mm_action_prediction/models/encoders/hierarchical_recurrent.py [36:62]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        )

    def forward(self, batch):
        """Forward pass through the encoder.

        Args:
            batch: Dict of batch variables.

        Returns:
            encoder_outputs: Dict of outputs from the forward pass.
        """
        encoder_out = {}
        # Flatten to encode sentences.
        batch_size, num_rounds, _ = batch["user_utt"].shape
        encoder_in = support.flatten(batch["user_utt"], batch_size, num_rounds)
        encoder_len = batch["user_utt_len"].reshape(-1)
        word_embeds_enc = self.word_embed_net(encoder_in)

        # Fake encoder_len to be non-zero even for utterances out of dialog.
        fake_encoder_len = encoder_len.eq(0).long() + encoder_len
        all_enc_states, enc_states = rnn.dynamic_rnn(
            self.encoder_unit, word_embeds_enc, fake_encoder_len, return_states=True
        )
        encoder_out["hidden_states_all"] = all_enc_states
        encoder_out["hidden_state"] = enc_states

        utterance_enc = enc_states[0][-1]
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



mm_action_prediction/models/encoders/memory_network.py [45:71]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        )

    def forward(self, batch):
        """Forward pass through the encoder.

        Args:
            batch: Dict of batch variables.

        Returns:
            encoder_outputs: Dict of outputs from the forward pass.
        """
        encoder_out = {}
        # Flatten to encode sentences.
        batch_size, num_rounds, _ = batch["user_utt"].shape
        encoder_in = support.flatten(batch["user_utt"], batch_size, num_rounds)
        encoder_len = batch["user_utt_len"].reshape(-1)
        word_embeds_enc = self.word_embed_net(encoder_in)

        # Fake encoder_len to be non-zero even for utterances out of dialog.
        fake_encoder_len = encoder_len.eq(0).long() + encoder_len
        all_enc_states, enc_states = rnn.dynamic_rnn(
            self.encoder_unit, word_embeds_enc, fake_encoder_len, return_states=True
        )
        encoder_out["hidden_states_all"] = all_enc_states
        encoder_out["hidden_state"] = enc_states

        utterance_enc = enc_states[0][-1]
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -