Where should the LSTM be placed in my CNN for NLP and how do I connect it?-CodePudding

I have a problem. I would like to use LSTM in my 1D-CNN to get an improvement in my NLP task. The problem is that I don't know exactly where to put the LSTM. I have found the following.

A CNN LSTM can be defined by adding CNN layers on the front end followed by LSTM layers with a Dense layer on the output.

(Source: https://machinelearningmastery.com/cnn-long-short-term-memory-networks/)

However, if I set it up like this (see code below), I get the following error

ValueError: Input 0 of layer "lstm_4" is incompatible with the layer: expected ndim=3, found ndim=2. Full shape received: (None, 128)

This is because LSTM expects a 3D input array. Is there an option to fix this error and use LSTM at this position? Or should it be somewhere else?

from keras.models import Sequential
from keras.layers import Input, Embedding, Dense, GlobalMaxPooling1D, Conv2D, MaxPool2D, LSTM, Bidirectional, Lambda, Conv1D, MaxPooling1D, GlobalMaxPooling1D

model_lstm = Sequential()

model_lstm.add(
        Embedding(vocab_size
                ,embed_size
                ,weights = [embedding_matrix] #Supplied embedding matrix created from glove
                ,input_length = maxlen
                ,trainable=False)
         )
model_lstm.add(SpatialDropout1D(rate = 0.4))
model_lstm.add(Conv1D(256, 7, activation="relu"))
model_lstm.add(MaxPooling1D())
#model_lstm.add(LSTM(128, dropout=0.3, recurrent_dropout=0.3, return_sequences=True))
model_lstm.add(Conv1D(128, 5, activation="relu"))
model_lstm.add(MaxPooling1D())
model_lstm.add(GlobalMaxPooling1D())
model_lstm.add(LSTM(128, dropout=0.3,return_sequences=True))
model_lstm.add(Dropout(0.3))
model_lstm.add(Dense(128, activation="relu"))
model_lstm.add(Dense(4, activation='softmax'))
print(model_lstm.summary())

Complete Code

print("Train shape : ",train_X2.shape)
print("Test shape : ",test_X2.shape)

## Tokenize the sentences
tokenizer = Tokenizer(num_words=num_unique_words)
tokenizer.fit_on_texts(list(train_X2))
train_X2 = tokenizer.texts_to_sequences(train_X2)
test_X2 = tokenizer.texts_to_sequences(test_X2)

## Pad the sentences 
train_X = pad_sequences(train_X2, maxlen=maxlen)
test_X = pad_sequences(test_X2, maxlen=maxlen)

word_index = tokenizer.word_index
vocab_size = len(tokenizer.word_index)   1

from sklearn.preprocessing import LabelEncoder
from tensorflow.keras.utils import to_categorical

#label encoding
le = LabelEncoder()
train_y = le.fit_transform(train_y2.tolist())
test_y = le.transform(test_y2.tolist())

#one hot encoding
train_y = to_categorical(train_y)
test_y = to_categorical(test_y)

# Word2Vec as pretrained embedding
import gensim
from gensim.models import Word2Vec
from gensim.utils import simple_preprocess

from gensim.models.keyedvectors import KeyedVectors
NUM_WORDS=20000
word_vectors = KeyedVectors.load_word2vec_format(r'./input/GoogleNews-vectors-negative300.bin', binary=True)

EMBEDDING_DIM=300
vocabulary_size=min(len(word_index) 1,NUM_WORDS)
embedding_matrix = np.zeros((vocabulary_size, EMBEDDING_DIM))
for word, i in word_index.items():
    if i>=NUM_WORDS:
        continue
    try:
        embedding_vector = word_vectors[word]
        embedding_matrix[i] = embedding_vector
    except KeyError:
        embedding_matrix[i]=np.random.normal(0,np.sqrt(0.25),EMBEDDING_DIM)

del(word_vectors)

from keras.layers import Embedding
embedding_layer = Embedding(vocabulary_size,
                            EMBEDDING_DIM,
                            weights=[embedding_matrix],
                            trainable=True)

from keras.layers import Embedding
EMBEDDING_DIM=300
vocabulary_size=min(len(word_index) 1,NUM_WORDS)

embedding_layer = Embedding(vocabulary_size,
                            EMBEDDING_DIM)

# CNN

CodePudding user response：

Maybe try removing the GlobalMaxPooling1D layer which reduces your tensor to 2D. For example try copy and run this:

from keras.models import Sequential
from keras.layers import Input, Embedding, Dense, GlobalMaxPooling1D, Conv2D, MaxPool2D, LSTM, Bidirectional, Lambda, Conv1D, MaxPooling1D, GlobalMaxPooling1D

model_lstm = Sequential()

model_lstm.add(
        Embedding(vocab_size
                ,embed_size
                ,weights = [embedding_matrix] #Supplied embedding matrix created from glove
                ,input_length = maxlen
                ,trainable=False)
         )
model_lstm.add(SpatialDropout1D(rate = 0.4))
model_lstm.add(Conv1D(256, 7, activation="relu"))
model_lstm.add(MaxPooling1D())
#model_lstm.add(LSTM(128, dropout=0.3, recurrent_dropout=0.3, return_sequences=True))
model_lstm.add(Conv1D(128, 5, activation="relu"))
model_lstm.add(MaxPooling1D())
model_lstm.add(LSTM(128, dropout=0.3, return_sequences=False))
model_lstm.add(Dropout(0.3))
model_lstm.add(Dense(128, activation="relu"))
model_lstm.add(Dense(4, activation='softmax'))
print(model_lstm.summary())