-
Notifications
You must be signed in to change notification settings - Fork 0
/
test_env.py
227 lines (176 loc) · 6.7 KB
/
test_env.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
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
import pandas
from tensorflow.keras import backend as K
from methodology.preprocessing.vectorization import *
# import tensorflow.contrib.eager as tfe
# tf.enable_eager_execution()
df = pandas.read_csv('methodology/data/dailydialog/dailydialog.csv')
print('Dataset', df.shape, df.columns.values, sep='\n',
end='\n========================================================\n')
columns_names = df.columns.values
features_names = columns_names[:-1]
label_name = columns_names[-1]
batch_size = 100
df = tf.data.experimental.CsvDataset('methodology/data/dailydialog/dailydialog.csv', columns_names)
# df = tf.keras.utils.get_file(fname='methodology/data/dailydialog/dailydialog.csv')
#
# df = tf.contrib.data.make_csv_dataset(df,batch_size,column_names=columns_names,label_name=label_name, num_epochs=1)
print(type(df))
print(df)
# for row in df.as_numpy_iterator():
# print(row[-1])
tok_df = df.map(tokenize)
tok_df = configure_dataset(tok_df)
vocab_dict = collections.defaultdict(lambda: 0)
# print(vocab_dict)
print('Loading Vocabulary...')
for item in tok_df.as_numpy_iterator():
# print(item[1])
for token in item[1]:
vocab_dict[token] += 1
print('Vocabulary loaded...')
vocab = sorted(vocab_dict.items(), key=lambda x: x[1], reverse=True)
vocab = [token for token, count in vocab]
vocab_size = len(vocab)
print("Vocab size: ", vocab_size)
print("First five vocab entries:", vocab[:10])
keys = vocab
values = range(2, len(vocab) + 2) # reserve 0 for padding, 1 for OOV
init = tf.lookup.KeyValueTensorInitializer(
keys, values, key_dtype=tf.string, value_dtype=tf.int64)
num_oov_buckets = 1
vocab_table = tf.lookup.StaticVocabularyTable(init, num_oov_buckets)
print(vocab_table)
# index, sentences, act_label, emotion_label, speaker, conv_id, utt_id, labels = next(iter(df))
_, sentences, _, _, _, _, _, labels = next(df.as_numpy_iterator())
print(sentences)
print(labels)
vxt, em = preprocess_text(sentences, vocab_table, labels)
print(vxt.numpy(), em)
def preprocess_punctuation(x):
for punctuation in '"\'!&?.,}-/<>#$%\()*+:;=?@[\\]^_`|\~':
x = x.replace(punctuation, ' ')
x = ' '.join(x.split())
x = x.lower()
return x
def preprocess_text_test(num, text, act_label, emotion_label, speaker, conv_id, utt_id, emotion_index):
tokenizer = tf_text.UnicodeScriptTokenizer()
standardized = tf_text.case_fold_utf8(text)
tokenized = tokenizer.tokenize(standardized)
vectorized = vocab_table.lookup(tokenized)
emotion_index = tf.strings.to_number(emotion_index, out_type=tf.int64)
return vectorized, emotion_index
encoded_df = df.map(preprocess_text_test)
# count = 0
# for row in encoded_df.as_numpy_iterator():
# count += 1
# print(count)
# for row in encoded_df.take(10):
# print(row[1])
"""Splitting Dataset into train and test"""
VALID_SIZE = 9268 + 6179
BATCH_SIZE = 32
TRAIN_SIZE = 87532
train_data = encoded_df.skip(VALID_SIZE)
valid_data = encoded_df.take(VALID_SIZE)
valid_data = valid_data.skip(6179)
test_data = valid_data.take(6179)
print('Dataset Split Successfully')
vocab_size += 2
max_len = 0
for row in encoded_df.as_numpy_iterator():
if max_len < len(row[0]):
max_len = len(row[0])
print(max_len)
print('Got MAX-LEN....')
