Sync model code with ace_step==1.6.0 to prevent runtime mutation

acestep-v15-base/configuration_acestep_v15.py

@@ -1,263 +1,5 @@
-# coding=utf-8
-# Copyright 2024 The Qwen team, Alibaba Group 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.
-"""AceStep model configuration"""
-
-from transformers.configuration_utils import PretrainedConfig, layer_type_validation
-from transformers.modeling_rope_utils import rope_config_validation
-from transformers.utils import logging
-
-
-logger = logging.get_logger(__name__)
-
-
-class AceStepConfig(PretrainedConfig):
-    r"""
-    This is the configuration class to store the configuration of a [`AceStepModel`]. It is used to instantiate an
-    AceStep model according to the specified arguments, defining the model 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 64003):
-            Vocabulary size of the AceStep model. Defines the number of different tokens that can be represented by the
-            `inputs_ids` passed when calling the model.
-        hidden_size (`int`, *optional*, defaults to 4096):
-            Dimension of the hidden representations.
-        intermediate_size (`int`, *optional*, defaults to 22016):
-            Dimension of the MLP representations.
-        num_hidden_layers (`int`, *optional*, defaults to 32):
-            Number of hidden layers in the Transformer encoder.
-        num_attention_heads (`int`, *optional*, defaults to 32):
-            Number of attention heads for each attention layer in the Transformer encoder.
-        num_key_value_heads (`int`, *optional*, defaults to 32):
-            This is the number of key_value heads that should be used to implement Grouped Query Attention. If
-            `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
-            `num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When
-            converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
-            by meanpooling all the original heads within that group. For more details, check out [this
-            paper](https://huggingface.co/papers/2305.13245). If it is not specified, will default to `32`.
-        head_dim (`int`, *optional*, defaults to 128):
-            The attention head dimension.
-        hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
-            The non-linear activation function (function or string) in the decoder.
-        max_position_embeddings (`int`, *optional*, defaults to 32768):
-            The maximum sequence length that this model might ever be used with.
-        initializer_range (`float`, *optional*, defaults to 0.02):
-            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
-        rms_norm_eps (`float`, *optional*, defaults to 1e-06):
-            The epsilon used by the rms normalization layers.
-        use_cache (`bool`, *optional*, defaults to `True`):
-            Whether or not the model should return the last key/values attentions (not used by all models). Only
-            relevant if `config.is_decoder=True`.
-        tie_word_embeddings (`bool`, *optional*, defaults to `False`):
-            Whether the model's input and output word embeddings should be tied.
-        rope_theta (`float`, *optional*, defaults to 10000.0):
-            The base period of the RoPE embeddings.
-        rope_scaling (`Dict`, *optional*):
-            Dictionary containing the scaling configuration for the RoPE embeddings. NOTE: if you apply new rope type
-            and you expect the model to work on longer `max_position_embeddings`, we recommend you to update this value
-            accordingly.
-            Expected contents:
-                `rope_type` (`str`):
-                    The sub-variant of RoPE to use. Can be one of ['default', 'linear', 'dynamic', 'yarn', 'longrope',
-                    'llama3'], with 'default' being the original RoPE implementation.
-                `factor` (`float`, *optional*):
-                    Used with all rope types except 'default'. The scaling factor to apply to the RoPE embeddings. In
-                    most scaling types, a `factor` of x will enable the model to handle sequences of length x *
-                    original maximum pre-trained length.
-                `original_max_position_embeddings` (`int`, *optional*):
-                    Used with 'dynamic', 'longrope' and 'llama3'. The original max position embeddings used during
-                    pretraining.
-                `attention_factor` (`float`, *optional*):
-                    Used with 'yarn' and 'longrope'. The scaling factor to be applied on the attention
-                    computation. If unspecified, it defaults to value recommended by the implementation, using the
-                    `factor` field to infer the suggested value.
