fasttext和cnn的比较，使用keras imdb看效果——cnn要慢10倍。
fasttext:
'''This example demonstrates the use of fasttext for text classification
Based on Joulin et al's paper:
Bags of Tricks for Efficient Text Classification
https://arxiv.org/abs/1607.01759
Results on IMDB datasets with uni and bi-gram embeddings:
Uni-gram: 0.8813 test accuracy after 5 epochs. 8s/epoch on i7 cpu.
Bi-gram : 0.9056 test accuracy after 5 epochs. 2s/epoch on GTx 980M gpu.
'''
from __future__ import print_function
import numpy as np
from keras.preprocessing import sequence
from keras.models import Sequential
from keras.layers import Dense
from keras.layers import Embedding
from keras.layers import GlobalAveragePooling1D
from keras.datasets import imdb
import numpy as np
import json
import warnings
def load_data(path='imdb.npz', num_words=None, skip_top=0,
maxlen=None, seed=113,
start_char=1, oov_char=2, index_from=3, **kwargs):
"""Loads the IMDB dataset.
# Arguments
path: where to cache the data (relative to `~/.keras/dataset`).
num_words: max number of words to include. Words are ranked
by how often they occur (in the training set) and only
the most frequent words are kept
skip_top: skip the top N most frequently occurring words
(which may not be informative).
maxlen: sequences longer than this will be filtered out.
seed: random seed for sample shuffling.
start_char: The start of a sequence will be marked with this character.
Set to 1 because 0 is usually the padding character.
oov_char: words that were cut out because of the `num_words`
or `skip_top` limit will be replaced with this character.
index_from: index actual words with this index and higher.
# Returns
Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`.
# Raises
ValueError: in case `maxlen` is so low
that no input sequence could be kept.
Note that the 'out of vocabulary' character is only used for
words that were present in the training set but are not included
because they're not making the `num_words` cut here.
Words that were not seen in the training set but are in the test set
have simply been skipped.
"""
# Legacy support
if 'nb_words' in kwargs:
warnings.warn('The `nb_words` argument in `load_data` '
'has been renamed `num_words`.')
num_words = kwargs.pop('nb_words')
if kwargs:
raise TypeError('Unrecognized keyword arguments: ' + str(kwargs))
#path = get_file(path,
# origin='https://s3.amazonaws.com/text-datasets/imdb.npz',
# file_hash='599dadb1135973df5b59232a0e9a887c')
with np.load(path) as f:
x_train, labels_train = f['x_train'], f['y_train']
x_test, labels_test = f['x_test'], f['y_test']
np.random.seed(seed)
indices = np.arange(len(x_train))
np.random.shuffle(indices)
x_train = x_train[indices]
labels_train = labels_train[indices]
indices = np.arange(len(x_test))
np.random.shuffle(indices)
x_test = x_test[indices]
labels_test = labels_test[indices]
xs = np.concatenate([x_train, x_test])
labels = np.concatenate([labels_train, labels_test])
if start_char is not None:
xs = [[start_char] + [w + index_from for w in x] for x in xs]
elif index_from:
xs = [[w + index_from for w in x] for x in xs]
if maxlen:
xs, labels = _remove_long_seq(maxlen, xs, labels)
if not xs:
raise ValueError('After filtering for sequences shorter than maxlen=' +
str(maxlen) + ', no sequence was kept. '
'Increase maxlen.')
if not num_words:
num_words = max([max(x) for x in xs])
# by convention, use 2 as OOV word
# reserve 'index_from' (=3 by default) characters:
# 0 (padding), 1 (start), 2 (OOV)
if oov_char is not None:
xs = [[w if (skip_top <= w < num_words) else oov_char for w in x] for x in xs]
else:
xs = [[w for w in x if skip_top <= w < num_words] for x in xs]
idx = len(x_train)
x_train, y_train = np.array(xs[:idx]), np.array(labels[:idx])
x_test, y_test = np.array(xs[idx:]), np.array(labels[idx:])
return (x_train, y_train), (x_test, y_test)
def create_ngram_set(input_list, ngram_value=2):
"""
Extract a set of n-grams from a list of integers.
>>> create_ngram_set([1, 4, 9, 4, 1, 4], ngram_value=2)
{(4, 9), (4, 1), (1, 4), (9, 4)}
>>> create_ngram_set([1, 4, 9, 4, 1, 4], ngram_value=3)
[(1, 4, 9), (4, 9, 4), (9, 4, 1), (4, 1, 4)]
"""
return set(zip(*[input_list[i:] for i in range(ngram_value)]))
def add_ngram(sequences, token_indice, ngram_range=2):
"""
Augment the input list of list (sequences) by appending n-grams values.
Example: adding bi-gram
>>> sequences = [[1, 3, 4, 5], [1, 3, 7, 9, 2]]
>>> token_indice = {(1, 3): 1337, (9, 2): 42, (4, 5): 2017}
>>> add_ngram(sequences, token_indice, ngram_range=2)
[[1, 3, 4, 5, 1337, 2017], [1, 3, 7, 9, 2, 1337, 42]]
Example: adding tri-gram
>>> sequences = [[1, 3, 4, 5], [1, 3, 7, 9, 2]]
>>> token_indice = {(1, 3): 1337, (9, 2): 42, (4, 5): 2017, (7, 9, 2): 2018}
>>> add_ngram(sequences, token_indice, ngram_range=3)
[[1, 3, 4, 5, 1337, 2017], [1, 3, 7, 9, 2, 1337, 42, 2018]]
"""
new_sequences = []
for input_list in sequences:
new_list = input_list[:]
for ngram_value in range(2, ngram_range + 1):
for i in range(len(new_list) - ngram_value + 1):
ngram = tuple(new_list[i:i + ngram_value])
if ngram in token_indice:
new_list.append(token_indice[ngram])
new_sequences.append(new_list)
return new_sequences
# Set parameters:
# ngram_range = 2 will add bi-grams features
ngram_range = 1
max_features = 20000
maxlen = 400
batch_size = 32
embedding_dims = 50
epochs = 5
print('Loading data...')
# the data, split between train and test sets
#(x_train, y_train), (x_test, y_test) = load_data()
(x_train, y_train), (x_test, y_test) = load_data(num_words=max_features)
print(len(x_train), 'train sequences')
print(len(x_test), 'test sequences')
print('Average train sequence length: {}'.format(np.mean(list(map(len, x_train)), dtype=int)))
print('Average test sequence length: {}'.format(np.mean(list(map(len, x_test)), dtype=int)))
if ngram_range > 1:
print('Adding {}-gram features'.format(ngram_range))
# Create set of unique n-gram from the training set.
ngram_set = set()
for input_list in x_train:
for i in range(2, ngram_range + 1):
set_of_ngram = create_ngram_set(input_list, ngram_value=i)
ngram_set.update(set_of_ngram)
# Dictionary mapping n-gram token to a unique integer.
# Integer values are greater than max_features in order
# to avoid collision with existing features.
start_index = max_features + 1
token_indice = {v: k + start_index for k, v in enumerate(ngram_set)}
indice_token = {token_indice[k]: k for k in token_indice}
# max_features is the highest integer that could be found in the dataset.
max_features = np.max(list(indice_token.keys())) + 1
# Augmenting x_train and x_test with n-grams features
x_train = add_ngram(x_train, token_indice, ngram_range)
x_test = add_ngram(x_test, token_indice, ngram_range)
print('Average train sequence length: {}'.format(np.mean(list(map(len, x_train)), dtype=int)))
print('Average test sequence length: {}'.format(np.mean(list(map(len, x_test)), dtype=int)))
print('Pad sequences (samples x time)')
x_train = sequence.pad_sequences(x_train, maxlen=maxlen)
x_test = sequence.pad_sequences(x_test, maxlen=maxlen)
print('x_train shape:', x_train.shape)
print('x_test shape:', x_test.shape)
print('Build model...')
model = Sequential()
# we start off with an efficient embedding layer which maps
# our vocab indices into embedding_dims dimensions
model.add(Embedding(max_features,
embedding_dims,
input_length=maxlen))
# we add a GlobalAveragePooling1D, which will average the embeddings
# of all words in the document
model.add(GlobalAveragePooling1D())
# We project onto a single unit output layer, and squash it with a sigmoid:
model.add(Dense(1, activation='sigmoid'))
model.compile(loss='binary_crossentropy',
optimizer='adam',
metrics=['accuracy'])
model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs, validation_data=(x_test, y_test))
效果：
Train on 25000 samples, validate on 25000 samples Epoch 1/50 2018-06-06 15:50:28.133461: I tensorflow/core/platform/cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA 25000/25000 [==============================] - 9s 379us/step - loss: 0.6125 - acc: 0.7431 - val_loss: 0.5050 - val_acc: 0.8227 Epoch 2/50 25000/25000 [==============================] - 10s 402us/step - loss: 0.4059 - acc: 0.8633 - val_loss: 0.3738 - val_acc: 0.8646 Epoch 3/50 25000/25000 [==============================] - 11s 441us/step - loss: 0.3061 - acc: 0.8934 - val_loss: 0.3219 - val_acc: 0.8783 Epoch 4/50 25000/25000 [==============================] - 9s 375us/step - loss: 0.2550 - acc: 0.9110 - val_loss: 0.2970 - val_acc: 0.8853 Epoch 5/50
可以看到一个epoch只需要10来秒，还是很快的！但是我训练到50个epoch后发现acc 100%，但是验证集上数据acc 86%，看来是过拟合了。
再看看传统cnn：
'''This example demonstrates the use of Convolution1D for text classification.
