将字嵌入层添加到CNN模型时的值错误

Question

我正在尝试使用CNN向着名的文本分类架构添加FastText嵌入层：https://github.com/dennybritz/cnn-text-classification-tf

我加载了我的FastText嵌入：

embedding = np.load('embedding/my_embedding_file')
# type(embedding): gensim.models.fasttext.FastText
vocab = embedding.wv.vocab
vocab_processor.fit(vocab)
x_train = np.array(list(vocab_processor.transform(x_train)))
x_dev = np.array(list(vocab_processor.transform(x_dev)))
vocabulary_size = len(vocab) #300k

以下是我在Tensorflow中添加嵌入图层的方法：

# Embedding layer
with tf.device('/cpu:0'), tf.name_scope("embedding"):
    embedding_weights_ = tf.Variable(tf.constant(0.0, shape=[vocab_size, embedding_size]), trainable=False, name="embedding_weights")
    embedding_weights = tf.assign(embedding_weights_, self.embedding_placeholder)

    self.embedded_chars = tf.nn.embedding_lookup(embedding_weights, self.input_x)
    self.embedded_chars_expanded = tf.expand_dims(self.embedded_chars, -1) 
# Create a convolution + maxpool layer for each filter size
pooled_outputs = []
for i, filter_size in enumerate(filter_sizes):
    with tf.name_scope("conv-maxpool-%s" % filter_size):
        # Convolution Layer
        filter_shape = [filter_size, embedding_size, 1, num_filters]
        W = tf.Variable(tf.truncated_normal(filter_shape, stddev=0.1), name="W")
        b = tf.Variable(tf.constant(0.1, shape=[num_filters]), name="b")
        conv = tf.nn.conv2d(
            self.embedded_chars_expanded, #my embedding to the next layer
            W,
            strides=[1, 1, 1, 1],
            padding="VALID",
            name="conv")

CNN模型的初始化：

with tf.Graph().as_default():
    session_conf = tf.ConfigProto(
      allow_soft_placement=FLAGS.allow_soft_placement,
      log_device_placement=FLAGS.log_device_placement)
    sess = tf.Session(config=session_conf)
    with sess.as_default():
        cnn = TextCNN(
            sentence_length = max_document_length,
            num_classes=17,
            vocab_size=len(vocab_processor.vocabulary_),
            embedding_size=FLAGS.embedding_dim,
            filter_sizes=list(map(int, FLAGS.filter_sizes.split(","))),
            num_filters=FLAGS.num_filters,
            l2_reg_lambda=FLAGS.l2_reg_lambda)

单一训练步骤：

def train_step(x_batch, y_batch):
            feed_dict = {
              cnn.input_x: x_batch,
              cnn.input_y: y_batch,
              cnn.dropout_keep_prob: FLAGS.dropout_keep_prob,
              cnn.embedding_placeholder:embedding
            }
            _, step, summaries, loss, accuracy = sess.run(
                [train_op, global_step, train_summary_op, cnn.loss, cnn.accuracy],
                feed_dict)
            time_str = datetime.datetime.now().isoformat()
            print("{}: step {}, loss {:g}, acc {:g}".format(time_str, step, loss, accuracy))
            train_summary_writer.add_summary(summaries, step)

作为参考，下面是我的x_train和y_train的样子：

当我开始训练时，我收到此错误：

  File "<ipython-input-24-2ec193e9123f>", line 122, in <module>
    train_step(x_batch, y_batch)

  File "<ipython-input-24-2ec193e9123f>", line 92, in train_step
    feed_dict)

  File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 889, in run
    run_metadata_ptr)

  File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 1089, in _run
    np_val = np.asarray(subfeed_val, dtype=subfeed_dtype)

  File "C:\ProgramData\Anaconda3\lib\site-packages\numpy\core\numeric.py", line 492, in asarray
    return array(a, dtype, copy=False, order=order)

ValueError: setting an array element with a sequence.

这个错误看起来像一个numpy数组创建错误，它尝试创建一个数组元素作为序列。 但： 当我移除嵌入时我没有收到任何错误，而我的列车数据完全相同，因此批次也是如此。

什么可能导致此错误？