4287aef6a3a82436b4e3e156b22ede235eb4e6ba,texar/modules/decoders/transformer_decoders.py,TransformerDecoder,__init__,#TransformerDecoder#Any#Any#Any#,28

Before Change

        ModuleBase.__init__(self, hparams)
        self._vocab_size = vocab_size
        self._embedding = None
        with tf.variable_scope(self.variable_scope):
            if self._hparams.embedding_enabled:
                if embedding is None and vocab_size == None:
                    raise ValueError("If "embedding" is not provided, "
                            ""vocab_size" must be specified.")
                if isinstance(embedding, tf.Variable):
                    self._embedding = embedding
                else:
                    self._embedding = layers.get_embedding(
                        self._hparams.embedding, embedding, vocab_size,
                        variable_scope="dec_embed")
                embed_dim = self._embedding.shape.as_list()[-1]
                if self._hparams.zero_pad:
                    self._embedding = tf.concat((tf.zeros(shape=[1, embed_dim]),
                        self._embedding[1:, :]), 0)
                if self._hparams.embedding.trainable:
                    self._add_trainable_variable(self._embedding)

    @staticmethod
    def default_hparams():
        return {
                "embedding_enabled": True,

After Change

            if self._hparams.embedding.trainable:
                self._add_trainable_variable(self._embedding)
            if self._vocab_size is None:
                self._vocab_size = self._embedding.get_shape().as_list()[0]

    @staticmethod
    def default_hparams():
        return {

Instances