-                `beta_fast` (`float`, *optional*):
-                    Only used with 'yarn'. Parameter to set the boundary for extrapolation (only) in the linear
-                    ramp function. If unspecified, it defaults to 32.
-                `beta_slow` (`float`, *optional*):
-                    Only used with 'yarn'. Parameter to set the boundary for interpolation (only) in the linear
-                    ramp function. If unspecified, it defaults to 1.
-                `short_factor` (`list[float]`, *optional*):
-                    Only used with 'longrope'. The scaling factor to be applied to short contexts (<
-                    `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden
-                    size divided by the number of attention heads divided by 2
-                `long_factor` (`list[float]`, *optional*):
-                    Only used with 'longrope'. The scaling factor to be applied to long contexts (<
-                    `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden
-                    size divided by the number of attention heads divided by 2
-                `low_freq_factor` (`float`, *optional*):
-                    Only used with 'llama3'. Scaling factor applied to low frequency components of the RoPE
-                `high_freq_factor` (`float`, *optional*):
-                    Only used with 'llama3'. Scaling factor applied to high frequency components of the RoPE
-        attention_bias (`bool`, defaults to `False`, *optional*, defaults to `False`):
-            Whether to use a bias in the query, key, value and output projection layers during self-attention.
-        use_sliding_window (`bool`, *optional*, defaults to `False`):
-            Whether to use sliding window attention.
-        sliding_window (`int`, *optional*, defaults to 4096):
-            Sliding window attention (SWA) window size. If not specified, will default to `4096`.
-        layer_types (`list`, *optional*):
-            Attention pattern for each layer.
-        attention_dropout (`float`, *optional*, defaults to 0.0):
-            The dropout ratio for the attention probabilities.
-
-    ```python
-    >>> from acestep.models import AceStepConfig
-
-    >>> # Initializing an AceStep configuration
-    >>> configuration = AceStepConfig()
-
-    >>> # Initializing a model from the configuration
-    >>> model = AceStepConditionGenerationModel(configuration)
-
-    >>> # Accessing the model configuration
-    >>> configuration = model.config
-    ```"""
-
-    model_type = "acestep"
-    keys_to_ignore_at_inference = ["past_key_values"]
-    
-    # Default tensor parallel plan for the base model
-    base_model_tp_plan = {
-        "layers.*.self_attn.q_proj": "colwise",
-        "layers.*.self_attn.k_proj": "colwise",
-        "layers.*.self_attn.v_proj": "colwise",
-        "layers.*.self_attn.o_proj": "rowwise",
-        "layers.*.mlp.gate_proj": "colwise",
-        "layers.*.mlp.up_proj": "colwise",
-        "layers.*.mlp.down_proj": "rowwise",
-    }
-    base_model_pp_plan = {
-        "embed_tokens": (["input_ids"], ["inputs_embeds"]),
-        "layers": (["hidden_states", "attention_mask"], ["hidden_states"]),
-        "norm": (["hidden_states"], ["hidden_states"]),
-    }
-    def __init__(
-        self,
-        vocab_size=64003,
-        fsq_dim=2048,
-        fsq_input_levels=[8, 8, 8, 5, 5, 5],
-        fsq_input_num_quantizers=1,
-        hidden_size=2048,
-        intermediate_size=6144,
-        num_hidden_layers=24,
-        num_attention_heads=16,
-        num_key_value_heads=8,
-        head_dim=128,
-        hidden_act="silu",
-        max_position_embeddings=32768,
-        initializer_range=0.02,
-        rms_norm_eps=1e-6,
-        use_cache=True,
-        tie_word_embeddings=True,
-        rope_theta=1000000,
-        rope_scaling=None,
-        attention_bias=False,
-        use_sliding_window=True,
