def pickle_labeled_sentiment_dataset():
dataset_with_label = get_labeled_sentiment_dataset()
pickle_as = os.path.join(get_project_root(),
'data/pickles/sentiment_data_labeled_v1.pickle')
with open(pickle_as, 'wb') as f:
pickle.dump(dataset_with_label, f)
def get_labeled_sentiment_data():
"""
Load pre-pickled sentiment data with labels.
"""
pickle_file = os.path.join(utils.get_project_root(),
'data/pickles/sentiment_data_labeled.pickle')
with open(pickle_file, 'rb') as f:
return pickle.load(f)
def pickle_word_features(num_feat=5000):
default_training_data = utils.get_training_data_path()
word_dict = get_word_freq_dict(default_training_data)
features = list(word_dict.keys())[:num_feat]
pickle_as = os.path.join(get_project_root(),
'data/pickles/word_features_5k.pickle')
with open(pickle_as, 'wb') as f:
pickle.dump(features, f)
def pickle_feature_sets():
dataset_with_label = get_labeled_sentiment_dataset()
# a list of tuples: e.g. [({'good': True, 'silly': False, ...}, 'pos'),
# ({'good': False, 'silly': True, ...}, 'neg'), (...)]
feature_sets = [(find_features(text), sentiment)
for (text, sentiment) in dataset_with_label]
random.shuffle(feature_sets)
pickle_as = os.path.join(get_project_root(),
'data/pickles/feature_sets.pickle')
with open(pickle_as, 'wb') as f:
pickle.dump(feature_sets, f)
def get_word_features():
"""
Returns the top word features in training data.
Use the existing pickled data.
:param num_feat: number of features to consider, defaulted to 5k.
:return: a list of top features (words)
"""
pickle_file = os.path.join(utils.get_project_root(),
'data/pickles/word_features_5k.pickle')
with open(pickle_file, 'rb') as f:
return pickle.load(f)
def train_mnb_clf(training_set, testing_set):
"""
accuracy: 73.28
"""
mnb_classifier = SklearnClassifier(MultinomialNB())
mnb_classifier.train(training_set)
print("Multinomial NB Classifier accuracy:",
(classify.accuracy(mnb_classifier, testing_set)) * 100)
pickle_as = os.path.join(utils.get_project_root(),
'data/classifiers/mnb_classifier_5k.pickle')
with open(pickle_as, 'wb') as f:
pickle.dump(mnb_classifier, f)
def get_feature_sets():
"""
Load pre-pickled feature sets data.
Feature sets structure: a list of tuples: e.g.
[({'good': True, 'silly': False, ...}, 'pos'),
({'good': False, 'silly': True, ...}, 'neg'), (...)]
:return: Feature sets.
"""
pickle_file = os.path.join(utils.get_project_root(),
'data/pickles/feature_sets.pickle')
with open(pickle_file, 'rb') as f:
return pickle.load(f)
def train_linear_svc_clf(training_set, testing_set):
"""
accuracy: 72.01
"""
linear_svc_classifier = SklearnClassifier(LinearSVC())
linear_svc_classifier.train(training_set)
print("LinearSVC Classifier accuracy:",
(classify.accuracy(linear_svc_classifier, testing_set)) * 100)
pickle_as = os.path.join(
utils.get_project_root(),
'data/classifiers/linear_svc_classifier_5k.pickle')
with open(pickle_as, 'wb') as f:
pickle.dump(linear_svc_classifier, f)
def train_bernoulli_nb_clf(training_set, testing_set):
"""
accuracy: 74.64
"""
bernoulli_nb_classifier = SklearnClassifier(BernoulliNB())
bernoulli_nb_classifier.train(training_set)
print("Bernoulli NB Classifier accuracy:",
(classify.accuracy(bernoulli_nb_classifier, testing_set)) * 100)
pickle_as = os.path.join(
utils.get_project_root(),
'data/classifiers/bernoulli_nb_classifier_5k.pickle')
with open(pickle_as, 'wb') as f:
pickle.dump(bernoulli_nb_classifier, f)
def train_naive_bayes_clf(training_set, testing_set):
"""
accuracy: 74.26
"""
naive_bayes_classifier = NaiveBayesClassifier.train(training_set)
print('Naive Bayes model accuracy:',
(classify.accuracy(naive_bayes_classifier, testing_set)) * 100)
naive_bayes_classifier.show_most_informative_features(15)
pickle_as = os.path.join(utils.get_project_root(),
'data/classifiers/naive_bayes_5k.pickle')
with open(pickle_as, 'wb') as f:
pickle.dump(naive_bayes_classifier, f)
def train_logistic_regression_clf(training_set, testing_set):
"""
accuracy: 74.59
"""
logistic_regression_classifier = SklearnClassifier(LogisticRegression())
logistic_regression_classifier.train(training_set)
print('Logistic Regression Classifier accuracy:',
(classify.accuracy(logistic_regression_classifier, testing_set)) *
100)
pickle_as = os.path.join(
utils.get_project_root(),
'data/classifiers/logistic_regression_classifier_5k.pickle')
with open(pickle_as, 'wb') as f:
pickle.dump(logistic_regression_classifier, f)
def load_naive_bayes_clf():
pickle_clf = os.path.join(utils.get_project_root(),
'data/classifiers/naive_bayes_5k.pickle')
with open(pickle_clf, 'rb') as f:
return pickle.load(f)
def load_linear_svc_clf():
pickle_clf = os.path.join(
utils.get_project_root(),
'data/classifiers/linear_svc_classifier_5k.pickle')
with open(pickle_clf, 'rb') as f:
return pickle.load(f)
def load_logistic_regression_clf():
pickle_clf = os.path.join(
utils.get_project_root(),
'data/classifiers/logistic_regression_classifier_5k.pickle')
with open(pickle_clf, 'rb') as f:
return pickle.load(f)
def load_bernoulli_nb_clf():
pickle_clf = os.path.join(
utils.get_project_root(),
'data/classifiers/bernoulli_nb_classifier_5k.pickle')
with open(pickle_clf, 'rb') as f:
return pickle.load(f)