GRU in Keras

In [1]:

    

import keras
from keras.datasets import imdb
from keras.preprocessing.sequence import pad_sequences
from keras.models import Sequential
from keras.layers import Dense, Dropout, Embedding, SpatialDropout1D
from keras.layers import GRU # new! 
from keras.callbacks import ModelCheckpoint
import os
from sklearn.metrics import roc_auc_score 
import matplotlib.pyplot as plt 
%matplotlib inline


    

    




Using TensorFlow backend.

In [2]:

    

# output directory name:
output_dir = 'model_output/gru'

# training:
epochs = 4
batch_size = 128

# vector-space embedding: 
n_dim = 64 
n_unique_words = 10000 
max_review_length = 100 # lowered due to vanishing gradient over time
pad_type = trunc_type = 'pre'
drop_embed = 0.2 

# GRU layer architecture:
n_gru = 256 
drop_gru = 0.2

# dense layer architecture: 
# n_dense = 256
# dropout = 0.2


    

In [3]:

    

(x_train, y_train), (x_valid, y_valid) = imdb.load_data(num_words=n_unique_words) # removed n_words_to_skip


    

In [4]:

    

x_train = pad_sequences(x_train, maxlen=max_review_length, padding=pad_type, truncating=trunc_type, value=0)
x_valid = pad_sequences(x_valid, maxlen=max_review_length, padding=pad_type, truncating=trunc_type, value=0)


    

In [5]:

    

model = Sequential()
model.add(Embedding(n_unique_words, n_dim, input_length=max_review_length)) 
model.add(SpatialDropout1D(drop_embed))
model.add(GRU(n_gru, dropout=drop_gru))
# model.add(Dense(n_dense, activation='relu')) # typically don't see top dense layer in NLP like in 
# model.add(Dropout(dropout))
model.add(Dense(1, activation='sigmoid'))


    

In [6]:

    

model.summary()


    

    




_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
embedding_1 (Embedding)      (None, 100, 64)           640000    
_________________________________________________________________
spatial_dropout1d_1 (Spatial (None, 100, 64)           0         
_________________________________________________________________
gru_1 (GRU)                  (None, 256)               246528    
_________________________________________________________________
dense_1 (Dense)              (None, 1)                 257       
=================================================================
Total params: 886,785
Trainable params: 886,785
Non-trainable params: 0
_________________________________________________________________

In [7]:

    

model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])


    

In [8]:

    

modelcheckpoint = ModelCheckpoint(filepath=output_dir+"/weights.{epoch:02d}.hdf5")
if not os.path.exists(output_dir):
    os.makedirs(output_dir)


    

In [9]:

    

# XXX% validation accuracy in epoch X
model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs, verbose=1, validation_data=(x_valid, y_valid), callbacks=[modelcheckpoint])


    

    




Train on 25000 samples, validate on 25000 samples
Epoch 1/4
25000/25000 [==============================] - 31s - loss: 0.5238 - acc: 0.7241 - val_loss: 0.3771 - val_acc: 0.8350
Epoch 2/4
25000/25000 [==============================] - 29s - loss: 0.3062 - acc: 0.8731 - val_loss: 0.3390 - val_acc: 0.8504
Epoch 3/4
25000/25000 [==============================] - 29s - loss: 0.2484 - acc: 0.9004 - val_loss: 0.3680 - val_acc: 0.8452
Epoch 4/4
25000/25000 [==============================] - 28s - loss: 0.2123 - acc: 0.9176 - val_loss: 0.3837 - val_acc: 0.8396






    
Out[9]:





<keras.callbacks.History at 0x7fd996b5e438>

In [10]:

    

model.load_weights(output_dir+"/weights.01.hdf5") # zero-indexed


    

In [11]:

    

y_hat = model.predict_proba(x_valid)


    

    




25000/25000 [==============================] - 24s    

In [12]:

    

plt.hist(y_hat)
_ = plt.axvline(x=0.5, color='orange')


    

In [13]:

    

"{:0.2f}".format(roc_auc_score(y_valid, y_hat)*100.0)


    

    
Out[13]:





'92.98'

In [ ]: