Fix word embeddings example.

examples/lstm_seq2seq.py

@@ -59,7 +59,7 @@ epochs = 100  # Number of epochs to train for.
 latent_dim = 256  # Latent dimensionality of the encoding space.
 num_samples = 10000  # Number of samples to train on.
 # Path to the data txt file on disk.
-data_path = '/Users/fchollet/Downloads/fra-eng/fra.txt'
+data_path = 'fra-eng/fra.txt'
 
 # Vectorize the data.
 input_texts = []

examples/pretrained_word_embeddings.py

@@ -19,14 +19,14 @@ import numpy as np
 from keras.preprocessing.text import Tokenizer
 from keras.preprocessing.sequence import pad_sequences
 from keras.utils import to_categorical
-from keras.layers import Dense, Input, Flatten
+from keras.layers import Dense, Input, GlobalMaxPooling1D
 from keras.layers import Conv1D, MaxPooling1D, Embedding
 from keras.models import Model
 
 
 BASE_DIR = ''
-GLOVE_DIR = BASE_DIR + '/glove.6B/'
-TEXT_DATA_DIR = BASE_DIR + '/20_newsgroup/'
+GLOVE_DIR = os.path.join(BASE_DIR, 'glove.6B')
+TEXT_DATA_DIR = os.path.join(BASE_DIR, '20_newsgroup')
 MAX_SEQUENCE_LENGTH = 1000
 MAX_NB_WORDS = 20000
 EMBEDDING_DIM = 100
@@ -106,7 +106,7 @@ print('Preparing embedding matrix.')
 
 # prepare embedding matrix
 num_words = min(MAX_NB_WORDS, len(word_index))
-embedding_matrix = np.zeros((num_words + 1, EMBEDDING_DIM))
+embedding_matrix = np.zeros((num_words, EMBEDDING_DIM))
 for word, i in word_index.items():
     if i >= MAX_NB_WORDS:
         continue
@@ -133,8 +133,7 @@ x = MaxPooling1D(5)(x)
 x = Conv1D(128, 5, activation='relu')(x)
 x = MaxPooling1D(5)(x)
 x = Conv1D(128, 5, activation='relu')(x)
-x = MaxPooling1D(35)(x)
-x = Flatten()(x)
+x = GlobalMaxPooling1D()(x)
 x = Dense(128, activation='relu')(x)
 preds = Dense(len(labels_index), activation='softmax')(x)