sentiment-analysis/mlp-nn.py at master · alexanderpp/sentiment-analysis · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
from sklearn.metrics import accuracy_score
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.neural_network import MLPClassifier
from nltk.corpus import stopwords
from stop_words import get_stop_words
import sys
sys.path.append('.')
import utils

max_features = 3000

print "Loading data ..."
x_train, x_test, y_train, y_test = utils.load_data()
# x_train, x_test, y_train, y_test = utils.load_data2()

sw = stopwords.words('english') + get_stop_words("english")

print "Extracting features..."
vectorizer = CountVectorizer(analyzer = "word",
                             max_features = max_features)

x_train, x_test, y_train, y_test = utils.extract_features(x_train, x_test, y_train, y_test, vectorizer)

print "Creating and training the model..."

model = MLPClassifier(solver='sgd',
                      hidden_layer_sizes=(10, 10, 10),
                      random_state=5,
                      verbose=True)

model.fit(x_train, y_train)

print "Predicting..."
y_pred = model.predict(x_test)

print "Evaluating..."
accuracy = accuracy_score(y_test, y_pred)
print("Accuracy: %.2f%%" % (accuracy * 100.0))