关于model.py中class Model_DNN的一个问题
codefish1990 opened this issue · comments
codefish1990 commented
Model_DNN实现中:
masks = tf.concat([tf.expand_dims(self.mask, -1) for _ in range(hidden_size)], axis=-1)是否应该改成
masks = tf.concat([tf.expand_dims(self.mask, -1) for _ in range(embedding_dim)], axis=-1)?
cenyk1230 commented
Thank you for pointing it out. It really should be embedding_dim
instead of hidden_size
. In our experiments, we set the same value for these two hyper-parameters, and thus I didn't find this mistake before.