def padding_data(inputs, label):
pad_size = abs(inputs.get_shape()[1] - max_len)
inputs = inputs.padded_batch(pad_size)
# print(inputs)
# print(inputs.get_shape(), type(inputs))
# inputs = inputs.eval(session=tf.compat.v1.Session())
# # inputs = tf.constant(inputs.numpy())
# # inputs = tf.make_tensor_proto(inputs)
# inputs = tf.make_ndarray(inputs)
# if len(inputs) < max_len:
# arr = numpy.zeros(max_len - len(inputs), dtype=int)
# inputs = numpy.append(inputs, arr)
return inputs, label
print(train_data)
print(valid_data)
print(test_data)
print('Padding Data...')
# train_data = train_data.map(padding_data)
# valid_data = valid_data.map(padding_data)
# test_data = test_data.map(padding_data)
# train_data = train_data.padded_batch(1, padded_shapes=max_len)
# valid_data = valid_data.padded_batch(1, padded_shapes=max_len)
# test_data = test_data.padded_batch(1, padded_shapes=max_len)
train_data = train_data.padded_batch(BATCH_SIZE)
valid_data = valid_data.padded_batch(BATCH_SIZE)
test_data = test_data.padded_batch(BATCH_SIZE)
train_data = configure_dataset(train_data)
valid_data = configure_dataset(valid_data)
test_data = configure_dataset(test_data)
count = 0
for row in train_data.as_numpy_iterator():
# print(row[0].shape)
count += 1
print(count)
count = 0
for row in valid_data.as_numpy_iterator():
count += 1
print(count)
count = 0
for row in test_data.as_numpy_iterator():
count += 1
print(count)
def create_model(vb_size, num_labels):
model_ = tf.keras.Sequential([
tf.keras.layers.Embedding(vb_size, 64, mask_zero=True),
tf.keras.layers.Conv1D(64, 5, padding='valid', activation='relu', strides=2),
tf.keras.layers.GlobalMaxPool1D(),
tf.keras.layers.Dense(num_labels)
])
return model_
def recall_m(y_true, y_predict):
true_positives = K.sum(K.round(K.clip(y_true * y_predict, 0, 1)))
possible_positives = K.sum(K.round(K.clip(y_true, 0, 1)))
recall = true_positives / (possible_positives + K.epsilon())
return recall
def precision_m(y_true, y_predict):
true_positives = K.sum(K.round(K.clip(y_true * y_predict, 0, 1)))
predicted_positives = K.sum(K.round(K.clip(y_predict, 0, 1)))
precision = true_positives / (predicted_positives + K.epsilon())
return precision
def f1_m(y_true, y_predict):
precision = precision_m(y_true, y_predict)
recall = recall_m(y_true, y_predict)
return 2 * ((precision * recall) / (precision + recall + K.epsilon()))
model = create_model(vocab_size, 7)
model.summary()
model.compile(
optimizer='adam',
loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
metrics=['accuracy', f1_m, precision_m, recall_m])
print(type(train_data))
print(train_data)
# train_data = to_categorical(train_data, 7)
history = model.fit(train_data, validation_data=valid_data, epochs=10)
y_predict_prob = model.predict(test_data)
# y_predict_classes = model.predict_classes(test_data)
# loss, acc = model.evaluate(test_data)
mtr = model.evaluate(test_data)
# print(y_predict[0])
print(y_predict_prob)
# print(y_predict_classes)
print("\nLoss: ", mtr[0])
print("Accuracy: {:2.2%}".format(mtr[1]))
print(mtr)
print(history.history['accuracy'])
# print(history.history['accuracy'])
for row, y in zip(test_data, y_predict_prob):
print(row[-1], round(max(y)), sep='\t-> ')
# score_f1_micro = f1_score(test_data[1], y_predict_prob, average='micro')
# score_f1_macro = f1_score(test_data[1], y_predict_prob, average='macro')
#
# print(score_f1_micro, score_f1_macro)