-
Notifications
You must be signed in to change notification settings - Fork 26
/
Copy pathsentence_convnet_final.py
112 lines (83 loc) · 2.9 KB
/
sentence_convnet_final.py
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
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
__author__ = 'mangate'
import cPickle
from model import Model
import process_data_mr
import process_data_tweets
import process_data_sst1
import process_data_sst2
import process_data_subj
import process_data_trec
import process_data_politeness2
import process_data_opi
import process_data_irony
import tensorflow as tf
import warnings
warnings.filterwarnings("ignore")
#Flags
tf.flags.DEFINE_boolean("random",False,"Initialize with random word embeddings (default: False)")
tf.flags.DEFINE_boolean("static",False,"Keep the word embeddings static (default: False)")
FLAGS =tf.flags.FLAGS
def evaluate(x, num_classes = 2, k_fold = 10):
revs, embedding, W2, word_idx_map, vocab, max_l = x[0], x[1], x[2], x[3], x[4], x[5]
if FLAGS.random:
embedding = W2
embedding_dim = 300
vocab_size = len(vocab) + 1
filter_sizes = [3, 4, 5]
num_filters = 100
vector_length = max_l + 2 * 4
cnn_model = Model()
trainable = not FLAGS.static
cnn_model.build_model(embedding_dim, vocab_size, filter_sizes, num_filters, vector_length, num_classes, trainable)
cnn_model.run(revs, embedding, word_idx_map, max_l, k_fold)
def evaluate_mr():
process_data_mr.process_data("data/processed/mr.p")
x = cPickle.load(open("data/processed/mr.p", "rb"))
evaluate(x, 2, 10)
def evaluate_tweets():
process_data_tweets.process_data("data/processed/twitter.p")
x = cPickle.load(open("data/processed/twitter.p", "rb"))
evaluate(x, 10, 1)
def evaluate_sst1():
process_data_sst1.process_data("data/processed/sst1.p")
x = cPickle.load(open("data/processed/sst1.p", "rb"))
evaluate(x, 5, 1)
def evaluate_sst2():
process_data_sst2.process_data("data/processed/sst2.p")
x = cPickle.load(open("data/processed/sst2.p", "rb"))
evaluate(x, 2, 1)
def evaluate_subj():
process_data_subj.process_data("data/processed/subj.p")
x = cPickle.load(open("data/processed/subj.p", "rb"))
evaluate(x, 2, 10)
def evaluate_trec():
process_data_trec.process_data("data/processed/trec.p")
x = cPickle.load(open("data/processed/trec.p", "rb"))
evaluate(x, 6, 1)
def evaluate_cr():
# couldn't find the dataset...
pass
def evaluate_mpqa():
# too complicated..
pass
def evaluate_politeness():
process_data_politeness2.process_data("data/processed/politeness.p")
x = cPickle.load(open("data/processed/politeness.p", "rb"))
evaluate(x, 2, 10)
def evaluate_opi():
process_data_opi.process_data("data/processed/opi.p")
x = cPickle.load(open("data/processed/opi.p", "rb"))
evaluate(x, 6, 10)
def evaluate_irony():
process_data_irony.process_data("data/processed/irony.p")
x = cPickle.load(open("data/processed/irony.p", "rb"))
evaluate(x, 2, 10)
if __name__=="__main__":
#evaluate_mr()
#evaluate_tweets()
#evaluate_sst1()
#evaluate_sst2()
#evaluate_subj()
#evaluate_trec()
#evaluate_politeness()
evaluate_irony()