-        sliding_window=128,
-        layer_types=None,
-        attention_dropout=0.0,
-        num_lyric_encoder_hidden_layers=8,
-        audio_acoustic_hidden_dim=64,
-        pool_window_size=5,
-        text_hidden_dim=1024,
-        in_channels=192,
-        data_proportion=0.5,
-        timestep_mu=-0.4,
-        timestep_sigma=1.0,
-        timbre_hidden_dim=64,
-        num_timbre_encoder_hidden_layers=4,
-        timbre_fix_frame=750,
-        patch_size=2,
-        num_attention_pooler_hidden_layers=2,
-        num_audio_decoder_hidden_layers=24,
-        model_version="turbo",
-        **kwargs,
-    ):
-        self.max_position_embeddings = max_position_embeddings
-        self.hidden_size = hidden_size
-        self.intermediate_size = intermediate_size
-        self.num_hidden_layers = num_hidden_layers
-        self.num_attention_heads = num_attention_heads
-        self.use_sliding_window = use_sliding_window
-        self.sliding_window = sliding_window if self.use_sliding_window else None
-        
-        # Text encoder configuration
-        self.text_hidden_dim = text_hidden_dim
-
-        # Lyric encoder configuration
-        self.num_lyric_encoder_hidden_layers = num_lyric_encoder_hidden_layers
-        self.patch_size = patch_size
-
-        # Audio semantic token generation configuration
-        self.audio_acoustic_hidden_dim = audio_acoustic_hidden_dim
-        self.pool_window_size = pool_window_size
-        self.in_channels = in_channels
-        self.data_proportion = data_proportion
-        self.timestep_mu = timestep_mu
-        self.timestep_sigma = timestep_sigma
-        
-        # FSQ (Finite Scalar Quantization) configuration
-        self.fsq_dim = fsq_dim
-        self.fsq_input_levels = fsq_input_levels
-        self.fsq_input_num_quantizers = fsq_input_num_quantizers
-        
-        # Timbre encoder configuration
-        self.timbre_hidden_dim = timbre_hidden_dim
-        self.num_timbre_encoder_hidden_layers = num_timbre_encoder_hidden_layers
-        self.timbre_fix_frame = timbre_fix_frame
-        self.num_attention_pooler_hidden_layers = num_attention_pooler_hidden_layers
-        self.num_audio_decoder_hidden_layers = num_audio_decoder_hidden_layers
-        self.vocab_size = vocab_size
-
-        # Backward compatibility: ensure num_key_value_heads is set
-        if num_key_value_heads is None:
-            num_key_value_heads = num_attention_heads
-
-        self.num_key_value_heads = num_key_value_heads
-        self.head_dim = head_dim
-        self.hidden_act = hidden_act
-        self.initializer_range = initializer_range
-        self.rms_norm_eps = rms_norm_eps
-        self.use_cache = use_cache
-        self.rope_theta = rope_theta
-        self.rope_scaling = rope_scaling
-        self.attention_bias = attention_bias
-        self.attention_dropout = attention_dropout
-        self.model_version = model_version
-        
-        # Validate rotary position embeddings parameters
-        # Backward compatibility: if there is a 'type' field, move it to 'rope_type'
-        if self.rope_scaling is not None and "type" in self.rope_scaling:
-            self.rope_scaling["rope_type"] = self.rope_scaling["type"]
-        rope_config_validation(self)
-
-        self.layer_types = layer_types
-
-        # Set default layer types if not specified
-        if self.layer_types is None:
-            self.layer_types = [
-                "sliding_attention" if bool((i + 1) % 2) else "full_attention" for i in range(self.num_hidden_layers)
-            ]
-        layer_type_validation(self.layer_types)
-
-        super().__init__(
-            tie_word_embeddings=tie_word_embeddings,
-            **kwargs,
-        )
-
+# Re-export from canonical location to avoid duplication.
+# All model variants share the same AceStepConfig.
+from acestep.models.common.configuration_acestep_v15 import AceStepConfig
 
 __all__ = ["AceStepConfig"]