Created
April 25, 2019 00:07
-
-
Save ngopal/ca7f139157136861d618ee0e77e6f2aa to your computer and use it in GitHub Desktop.
Loading pre-trained vectors into keras models
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# The first step is to load the pre-trained vectors into python. The example below uses glove data. | |
import os | |
GLOVE_DIR = "/path/to/pretrained/embeddings/glove.6B/" | |
embeddings_index = {} | |
f = open(os.path.join(GLOVE_DIR, 'glove.6B.100d.txt'), "r") | |
for line in f: | |
values = line.split() | |
word = values[0] | |
coefs = np.asarray(values[1:], dtype='float32') | |
embeddings_index[word] = coefs | |
f.close() | |
EMBEDDING_DIM = 100 | |
embedding_matrix = np.random.random((len(word_index) + 1, EMBEDDING_DIM)) | |
for word, i in word_index.items(): | |
embedding_vector = embeddings_index.get(word) | |
if embedding_vector is not None: | |
# words not found in embedding index will be all-zeros. | |
embedding_matrix[i] = embedding_vector | |
# The second step is to load this data into a keras embedding layer | |
from keras.layers import Embedding | |
MAX_SEQUENCE_LENGTH = 1000 | |
embedding_layer = Embedding(len(word_index) + 1, | |
EMBEDDING_DIM, | |
weights=[embedding_matrix], | |
input_length=MAX_SEQUENCE_LENGTH, | |
trainable=False) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment