T^(h{dZddlmZddlmZddlmZddlmZddl m Z e je Z Gdd eZGd d eZd d gZy ) zRemBERT model configuration) OrderedDict)Mapping)PretrainedConfig) OnnxConfig)loggingcNeZdZdZdZ dfd ZxZS) RemBertConfiga This is the configuration class to store the configuration of a [`RemBertModel`]. It is used to instantiate an RemBERT 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 RemBERT [google/rembert](https://huggingface.co/google/rembert) 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 250300): Vocabulary size of the RemBERT model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`RemBertModel`] or [`TFRemBertModel`]. Vocabulary size of the model. Defines the different tokens that can be represented by the *inputs_ids* passed to the forward method of [`RemBertModel`]. hidden_size (`int`, *optional*, defaults to 1152): Dimensionality of the encoder layers and the pooler layer. num_hidden_layers (`int`, *optional*, defaults to 32): Number of hidden layers in the Transformer encoder. num_attention_heads (`int`, *optional*, defaults to 18): Number of attention heads for each attention layer in the Transformer encoder. input_embedding_size (`int`, *optional*, defaults to 256): Dimensionality of the input embeddings. output_embedding_size (`int`, *optional*, defaults to 1664): Dimensionality of the output embeddings. intermediate_size (`int`, *optional*, defaults to 4608): Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder. hidden_act (`str` or `function`, *optional*, defaults to `"gelu"`): The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`, `"relu"`, `"selu"` and `"gelu_new"` are supported. hidden_dropout_prob (`float`, *optional*, defaults to 0): The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. attention_probs_dropout_prob (`float`, *optional*, defaults to 0): The dropout ratio for the attention probabilities. classifier_dropout_prob (`float`, *optional*, defaults to 0.1): The dropout ratio for the classifier layer when fine-tuning. 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 [`RemBertModel`] or [`TFRemBertModel`]. 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): The epsilon used by the layer normalization layers. is_decoder (`bool`, *optional*, defaults to `False`): Whether the model is used as a decoder or not. If `False`, the model is used as an encoder. 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`. Example: ```python >>> from transformers import RemBertModel, RemBertConfig >>> # Initializing a RemBERT rembert style configuration >>> configuration = RemBertConfig() >>> # Initializing a model from the rembert style configuration >>> model = RemBertModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```rembertc t|d|||d|||_||_||_| |_||_||_||_||_ ||_ | |_ | |_ | |_ ||_| |_||_||_d|_y)N) pad_token_id bos_token_id eos_token_idF)super__init__ vocab_sizeinput_embedding_sizeoutput_embedding_sizemax_position_embeddings hidden_sizenum_hidden_layersnum_attention_headsintermediate_size hidden_acthidden_dropout_probattention_probs_dropout_probclassifier_dropout_probinitializer_rangetype_vocab_sizelayer_norm_eps use_cachetie_word_embeddings)selfrrrrrrrrrrrrr rr!r"r rrkwargs __class__s o/var/www/skyplay_api_hub/venv/lib/python3.12/site-packages/transformers/models/rembert/configuration_rembert.pyrzRemBertConfig.__init__bs. sl\hslrs$$8!%:"'>$&!2#6 !2$#6 ,H)'>$!2.,"#( )ii iigelur-g?ig{Gz?g-q=Tri8i9)__name__ __module__ __qualname____doc__ model_typer __classcell__)r&s@r'r r sVA FJ "%( # #)))))r(r cLeZdZedeeeeefffdZedefdZ y)RemBertOnnxConfigreturnc`|jdk(rdddd}nddd}td|fd|fd |fgS) Nzmultiple-choicebatchchoicesequence)rr.)rr< input_idsattention_masktoken_type_ids)taskr)r$ dynamic_axiss r'inputszRemBertOnnxConfig.inputssO 99) )&8 CL&:6Ll+!<0!<0   r(cy)Ng-C6?r)r$s r'atol_for_validationz%RemBertOnnxConfig.atol_for_validationsr(N) r/r0r1propertyrstrintrBfloatrDrr(r'r6r6sI   WS#X%6 67    Ur(r6N)r2 collectionsrtypingrconfiguration_utilsronnxrutilsr get_loggerr/loggerr r6__all__rr(r'rQsV"#3   H %o)$o)d ( / 0r(