o ۷i/@sdZddlmZddlmZddlmZmZddlm Z m Z m Z ddl m Z ddlmZmZdd lmZeeZGd d d e ZGd d d eZd d gZdS)zOpenAI GPT-2 configuration) OrderedDict)Mapping)AnyOptional)PreTrainedTokenizer TensorTypeis_torch_available)PretrainedConfig)OnnxConfigWithPast PatchingSpec)loggingcsheZdZdZdZdgZdddddZ                dfdd ZZS) GPT2ConfigaK This is the configuration class to store the configuration of a [`GPT2Model`] or a [`TFGPT2Model`]. It is used to instantiate a GPT-2 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 GPT-2 [openai-community/gpt2](https://huggingface.co/openai-community/gpt2) 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 50257): Vocabulary size of the GPT-2 model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`GPT2Model`] or [`TFGPT2Model`]. n_positions (`int`, *optional*, defaults to 1024): 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). n_embd (`int`, *optional*, defaults to 768): Dimensionality of the embeddings and hidden states. n_layer (`int`, *optional*, defaults to 12): Number of hidden layers in the Transformer encoder. n_head (`int`, *optional*, defaults to 12): Number of attention heads for each attention layer in the Transformer encoder. n_inner (`int`, *optional*): Dimensionality of the inner feed-forward layers. `None` will set it to 4 times n_embd activation_function (`str`, *optional*, defaults to `"gelu_new"`): Activation function, to be selected in the list `["relu", "silu", "gelu", "tanh", "gelu_new"]`. resid_pdrop (`float`, *optional*, defaults to 0.1): The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. embd_pdrop (`float`, *optional*, defaults to 0.1): The dropout ratio for the embeddings. attn_pdrop (`float`, *optional*, defaults to 0.1): The dropout ratio for the attention. layer_norm_epsilon (`float`, *optional*, defaults to 1e-05): The epsilon to use in the layer normalization layers. initializer_range (`float`, *optional*, defaults to 0.02): The standard deviation of the truncated_normal_initializer for initializing all weight matrices. summary_type (`string`, *optional*, defaults to `"cls_index"`): Argument used when doing sequence summary, used in the models [`GPT2DoubleHeadsModel`] and [`TFGPT2DoubleHeadsModel`]. Has to be one of the following options: - `"last"`: Take the last token hidden state (like XLNet). - `"first"`: Take the first token hidden state (like BERT). - `"mean"`: Take the mean of all tokens hidden states. - `"cls_index"`: Supply a Tensor of classification token position (like GPT/GPT-2). - `"attn"`: Not implemented now, use multi-head attention. summary_use_proj (`bool`, *optional*, defaults to `True`): Argument used when doing sequence summary, used in the models [`GPT2DoubleHeadsModel`] and [`TFGPT2DoubleHeadsModel`]. Whether or not to add a projection after the vector extraction. summary_activation (`str`, *optional*): Argument used when doing sequence summary. Used in for the multiple choice head in [`GPT2DoubleHeadsModel`]. Pass `"tanh"` for a tanh activation to the output, any other value will result in no activation. summary_proj_to_labels (`bool`, *optional*, defaults to `True`): Argument used when doing sequence summary, used in the models [`GPT2DoubleHeadsModel`] and [`TFGPT2DoubleHeadsModel`]. Whether the projection outputs should have `config.num_labels` or `config.hidden_size` classes. summary_first_dropout (`float`, *optional*, defaults to 0.1): Argument used when doing sequence summary, used in the models [`GPT2DoubleHeadsModel`] and [`TFGPT2DoubleHeadsModel`]. The dropout