Class EncoderDecoderModel

src/transformers/modeling_encoder_decoder.py:29–316

r""" :class:`~transformers.EncoderDecoder` is a generic model class that will be instantiated as a transformer architecture with one of the base model classes of the library as encoder and another one as decoder when created with the `AutoModel.from_pretrained(pretrai

class EncoderDecoderModel(PreTrainedModel):
    r"""
        :class:`~transformers.EncoderDecoder` is a generic model class that will be
        instantiated as a transformer architecture with one of the base model
        classes of the library as encoder and another one as
        decoder when created with the `AutoModel.from_pretrained(pretrained_model_name_or_path)`
        class method for the encoder and `AutoModelForCausalLM.from_pretrained(pretrained_model_name_or_path)` class method for the decoder.
    """
    config_class = EncoderDecoderConfig
    base_model_prefix = "encoder_decoder"

    def __init__(
        self,
        config: Optional[PretrainedConfig] = None,
        encoder: Optional[PreTrainedModel] = None,
        decoder: Optional[PreTrainedModel] = None,
    ):
        assert config is not None or (
            encoder is not None and decoder is not None
        ), "Either a configuration or an Encoder and a decoder has to be provided"
        if config is None:
            config = EncoderDecoderConfig.from_encoder_decoder_configs(encoder.config, decoder.config)
        else:
            assert isinstance(config, self.config_class), "config: {} has to be of type {}".format(
                config, self.config_class
            )
        # initialize with config
        super().__init__(config)

        if encoder is None:
            from transformers import AutoModel

            encoder = AutoModel.from_config(config.encoder)

        if decoder is None:
            from transformers import AutoModelForCausalLM

            decoder = AutoModelForCausalLM.from_config(config.decoder)

        self.encoder = encoder
        self.decoder = decoder
        assert (
            self.encoder.get_output_embeddings() is None
        ), "The encoder {} should not have a LM Head. Please use a model without LM Head"

    def tie_weights(self):
        # for now no weights tying in encoder-decoder
        pass

    def get_encoder(self):
        return self.encoder

    def get_decoder(self):
        return self.decoder

    def get_input_embeddings(self):
        return self.encoder.get_input_embeddings()