qi`dZddlmZddlmZej eZGddeZdgZ y)zSqueezeBERT model configuration)PreTrainedConfig)loggingcVeZdZdZdZ dfd ZxZS)SqueezeBertConfiga This is the configuration class to store the configuration of a [`SqueezeBertModel`]. It is used to instantiate a SqueezeBERT model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the SqueezeBERT [squeezebert/squeezebert-uncased](https://huggingface.co/squeezebert/squeezebert-uncased) 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 30522): Vocabulary size of the SqueezeBERT model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`SqueezeBertModel`]. hidden_size (`int`, *optional*, defaults to 768): Dimensionality of the encoder layers and the pooler layer. num_hidden_layers (`int`, *optional*, defaults to 12): Number of hidden layers in the Transformer encoder. num_attention_heads (`int`, *optional*, defaults to 12): Number of attention heads for each attention layer in the Transformer encoder. intermediate_size (`int`, *optional*, defaults to 3072): Dimensionality of the "intermediate" (often named feed-forward) layer in the Transformer encoder. hidden_act (`str` or `Callable`, *optional*, defaults to `"gelu"`): The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`, `"relu"`, `"silu"` and `"gelu_new"` are supported. hidden_dropout_prob (`float`, *optional*, defaults to 0.1): The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. attention_probs_dropout_prob (`float`, *optional*, defaults to 0.1): The dropout ratio for the attention probabilities. max_position_embeddings (`int`, *optional*, defaults to 512): 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). type_vocab_size (`int`, *optional*, defaults to 2): The vocabulary size of the `token_type_ids` passed when calling [`BertModel`]. initializer_range (`float`, *optional*, defaults to 0.02): The standard deviation of the truncated_normal_initializer for initializing all weight matrices. layer_norm_eps (`float`, *optional*, defaults to 1e-12): pad_token_id (`int`, *optional*, defaults to 0): The ID of the token in the word embedding to use as padding. bos_token_id (`int`, *optional*): Beginning of stream token id. eos_token_id (`int`, *optional*): End of stream token id. embedding_size (`int`, *optional*, defaults to 768): The dimension of the word embedding vectors. q_groups (`int`, *optional*, defaults to 4): The number of groups in Q layer. k_groups (`int`, *optional*, defaults to 4): The number of groups in K layer. v_groups (`int`, *optional*, defaults to 4): The number of groups in V layer. post_attention_groups (`int`, *optional*, defaults to 1): The number of groups in the first feed forward network layer. intermediate_groups (`int`, *optional*, defaults to 4): The number of groups in the second feed forward network layer. output_groups (`int`, *optional*, defaults to 4): The number of groups in the third feed forward network layer. tie_word_embeddings (`bool`, *optional*, defaults to `True`): Whether to tie weight embeddings Examples: ```python >>> from transformers import SqueezeBertConfig, SqueezeBertModel >>> # Initializing a SqueezeBERT configuration >>> configuration = SqueezeBertConfig() >>> # Initializing a model (with random weights) from the configuration above >>> model = SqueezeBertModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ``` squeezebertc ft|di|||_| |_||_||_||_||_||_||_ ||_ ||_ ||_ ||_ | |_| |_| |_| |_||_||_||_||_||_||_||_y)N)super__init__tie_word_embeddings pad_token_id bos_token_id eos_token_id vocab_size hidden_sizenum_hidden_layersnum_attention_heads hidden_actintermediate_sizehidden_dropout_probattention_probs_dropout_probmax_position_embeddingstype_vocab_sizeinitializer_rangelayer_norm_epsembedding_sizeq_groupsk_groupsv_groupspost_attention_groupsintermediate_groups output_groups)selfrrrrrrrrrrrrr rrrrrrr r!r"r kwargs __class__s k/opt/pipecat/venv/lib/python3.12/site-packages/transformers/models/squeezebert/configuration_squeezebert.pyr zSqueezeBertConfig.__init__hs6 "6"#6 ((($&!2#6 $!2#6 ,H)'>$.!2,,      %:"#6 *)i:w r)i gelu皙?r+ig{Gz?g-q=NNr(r.r.r.r.T)__name__ __module__ __qualname____doc__ model_typer __classcell__)r%s@r&rrsbL\J%( # 13+3+r'rN) r3configuration_utilsrutilsr get_loggerr0loggerr__all__r r'r&r;s>&3   H %D+(D+N  r'