python - Tensorflow tf.constant_initializer 非常慢

Question

尝试使用 100 dim 的预训练 word2vec 嵌入来训练 LSTM

@staticmethod
def load_embeddings(pre_trained_embeddings_path, word_embed_size):
    embd = []
    import time
    start_time = time.time()
    cnt = 4
    with codecs.open(pre_trained_embeddings_path, mode="r", encoding='utf-8') as f:
        for line in f.readlines():
            values = line.strip().split(' ')
            embd.append(values[1:])
            cnt += 1
            if cnt % 100000 == 0:
                print("word-vectors loaded: %d" % cnt)

    embedding, vocab_size, embed_dim = embd, len(embd), len(embd[0])

    load_end_time = time.time()
    print("word vectors loaded from and start initialising, cnt: %d, time taken: %d secs " % (vocab_size, load_end_time - start_time))

    embedding_init = tf.constant_initializer(embedding, dtype=tf.float16)
    src_word_embedding = tf.get_variable(shape=[vocab_size, embed_dim], initializer=embedding_init, trainable=False, name='word_embedding', dtype=tf.float16)

    print("word-vectors loaded and initialised, cnt: %d, time taken: %d secs" % (vocab_size, time.time() - load_end_time))

    return src_word_embedding

运行此方法时的输出如下:

word vectors loaded from and start initialising, cnt: 2419080, time taken: 74 secs
word-vectors loaded and initialised, cnt: 2419080, time taken: 1647 secs

系统信息:tensorflow 1.1.0，tcmalloc，python 3.6，ubuntu 14.04

半小时初始化似乎很慢还是正常行为？知道可能是什么问题吗？

更新:使用@sirfz 方法提供嵌入使得加载嵌入变得非常快初始化在 85 秒内完成

Answer 1

将大常量加载到图形中不仅速度较慢，而且还会泄漏大量内存。我有一个类似的问题 I reported not long ago对我来说最好的解决方法是:

# placeholder for loading your saved embeddings
embedding_init = tf.placeholder(tf.float16, shape=[vocab_size, embed_dim])
src_word_embedding = tf.get_variable(initializer=embedding_init, trainable=False, name='word_embedding', dtype=tf.float16)

# run initialization with the value of embeddings placeholder
session.run(tf.global_variables_initializer(), feed_dict={embedding_init: embedding})