat, clas)
total = self._totals[param]
total += (self.i - self._tstamps[param]) * weight
averaged = round(total / self.i, 3)
if averaged:
new_feat_weights[clas] = averaged
self.weights[feat] = new_feat_weights
[docs]
def save(self, path):
"""Save the model weights as json"""
with open(path, "w") as fout:
return json.dump(self.weights, fout)
[docs]
def load(self, path):
"""Load the json model weights."""
with open(path) as fin:
self.weights = json.load(fin)
[docs]
def encode_json_obj(self):
return self.weights
[docs]
@classmethod
def decode_json_obj(cls, obj):
return cls(obj)
[docs]
@jsontags.register_tag
class PerceptronTagger(TaggerI):
"""
Greedy Averaged Perceptron tagger, as implemented by Matthew Honnibal.
See more implementation details here:
https://explosion.ai/blog/part-of-speech-pos-tagger-in-python
>>> from nltk.tag.perceptron import PerceptronTagger
>>> tagger = PerceptronTagger(load=False)
Train and save the model:
>>> tagger.train([[('today','NN'),('is','VBZ'),('good','JJ'),('day','NN')],
... [('yes','NNS'),('it','PRP'),('beautiful','JJ')]], save_loc=tagger.save_dir)
Load the saved model:
>>> tagger2 = PerceptronTagger(loc=tagger.save_dir)
>>> print(sorted(list(tagger2.classes)))
['JJ', 'NN', 'NNS', 'PRP', 'VBZ']
>>> print(tagger2.classes == tagger.classes)
True
>>> tagger2.tag(['today','is','a','beautiful','day'])
[('today', 'NN'), ('is', 'PRP'), ('a', 'PRP'), ('beautiful', 'JJ'), ('day', 'NN')]
Use the pretrain model (the default constructor)
>>> pretrain = PerceptronTagger()
>>> pretrain.tag('The quick brown fox jumps over the lazy dog'.split())
[('The', 'DT'), ('quick', 'JJ'), ('brown', 'NN'), ('fox', 'NN'), ('jumps', 'VBZ'), ('over', 'IN'), ('the', 'DT'), ('lazy', 'JJ'), ('dog', 'NN')]
>>> pretrain.tag("The red cat".split())
[('The', 'DT'), ('red', 'JJ'), ('cat', 'NN')]
"""
json_tag = "nltk.tag.perceptron.PerceptronTagger"
START = ["-START-", "-START2-"]
END = ["-END-", "-END2-"]
[docs]
def __init__(self, load=True, lang="eng", loc=None):
"""
:param load: Load the json model upon instantiation.
"""
self.model = AveragedPerceptron()
self.tagdict = {}
self.classes = set()
self.lang = lang
# Save trained models in tmp directory by default:
self.TRAINED_TAGGER_PATH = gettempdir()
self.TAGGER_NAME = "averaged_perceptron_tagger"
self.save_dir = path_join(
self.TRAINED_TAGGER_PATH, f"{self.TAGGER_NAME}_{self.lang}"
)
if load:
self.load_from_json(lang, loc)
[docs]
def param_files(self, lang="eng"):
return (
f"{self.TAGGER_NAME}_{lang}.{attr}.json"
for attr in ["weights", "tagdict", "classes"]
)
[docs]
def tag(self, tokens, return_conf=False, use_tagdict=True):
"""
Tag tokenized sentences.
:params tokens: list of word
:type tokens: list(str)
"""
prev, prev2 = self.START
output = []
context = self.START + [self.normalize(w) for w in tokens] + self.END
for i, word in enumerate(tokens):
tag, conf = (
(self.tagdict.get(word), 1.0) if use_tagdict == True else (None, None)
)
if not tag:
features = self._get_features(i, word, context, prev, prev2)
tag, conf = self.model.predict(features, return_conf)
output.append((word, tag, conf) if return_conf == True else (word, tag))
prev2 = prev
prev = tag
return output
[docs]
def train(self, sentences, save_loc=None, nr_iter=5):
"""Train a model from sentences, and save it at ``save_loc``. ``nr_iter``
controls the number of Perceptron training iterations.
:param sentences: A list or iterator of sentences, where each sentence
is a list of (words, tags) tuples.
:param save_loc: If not ``None``, saves a json model in this location.
:param nr_iter: Number of training iterations.
"""
# We'd like to allow ``sentences`` to be either a list or an iterator,
# the latter being especially important for a large training dataset.
# Because ``self._make_tagdict(sentences)`` runs regardless, we make
# it populate ``self._sentences`` (a list) with all the sentences.
# This saves the overheard of just iterating through ``sentences`` to
# get the list by ``sentences = list(sentences)``.
self._sentences = list() # to be populated by self._make_tagdict...
self._make_tagdict(sentences)
self.model.classes = self.classes
for iter_ in range(nr_iter):
c = 0
n = 0
for sentence in self._sentences:
words, tags = zip(*sentence)
prev, prev2 = self.START
context = self.START + [self.normalize(w) for w in words] + self.END
for i, word in enumerate(words):
guess = self.tagdict.get(word)
if not guess:
feats = self._get_features(i, word, context, prev, prev2)
guess, _ = self.model.predict(feats)
self.model.update(tags[i], guess, feats)
prev2 = prev
prev = guess
c += guess == tags[i]
n += 1
random.shuffle(self._sentences)
logging.info(f"Iter {iter_}: {c}/{n}={_pc(c, n)}")
# We don't need the training sentences anymore, and we don't want to
# waste space on them when we the trained tagger.
self._sentences = None
self.model.average_weights()
# Save to json files.
if save_loc is not None:
self.save_to_json(lang=self.lang, loc=save_loc)
[docs]
def save_to_json(self, lang="xxx", loc=None):
from os import mkdir
from os.path import isdir
if not loc:
loc = self.save_dir
if not isdir(loc):
mkdir(loc)
for param, json_file in zip(self.encode_json_obj(), self.param_files(lang)):
with open(path_join(loc, json_file), "w") as fout:
json.dump(param, fout)
[docs]
def load_from_json(self, lang="eng", loc=None):
# Automatically find path to the tagger if location is not specified.
if not loc:
loc = find(f"taggers/averaged_perceptron_tagger_{lang}")
def load_param(json_file):
with open(path_join(loc, json_file)) as fin:
return json.load(fin)
self.decode_json_params(
load_param(js_file) for js_file in self.param_files(lang)
)
[docs]
def decode_json_params(self, params):
weights, tagdict, class_list = params
self.model.weights = weights
self.tagdict = tagdict
self.classes = self.model.classes = set(class_list)
[docs]
def encode_json_obj(self):
return self.model.weights, self.tagdict, list(self.classes)
[docs]
@classmethod
def decode_json_obj(cls, obj):
tagger = cls(load=False)
tagger.decode_json_params(obj)
return tagger
[docs]
def normalize(self, word):
"""
Normalization used in pre-processing.
- All words are lower cased
- Groups of digits of length 4 are represented as !YEAR;
- Other digits are represented as !DIGITS
:rtype: str
"""
if "-" in word and word[0] != "-":
return "!HYPHEN"
if word.isdigit() and len(word) == 4:
return "!YEAR"
if word and word[0].isdigit():
return "!DIGITS"
return word.lower()
def _get_features(self, i, word, context, prev, prev2):
"""Map tokens into a feature representation, implemented as a
{hashable: int} dict. If the features change, a new model must be
trained.
"""
def add(name, *args):
features[" ".join((name,) + tuple(args))] += 1
i += len(self.START)
features = defaultdict(int)
# It's useful to have a constant feature, which acts sort of like a prior
add("bias")
add("i suffix", word[-3:])
add("i pref1", word[0] if word else "")
add("i-1 tag", prev)
add("i-2 tag", prev2)
add("i tag+i-2 tag", prev, prev2)
add("i word", context[i])
add("i-1 tag+i word", prev, context[i])
add("i-1 word", context[i - 1])
add("i-1 suffix", context[i - 1][-3:])
add("i-2 word", context[i - 2])
add("i+1 word", context[i + 1])
add("i+1 suffix", context[i + 1][-3:])
add("i+2 word", context[i + 2])
return features
def _make_tagdict(self, sentences):
"""
Make a tag dictionary for single-tag words.
:param sentences: A list of list of (word, tag) tuples.
"""
counts = defaultdict(lambda: defaultdict(int))
for sentence in sentences:
self._sentences.append(sentence)
for word, tag in sentence:
counts[word][tag] += 1
self.classes.add(tag)
freq_thresh = 20
ambiguity_thresh = 0.97
for word, tag_freqs in counts.items():
tag, mode = max(tag_freqs.items(), key=lambda item: item[1])
n = sum(tag_freqs.values())
# Don't add rare words to the tag dictionary
# Only add quite unambiguous words
if n >= freq_thresh and (mode / n) >= ambiguity_thresh:
self.tagdict[word] = tag
def _pc(n, d):
return (n / d) * 100
def _train_and_test(lang="sv"):
"""
Train and test on 'lang' part of universal_treebanks corpus, which includes
train and test sets in conll format for 'de', 'es', 'fi', 'fr' and 'sv'.
Finds 0.94 accuracy on 'sv' (Swedish) test set.
"""
from nltk.corpus import universal_treebanks as utb
tagger = PerceptronTagger(load=False, lang=lang)
training = utb.tagged_sents(f"ch/{lang}/{lang}-universal-ch-train.conll")
testing = utb.tagged_sents(f"ch/{lang}/{lang}-universal-ch-test.conll")
print(
f"(Lang = {lang}) training on {len(training)} and testing on {len(testing)} sentences"
)
# Train and save the model
tagger.train(training, save_loc=tagger.save_dir)
print("Accuracy : ", tagger.accuracy(testing))
if __name__ == "__main__":
_train_and_test()