Gets to 0.89 test accuracy after 2 epochs.
90s/epoch on Intel i5 2.4Ghz CPU.
10s/epoch on Tesla K40 GPU.
'''
from __future__ import print_function
from keras.preprocessing import sequence
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation
from keras.layers import Embedding
from keras.layers import Conv1D, GlobalMaxPooling1D
from keras.datasets import imdb
# set parameters:
max_features = 5000
import numpy as np
import json
import warnings
def load_data(path='imdb.npz', num_words=None, skip_top=0,
maxlen=None, seed=113,
start_char=1, oov_char=2, index_from=3, **kwargs):
"""Loads the IMDB dataset.
# Arguments
path: where to cache the data (relative to `~/.keras/dataset`).
num_words: max number of words to include. Words are ranked
by how often they occur (in the training set) and only
the most frequent words are kept
skip_top: skip the top N most frequently occurring words
(which may not be informative).
maxlen: sequences longer than this will be filtered out.
seed: random seed for sample shuffling.
start_char: The start of a sequence will be marked with this character.
Set to 1 because 0 is usually the padding character.
oov_char: words that were cut out because of the `num_words`
or `skip_top` limit will be replaced with this character.
index_from: index actual words with this index and higher.
# Returns
Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`.
# Raises
ValueError: in case `maxlen` is so low
that no input sequence could be kept.
Note that the 'out of vocabulary' character is only used for
words that were present in the training set but are not included
because they're not making the `num_words` cut here.
Words that were not seen in the training set but are in the test set
have simply been skipped.
"""
# Legacy support
if 'nb_words' in kwargs:
warnings.warn('The `nb_words` argument in `load_data` '
'has been renamed `num_words`.')
num_words = kwargs.pop('nb_words')
if kwargs:
raise TypeError('Unrecognized keyword arguments: ' + str(kwargs))
#path = get_file(path,
# origin='https://s3.amazonaws.com/text-datasets/imdb.npz',
# file_hash='599dadb1135973df5b59232a0e9a887c')
with np.load(path) as f:
x_train, labels_train = f['x_train'], f['y_train']
x_test, labels_test = f['x_test'], f['y_test']
np.random.seed(seed)
indices = np.arange(len(x_train))
np.random.shuffle(indices)
x_train = x_train[indices]
labels_train = labels_train[indices]
indices = np.arange(len(x_test))
np.random.shuffle(indices)
x_test = x_test[indices]
labels_test = labels_test[indices]
xs = np.concatenate([x_train, x_test])
labels = np.concatenate([labels_train, labels_test])
if start_char is not None:
xs = [[start_char] + [w + index_from for w in x] for x in xs]
elif index_from:
xs = [[w + index_from for w in x] for x in xs]
if maxlen:
xs, labels = _remove_long_seq(maxlen, xs, labels)
if not xs:
raise ValueError('After filtering for sequences shorter than maxlen=' +
str(maxlen) + ', no sequence was kept. '
'Increase maxlen.')
if not num_words:
num_words = max([max(x) for x in xs])
# by convention, use 2 as OOV word
# reserve 'index_from' (=3 by default) characters:
# 0 (padding), 1 (start), 2 (OOV)
if oov_char is not None:
xs = [[w if (skip_top <= w < num_words) else oov_char for w in x] for x in xs]
else:
xs = [[w for w in x if skip_top <= w < num_words] for x in xs]
idx = len(x_train)
x_train, y_train = np.array(xs[:idx]), np.array(labels[:idx])
x_test, y_test = np.array(xs[idx:]), np.array(labels[idx:])
return (x_train, y_train), (x_test, y_test)
def create_ngram_set(input_list, ngram_value=2):
"""
Extract a set of n-grams from a list of integers.