ratio to be used after the projection and activation. scale_attn_weights (`bool`, *optional*, defaults to `True`): Scale attention weights by dividing by sqrt(hidden_size).. use_cache (`bool`, *optional*, defaults to `True`): Whether or not the model should return the last key/values attentions (not used by all models). bos_token_id (`int`, *optional*, defaults to 50256): Id of the beginning of sentence token in the vocabulary. eos_token_id (`int`, *optional*, defaults to 50256): Id of the end of sentence token in the vocabulary. scale_attn_by_inverse_layer_idx (`bool`, *optional*, defaults to `False`): Whether to additionally scale attention weights by `1 / layer_idx + 1`. reorder_and_upcast_attn (`bool`, *optional*, defaults to `False`): Whether to scale keys (K) prior to computing attention (dot-product) and upcast attention dot-product/softmax to float() when training with mixed precision. Example: ```python >>> from transformers import GPT2Config, GPT2Model >>> # Initializing a GPT2 configuration >>> configuration = GPT2Config() >>> # Initializing a model (with random weights) from the configuration >>> model = GPT2Model(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```gpt2past_key_valuesn_embd n_positionsn_headn_layer) hidden_sizemax_position_embeddingsnum_attention_headsnum_hidden_layersQ Ngelu_new皙?h㈵>{Gz? cls_indexTPFc s||_||_||_||_||_||_||_||_| |_| |_ | |_ | |_ | |_ ||_ ||_||_||_||_||_||_||_||_||_tjd||d|dS)N) bos_token_id eos_token_id) vocab_sizerrrrn_inneractivation_function resid_pdrop embd_pdrop attn_pdroplayer_norm_epsiloninitializer_range summary_typesummary_use_projsummary_activationsummary_first_dropoutsummary_proj_to_labelsscale_attn_weights use_cachescale_attn_by_inverse_layer_idxreorder_and_upcast_attnr#r$super__init__)selfr&rrrrr'r(r)r*r+r,r-r.r/r0r2r1r3r4r#r$r5r6kwargs __class__r%a/home/ubuntu/vllm_env/lib/python3.10/site-packages/transformers/models/gpt2/configuration_gpt2.pyr8s0zGPT2Config.__init__)rrrrrNrrrrrr r!TNTrTTr"r"FF) __name__ __module__ __qualname____doc__ model_typekeys_to_ignore_at_inference attribute_mapr8 __classcell__r%r%r;r=rsBa rcseZdZ   ddededeeedeffdd Z e d e ee e efffd d Z e d e fd dZe d e fddZ    ddede de dedeed e eeff fdd Ze d e fddZZS)GPT2OnnxConfigdefaultNFconfigtaskpatching_specsuse_pastcs2tj||||dt|jddsd|j_dSdS)N)rIrJrK pad_token_idr)r7r8getattr_configrL)r9rHrIrJrKr;r%r=r8s zGPT2OnnxConfig.__init__returncCsJtddddi}|jr|j|ddddd|d<|Sddd|d<|S) N input_idsbatchsequence)rinputs) directionzpast_sequence + sequenceattention_mask)rrKfill_with_past_key_values_)r9 common_inputsr%r%r=rTszGPT2OnnxConfig.inputscC|jjSN)rNrr9r%r%r= num_layerszGPT2OnnxConfig.num_layerscCrYrZ)rNrr[r%r%r=rr]z"GPT2OnnxConfig.num_attention_heads tokenizer batch_size seq_lengthis_pair frameworkc stt|j|||||d}td|di}|jrItstdddl|dj\}} | d} ||j | |j j |j ffddt |j D|d<|d |d <|jrj|d j} j|d j|| | d gd d |d <|S) N)r`rarbrcrPzACannot generate dummy past_keys inputs without PyTorch installed.rcs g|] }fqSr%)zeros).0_ past_shapetorchr%r= sz8GPT2OnnxConfig.generate_dummy_inputs..rrV)dtyperS)dim)r7r generate_dummy_inputsrrKr ValueErrorrjshaperrNrranger\rlcatones) r9r_r`rarbrcrXordered_inputsrQseqlenpast_key_values_length mask_dtyper;rhr=rns2       z$GPT2OnnxConfig.generate_dummy_inputscCsdS)N r%r[r%r%r=default_onnx_opset sz!GPT2OnnxConfig.default_onnx_opset)rGNF)r^r^FN)r>r?r@r strrlistr boolr8propertyrintrTr\rrrrrnryrEr%r%r;r=rFsL    ,rFN)rA collectionsrcollections.abcrtypingrrrrr configuration_utilsr onnxr r utilsr get_loggerr>loggerrrF__all__r%r%r%r=s     # Q