-
Notifications
You must be signed in to change notification settings - Fork 1
/
one_hot_bilstm_functional.py
139 lines (119 loc) · 5.44 KB
/
one_hot_bilstm_functional.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
# -*- coding: utf-8 -*-
"""
This is an attempt to use a single word embedding for multiple filters by combining
the Graph and Sequential containers.
@author: ameasure
1626s/epoch gpu
4480s/epoch cpu (estimated)
"""
import datetime
import os
import numpy as np
np.random.seed(42)
from sklearn.preprocessing import LabelEncoder
import keras
from keras.models import Model
from keras.engine.topology import Merge
from keras.layers import Input, Dense, Dropout, Activation, Flatten, Reshape, merge
from keras.layers.recurrent import LSTM
from keras.layers.convolutional import MaxPooling1D
from keras.layers.normalization import BatchNormalization
from keras.utils import np_utils
from keras.preprocessing.text import Tokenizer
from keras.preprocessing.sequence import pad_sequences
from msha_extractor import get_data
#output needs to be (nb_samples, timesteps, input_dim)
def generate_one_hot(X, Y, vocab_size, batch_size):
"""
Inputs:
X: [n_samples, timesteps] each value is the index of a token
Y: [n_samples, n_categories] already one hot
Returns: dictionary with 'input': [n_samples, n_timesteps, vocab_size] and 'output': [n_samples, n_categories]
"""
n_samples = len(X)
seq_len = len(X[0])
start = 0
while 1:
stop = start + batch_size
X_subset = X[start: stop]
X_out = np.zeros([batch_size, seq_len, vocab_size])
index_1 = np.repeat(np.arange(batch_size), seq_len).reshape(batch_size, seq_len)
index_2 = np.arange(seq_len)
X_out[index_1, index_2, X_subset] = 1
Y_out = Y[start: stop]
start += batch_size
if (start + batch_size) > n_samples:
print 'reshuffling, %s + %s > %s' % (start, batch_size, n_samples)
remaining_X = X[start: start + batch_size]
remaining_Y = Y[start: start + batch_size]
random_index = np.random.permutation(n_samples)
X = np.vstack((remaining_X, X[random_index, :]))
Y = np.vstack((remaining_Y, Y[random_index, :]))
start = 0
n_samples = len(X)
yield X_out, Y_out
max_len = 100
gen_batch_size = 16
checkpoint_dir = r'C:\Users\ameasure\Desktop\Programming Projects\cnn\checkpoints'
code_type = 'ACTIVITY_CD'
raw_train, raw_valid, raw_test = get_data(n_train=47500, n_valid=2500, n_test=10000)
raw_train_labels = raw_train[code_type]
raw_valid_labels = raw_valid[code_type]
raw_test_labels = raw_test[code_type]
labeler = LabelEncoder()
labels = set(raw_train[code_type].tolist() + raw_test[code_type].tolist())
labeler.fit(list(labels))
nb_classes = len(set(labels))
print('nb_classes = %s' % nb_classes)
y_train = labeler.transform(raw_train_labels)
Y_train = np_utils.to_categorical(y_train, nb_classes)
y_valid = labeler.transform(raw_valid_labels)
Y_valid = np_utils.to_categorical(y_valid, nb_classes)
y_test = labeler.transform(raw_test_labels)
Y_test = np_utils.to_categorical(y_test, nb_classes)
print 'Tokenizing X_train'
tokenizer = Tokenizer(nb_words=5000)
tokenizer.fit_on_texts(raw_train['NARRATIVE'])
tokenizer.word_index['BLANK_WORD'] = 0
X_train = tokenizer.texts_to_sequences(raw_train['NARRATIVE'])
X_valid = tokenizer.texts_to_sequences(raw_valid['NARRATIVE'])
X_test = tokenizer.texts_to_sequences(raw_test['NARRATIVE'])
X_train = pad_sequences(X_train, maxlen=max_len)
X_valid = pad_sequences(X_valid, maxlen=max_len)
X_test = pad_sequences(X_test, maxlen=max_len)
print('X_train shape:', X_train.shape)
print('X_valid shape:', X_valid.shape)
print('X_test shape:', X_test.shape)
vocab_size = len(tokenizer.word_index)
print 'specifying model'
input_layer = Input(shape=(max_len, vocab_size), dtype='float32') # [n_samples, n_steps, n_features]
lstm_left = LSTM(output_dim=10, dropout_W=0.5, dropout_U=0.5,
return_sequences=True, consume_less='mem')(input_layer) # [n_samples, n_steps, n_units]
lstm_right = LSTM(output_dim=10, dropout_W=0.5, dropout_U=0.5,
return_sequences=True, consume_less='mem', go_backwards=True)(input_layer)
merged_layer = merge([lstm_left, lstm_right], mode='concat', concat_axis=2) # [n_samples, ]
pooling = MaxPooling1D(pool_length=max_len)(merged_layer)
flatten = Flatten()(pooling)
batch_norm = BatchNormalization()(flatten)
dropout = Dropout(0.5)(flatten)
output = Dense(output_dim=nb_classes, activation='softmax')(dropout)
print 'instantiate model'
model = Model(input=input_layer, output=output)
print 'compiling model'
model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])
checkpointer = keras.callbacks.ModelCheckpoint(filepath=os.path.join(checkpoint_dir, 'weights.{epoch:02d}--{val_acc:.3f}.hdf5'),
monitor='val_acc',
verbose=1,
save_best_only=False)
train_generator = generate_one_hot(X_train, Y_train, vocab_size, batch_size=gen_batch_size)
valid_generator = generate_one_hot(X_valid, Y_valid, vocab_size, batch_size=gen_batch_size)
print 'fitting model'
model.fit_generator(generator=train_generator,
samples_per_epoch=len(X_train),
nb_epoch=60,
validation_data=valid_generator,
nb_val_samples=len(X_valid))
test_generator = generate_one_hot(X_test, Y_test, vocab_size, batch_size=gen_batch_size)
score = model.evaluate_generator(test_generator, len(X_test), verbose=0)
print('Test score:', score[0])
print('Test accuracy:', score[1])