models/configuration_audio_encoder_decoder.py

#!/usr/bin/env python3
# coding: utf-8
# @Author  : Xinhao Mei @CVSSP, University of Surrey
# @E-mail  : x.mei@surrey.ac.uk
# coding=utf-8
# Copyright 2021 The HuggingFace Inc. team.
# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

import copy

from transformers import PretrainedConfig, AutoConfig, BertConfig

from models.audio_encoder_config import AudioEncoderConfig
from models.audio_encoder import AudioEncoderModel


class AudioEncoderDecoderConfig(PretrainedConfig):
    r"""
    [`VisionEncoderDecoderConfig`] is the configuration class to store the configuration of a
    [`VisionEncoderDecoderModel`]. It is used to instantiate a Vision-Encoder-Text-Decoder model according to the
    specified arguments, defining the encoder and decoder configs.

    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
    documentation from [`PretrainedConfig`] for more information.

    Args:
        kwargs (*optional*):
            Dictionary of keyword arguments. Notably:

                - **encoder** ([`PretrainedConfig`], *optional*) -- An instance of a configuration object that defines
                  the encoder config.
                - **decoder** ([`PretrainedConfig`], *optional*) -- An instance of a configuration object that defines
                  the decoder config.

    Examples:

    ```python
    >>> from transformers import BertConfig, ViTConfig, VisionEncoderDecoderConfig, VisionEncoderDecoderModel

    >>> # Initializing a ViT & BERT style configuration
    >>> config_encoder = ViTConfig()
    >>> config_decoder = BertConfig()

    >>> config = VisionEncoderDecoderConfig.from_encoder_decoder_configs(config_encoder, config_decoder)

    >>> # Initializing a ViTBert model from a ViT & bert-base-uncased style configurations
    >>> model = VisionEncoderDecoderModel(config=config)

    >>> # Accessing the model configuration
    >>> config_encoder = model.config.encoder
    >>> config_decoder = model.config.decoder
    >>> # set decoder config to causal lm
    >>> config_decoder.is_decoder = True
    >>> config_decoder.add_cross_attention = True

    >>> # Saving the model, including its configuration
    >>> model.save_pretrained("my-model")

    >>> # loading model and config from pretrained folder
    >>> encoder_decoder_config = VisionEncoderDecoderConfig.from_pretrained("my-model")
    >>> model = VisionEncoderDecoderModel.from_pretrained("my-model", config=encoder_decoder_config)
    ```"""
    model_type = "audio-encoder-decoder"
    is_composition = True

    def __init__(self, **kwargs):
        super().__init__(**kwargs)
        if "encoder" not in kwargs or "decoder" not in kwargs:
            raise ValueError(
                f"A configuraton of type {self.model_type} cannot be instantiated because "
                f"not both `encoder` and `decoder` sub-configurations are passed, but only {kwargs}"
            )

        encoder_config = kwargs.pop("encoder")
        # encoder_model_type = encoder_config.pop("model_type")
        decoder_config = kwargs.pop("decoder")
        decoder_model_type = decoder_config.pop("model_type")

        self.encoder = AudioEncoderConfig(**encoder_config)
        # self.encoder = AudioEncoderModel(encoder_config)
        self.decoder = AutoConfig.for_model(decoder_model_type, **decoder_config)
        self.is_encoder_decoder = True
        # self.num_labels = 0

    @classmethod
    def from_encoder_decoder_configs(
        cls, encoder_config: PretrainedConfig, decoder_config: PretrainedConfig, **kwargs
    ) -> PretrainedConfig:
        r"""
        Instantiate a [`VisionEncoderDecoderConfig`] (or a derived class) from a pre-trained encoder model
        configuration and decoder model configuration.

        Returns:
            [`VisionEncoderDecoderConfig`]: An instance of a configuration object
        """
        decoder_config.is_decoder = True
        decoder_config.add_cross_attention = True

        return cls(encoder=encoder_config.to_dict(), decoder=decoder_config.to_dict(), **kwargs)

    def to_dict(self):
        """
        Serializes this instance to a Python dictionary. Override the default *to_dict()* from *PretrainedConfig*.

        Returns:
            `Dict[str, any]`: Dictionary of all the attributes that make up this configuration instance,
        """
        output = copy.deepcopy(self.__dict__)
        output["encoder"] = self.encoder.to_dict()
        output["decoder"] = self.decoder.to_dict()
        output["model_type"] = self.__class__.model_type
        return output


if __name__ == '__main__':
    import os
    import ruamel.yaml as yaml

    os.chdir("../")
    with open("settings/settings.yaml", "r") as f:
        config = yaml.safe_load(f)
    encoder_config = AudioEncoderConfig(**config["audio_encoder_args"])

    config_decoder = BertConfig()

    config = AudioEncoderDecoderConfig.from_encoder_decoder_configs(encoder_config, config_decoder)
    print(config)