You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
230 lines
10 KiB
230 lines
10 KiB
# Copyright 2023 Meta AI and The HuggingFace Inc. team. 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.
|
|
"""MusicGen model configuration"""
|
|
|
|
from ...configuration_utils import PreTrainedConfig
|
|
from ...utils import logging
|
|
from ..auto.configuration_auto import AutoConfig
|
|
|
|
|
|
logger = logging.get_logger(__name__)
|
|
|
|
|
|
class MusicgenDecoderConfig(PreTrainedConfig):
|
|
r"""
|
|
This is the configuration class to store the configuration of an [`MusicgenDecoder`]. It is used to instantiate a
|
|
MusicGen decoder according to the specified arguments, defining the model architecture. Instantiating a
|
|
configuration with the defaults will yield a similar configuration to that of the MusicGen
|
|
[facebook/musicgen-small](https://huggingface.co/facebook/musicgen-small) architecture.
|
|
|
|
Configuration objects inherit from [`PreTrainedConfig`] and can be used to control the model outputs. Read the
|
|
documentation from [`PreTrainedConfig`] for more information.
|
|
|
|
|
|
Args:
|
|
vocab_size (`int`, *optional*, defaults to 2048):
|
|
Vocabulary size of the MusicgenDecoder model. Defines the number of different tokens that can be
|
|
represented by the `inputs_ids` passed when calling [`MusicgenDecoder`].
|
|
hidden_size (`int`, *optional*, defaults to 1024):
|
|
Dimensionality of the layers and the pooler layer.
|
|
num_hidden_layers (`int`, *optional*, defaults to 24):
|
|
Number of decoder layers.
|
|
num_attention_heads (`int`, *optional*, defaults to 16):
|
|
Number of attention heads for each attention layer in the Transformer block.
|
|
ffn_dim (`int`, *optional*, defaults to 4096):
|
|
Dimensionality of the "intermediate" (often named feed-forward) layer in the Transformer block.
|
|
activation_function (`str` or `function`, *optional*, defaults to `"gelu"`):
|
|
The non-linear activation function (function or string) in the decoder and pooler. If string, `"gelu"`,
|
|
`"relu"`, `"silu"` and `"gelu_new"` are supported.
|
|
dropout (`float`, *optional*, defaults to 0.1):
|
|
The dropout probability for all fully connected layers in the embeddings, text_encoder, and pooler.
|
|
attention_dropout (`float`, *optional*, defaults to 0.0):
|
|
The dropout ratio for the attention probabilities.
|
|
activation_dropout (`float`, *optional*, defaults to 0.0):
|
|
The dropout ratio for activations inside the fully connected layer.
|
|
max_position_embeddings (`int`, *optional*, defaults to 2048):
|
|
The maximum sequence length that this model might ever be used with. Typically, set this to something large
|
|
just in case (e.g., 512 or 1024 or 2048).
|
|
initializer_factor (`float`, *optional*, defaults to 0.02):
|
|
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
|
|
layerdrop (`float`, *optional*, defaults to 0.0):
|
|
The LayerDrop probability for the decoder. See the [LayerDrop paper](see https://huggingface.co/papers/1909.11556)
|
|
for more details.
|
|
scale_embedding (`bool`, *optional*, defaults to `False`):
|
|
Scale embeddings by diving by sqrt(hidden_size).
|
|
use_cache (`bool`, *optional*, defaults to `True`):
|
|
Whether the model should return the last key/values attentions (not used by all models)
|
|
num_codebooks (`int`, *optional*, defaults to 4):
|
|
The number of parallel codebooks forwarded to the model.
|
|
tie_word_embeddings(`bool`, *optional*, defaults to `False`):
|
|
Whether input and output word embeddings should be tied.
|
|
audio_channels (`int`, *optional*, defaults to 1
|
|
Number of channels in the audio data. Either 1 for mono or 2 for stereo. Stereo models generate a separate
|
|
audio stream for the left/right output channels. Mono models generate a single audio stream output.
|
|
"""
|
|
|
|
model_type = "musicgen_decoder"
|
|
base_config_key = "decoder_config"
|
|
keys_to_ignore_at_inference = ["past_key_values"]
|
|
|
|
def __init__(
|
|
self,
|
|
vocab_size=2048,
|
|
max_position_embeddings=2048,
|
|
num_hidden_layers=24,
|
|
ffn_dim=4096,
|
|
num_attention_heads=16,
|
|
layerdrop=0.0,
|
|
use_cache=True,
|
|
activation_function="gelu",
|
|
hidden_size=1024,
|
|
dropout=0.1,
|
|
attention_dropout=0.0,
|
|
activation_dropout=0.0,
|
|
initializer_factor=0.02,
|
|
scale_embedding=False,
|
|
num_codebooks=4,
|
|
audio_channels=1,
|
|
pad_token_id=2048,
|
|
bos_token_id=2048,
|
|
eos_token_id=None,
|
|
tie_word_embeddings=False,
|
|
is_decoder=False,
|
|
add_cross_attention=False,
|
|
cross_attention_hidden_size=None,
|
|
**kwargs,
|
|
):
|
|
self.is_decoder = is_decoder
|
|
self.add_cross_attention = add_cross_attention
|
|
self.cross_attention_hidden_size = cross_attention_hidden_size
|
|
self.vocab_size = vocab_size
|
|
self.max_position_embeddings = max_position_embeddings
|
|
self.hidden_size = hidden_size
|
|
self.ffn_dim = ffn_dim
|
|
self.num_hidden_layers = num_hidden_layers
|
|
self.num_attention_heads = num_attention_heads
|
|
self.dropout = dropout
|
|
self.attention_dropout = attention_dropout
|
|
self.activation_dropout = activation_dropout
|
|
self.activation_function = activation_function
|
|
self.initializer_factor = initializer_factor
|
|
self.layerdrop = layerdrop
|
|
self.use_cache = use_cache
|
|
self.scale_embedding = scale_embedding # scale factor will be sqrt(d_model) if True
|
|
self.num_codebooks = num_codebooks
|
|
|
|
if audio_channels not in [1, 2]:
|
|
raise ValueError(f"Expected 1 (mono) or 2 (stereo) audio channels, got {audio_channels} channels.")
|
|
self.audio_channels = audio_channels
|
|
|
|
self.tie_word_embeddings = tie_word_embeddings
|
|
self.pad_token_id = pad_token_id
|
|
self.bos_token_id = bos_token_id
|
|
self.eos_token_id = eos_token_id
|
|
super().__init__(**kwargs)
|
|
|
|
|
|
class MusicgenConfig(PreTrainedConfig):
|
|
r"""
|
|
This is the configuration class to store the configuration of a [`MusicgenModel`]. It is used to instantiate a
|
|
MusicGen model according to the specified arguments, defining the text encoder, audio encoder and MusicGen 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:
|
|
text_encoder (`Union[dict, `PretrainedConfig`]`):
|
|
An instance of a configuration object that defines the text encoder config.
|
|
audio_encoder (`Union[dict, `PretrainedConfig`]`):
|
|
An instance of a configuration object that defines the audio encoder config.
|
|
decoder (`Union[dict, `PretrainedConfig`]`):
|
|
An instance of a configuration object that defines the decoder config.
|
|
|
|
Example:
|
|
|
|
```python
|
|
>>> from transformers import (
|
|
... MusicgenConfig,
|
|
... MusicgenDecoderConfig,
|
|
... T5Config,
|
|
... EncodecConfig,
|
|
... MusicgenForConditionalGeneration,
|
|
... )
|
|
|
|
>>> # Initializing text encoder, audio encoder, and decoder model configurations
|
|
>>> text_encoder_config = T5Config()
|
|
>>> audio_encoder_config = EncodecConfig()
|
|
>>> decoder_config = MusicgenDecoderConfig()
|
|
|
|
>>> configuration = MusicgenConfig(
|
|
... text_encoder=text_encoder_config,
|
|
... audio_encoder=audio_encoder_config,
|
|
... decoder=decoder_config,
|
|
... )
|
|
|
|
>>> # Initializing a MusicgenForConditionalGeneration (with random weights) from the facebook/musicgen-small style configuration
|
|
>>> model = MusicgenForConditionalGeneration(configuration)
|
|
|
|
>>> # Accessing the model configuration
|
|
>>> configuration = model.config
|
|
>>> config_text_encoder = model.config.text_encoder
|
|
>>> config_audio_encoder = model.config.audio_encoder
|
|
>>> config_decoder = model.config.decoder
|
|
|
|
>>> # Saving the model, including its configuration
|
|
>>> model.save_pretrained("musicgen-model")
|
|
|
|
>>> # loading model and config from pretrained folder
|
|
>>> musicgen_config = MusicgenConfig.from_pretrained("musicgen-model")
|
|
>>> model = MusicgenForConditionalGeneration.from_pretrained("musicgen-model", config=musicgen_config)
|
|
```"""
|
|
|
|
model_type = "musicgen"
|
|
sub_configs = {
|
|
"text_encoder": AutoConfig,
|
|
"audio_encoder": AutoConfig,
|
|
"decoder": MusicgenDecoderConfig,
|
|
}
|
|
has_no_defaults_at_init = True
|
|
|
|
def __init__(self, text_encoder, audio_encoder, decoder, **kwargs):
|
|
if isinstance(text_encoder, dict):
|
|
text_encoder_model_type = text_encoder.pop("model_type")
|
|
text_encoder = AutoConfig.for_model(text_encoder_model_type, **text_encoder)
|
|
|
|
if isinstance(audio_encoder, dict):
|
|
audio_encoder_model_type = audio_encoder.pop("model_type")
|
|
audio_encoder = AutoConfig.for_model(audio_encoder_model_type, **audio_encoder)
|
|
|
|
if isinstance(decoder, dict):
|
|
decoder = MusicgenDecoderConfig(**decoder)
|
|
|
|
self.text_encoder = text_encoder
|
|
self.audio_encoder = audio_encoder
|
|
self.decoder = decoder
|
|
self.initializer_factor = self.decoder.initializer_factor
|
|
self.tie_encoder_decoder = kwargs.get("tie_encoder_decoder", False)
|
|
|
|
kwargs["is_encoder_decoder"] = True
|
|
super().__init__(**kwargs)
|
|
|
|
@property
|
|
# This is a property because you might want to change the codec model on the fly
|
|
def sampling_rate(self):
|
|
return self.audio_encoder.sampling_rate
|
|
|
|
|
|
__all__ = ["MusicgenConfig", "MusicgenDecoderConfig"]
|