sg.dZddlmZddlmZmZmZmZddlm Z m Z m Z ddl m Z ddlmZmZddlmZej&eZGd d e ZGd d eZy )zOpenAI GPT-2 configuration) OrderedDict)AnyListMappingOptional)PreTrainedTokenizer TensorTypeis_torch_available)PretrainedConfig)OnnxConfigWithPast PatchingSpec)loggingcjeZdZdZdZdgZdddddZ d fd ZxZS) 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_layersc l||_||_||_||_||_||_||_||_| |_| |_ | |_ | |_ | |_ ||_ ||_||_||_||_||_||_||_||_||_t/|`d||d|y)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_attnrrsuper__init__)selfr rrrrr!r"r#r$r%r&r'r(r)r*r,r+r-r.rrr/r0kwargs __class__s ^/var/www/html/venv/lib/python3.12/site-packages/transformers/models/gpt2/configuration_gpt2.pyr2zGPT2Config.__init__s6%&    #6 &$$"4!2( 0"4%:"&<#"4"/N,'>$(( XlXQWX)iQii r8Ngelu_new皙?r:r:gh㈵>g{Gz? cls_indexTNTr:TTPr<FF) __name__ __module__ __qualname____doc__ model_typekeys_to_ignore_at_inference attribute_mapr2 __classcell__r5s@r6rrs_ BJ#4"5#0'& M& #!(- %14Y4Yr7rceZdZ ddededeedeffd Ze de ee e efffdZ e de fdZ e de fd Z dd ed e d e d edeede eeff fd Ze de fdZxZS)GPT2OnnxConfigconfigtaskpatching_specsuse_pastc~t|||||t|jddsd|j_yy)N)rIrJrK pad_token_idr)r1r2getattr_configrM)r3rHrIrJrKr5s r6r2zGPT2OnnxConfig.__init__s= d>T\]t||^T:()DLL %;r7returnctddddi}|jr|j|dddd|d<|Sddd|d<|S) N input_idsbatchsequence)rinputs) directionzpast_sequence + sequenceattention_mask)rrKfill_with_past_key_values_)r3 common_inputss r6rVzGPT2OnnxConfig.inputssa#[g*2M$NO ==  + +MX + N29>X.YM* +3:j.IM* +r7c.|jjSN)rOrr3s r6 num_layerszGPT2OnnxConfig.num_layerss||###r7c.|jjSr\)rOrr]s r6rz"GPT2OnnxConfig.num_attention_headss||"""r7 tokenizer batch_size seq_lengthis_pair frameworkchtt| |||||}td|di}|jrt s t dddl}|dj\} } | dz} | |j| |jj|jzf} t|jD cgc]$} |j| |j| f&c} |d<|d|d<|jr<|dj}j!|d|j#  |gd |d<|Scc} w) N)rarbrcrdrRzACannot generate dummy past_keys inputs without PyTorch installed.rrrX)dtyperU)dim)r1r generate_dummy_inputsrrKr ValueErrortorchshaperrOrranger^zerosrgcatones)r3r`rarbrcrdrZordered_inputsrkrSseqlenpast_key_values_length past_shape_ mask_dtyper5s r6riz$GPT2OnnxConfig.generate_dummy_inputss^0$M *W`iN %k=3M%NO ==%' !dee -k : @ @ v)/!&,,*LL,,0H0HH  QVVZVeVePf5KLU[[,ekk*.EF501,99I+J'( =='(89??J/4yy 015::eE[cm:3nouv090N+ ,5s.)D/cy)N rr]s r6default_onnx_opsetz!GPT2OnnxConfig.default_onnx_opset sr7)defaultNF)r{FN)r=r>r?r strrrboolr2propertyrintrVr^rr rr rriryrDrEs@r6rGrGs-1 *  * *\* *  *WS#X%6 67$C$$#S## *. *&** *  * J' * c *XCr7rGN)r@ collectionsrtypingrrrrr r r configuration_utilsr onnxr rutilsr get_loggerr=loggerrrGrr7r6rsW !#//CC34   H %_Y!_YDN'Nr7