>>> create_ngram_set([1, 4, 9, 4, 1, 4], ngram_value=2)
{(4, 9), (4, 1), (1, 4), (9, 4)}
>>> create_ngram_set([1, 4, 9, 4, 1, 4], ngram_value=3)
[(1, 4, 9), (4, 9, 4), (9, 4, 1), (4, 1, 4)]
"""
return set(zip(*[input_list[i:] for i in range(ngram_value)]))
def add_ngram(sequences, token_indice, ngram_range=2):
"""
Augment the input list of list (sequences) by appending n-grams values.
Example: adding bi-gram
>>> sequences = [[1, 3, 4, 5], [1, 3, 7, 9, 2]]
>>> token_indice = {(1, 3): 1337, (9, 2): 42, (4, 5): 2017}
>>> add_ngram(sequences, token_indice, ngram_range=2)
[[1, 3, 4, 5, 1337, 2017], [1, 3, 7, 9, 2, 1337, 42]]
Example: adding tri-gram
>>> sequences = [[1, 3, 4, 5], [1, 3, 7, 9, 2]]
>>> token_indice = {(1, 3): 1337, (9, 2): 42, (4, 5): 2017, (7, 9, 2): 2018}
>>> add_ngram(sequences, token_indice, ngram_range=3)
[[1, 3, 4, 5, 1337, 2017], [1, 3, 7, 9, 2, 1337, 42, 2018]]
"""
new_sequences = []
for input_list in sequences:
new_list = input_list[:]
for ngram_value in range(2, ngram_range + 1):
for i in range(len(new_list) - ngram_value + 1):
ngram = tuple(new_list[i:i + ngram_value])
if ngram in token_indice:
new_list.append(token_indice[ngram])
new_sequences.append(new_list)
return new_sequences
maxlen = 400
batch_size = 32
embedding_dims = 50
filters = 250
kernel_size = 3
hidden_dims = 250
epochs = 5
print('Loading data...')
(x_train, y_train), (x_test, y_test) = load_data(num_words=max_features)
print(len(x_train), 'train sequences')
print(len(x_test), 'test sequences')
print('Pad sequences (samples x time)')
x_train = sequence.pad_sequences(x_train, maxlen=maxlen)
x_test = sequence.pad_sequences(x_test, maxlen=maxlen)
print('x_train shape:', x_train.shape)
print('x_test shape:', x_test.shape)
print('Build model...')
model = Sequential()
# we start off with an efficient embedding layer which maps
# our vocab indices into embedding_dims dimensions
model.add(Embedding(max_features,
embedding_dims,
input_length=maxlen))
model.add(Dropout(0.2))
# we add a Convolution1D, which will learn filters
# word group filters of size filter_length:
model.add(Conv1D(filters,
kernel_size,
padding='valid',
activation='relu',
strides=1))
# we use max pooling:
model.add(GlobalMaxPooling1D())
# We add a vanilla hidden layer:
model.add(Dense(hidden_dims))
model.add(Dropout(0.2))
model.add(Activation('relu'))
# We project onto a single unit output layer, and squash it with a sigmoid:
model.add(Dense(1))
model.add(Activation('sigmoid'))
model.compile(loss='binary_crossentropy',
optimizer='adam',
metrics=['accuracy'])
model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs,
validation_data=(x_test, y_test))
效果：
Train on 25000 samples, validate on 25000 samples
Epoch 1/5
2018-06-06 16:10:34.733973: I tensorflow/core/platform/cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
25000/25000 [==============================] - 117s 5ms/step - loss: 0.4044 - acc: 0.8007 - val_loss: 0.3212 - val_acc: 0.8600
Epoch 2/5
25000/25000 [==============================] - 121s 5ms/step - loss: 0.2323 - acc: 0.9057 - val_loss: 0.2903 - val_acc: 0.8801
Epoch 3/5
25000/25000 [==============================] - 124s 5ms/step - loss: 0.1640 - acc: 0.9377 - val_loss: 0.2720 - val_acc: 0.8900
Epoch 4/5
25000/25000 [==============================] - 116s 5ms/step - loss: 0.1136 - acc: 0.9579 - val_loss: 0.3353 - val_acc: 0.8811
Epoch 5/5
25000/25000 [==============================] - 114s 5ms/step - loss: 0.0764 - acc: 0.9726 - val_loss: 0.3958 - val_acc: 0.8793
可以看出cnn的确要慢10倍。
