sg%dZddlmZddlmZmZmZmZddlm Z m Z m Z ddl m Z ddlmZmZddlmZej&eZGd d e ZGd d eZy )zCodeGen model configuration) OrderedDict)AnyListMappingOptional)PreTrainedTokenizer TensorTypeis_torch_available)PretrainedConfig)OnnxConfigWithPast PatchingSpec)loggingcZeZdZdZdZdddddZ d fd ZxZS) CodeGenConfiga This is the configuration class to store the configuration of a [`CodeGenModel`]. It is used to instantiate a CodeGen 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 CodeGen [Salesforce/codegen-2B-mono](https://huggingface.co/Salesforce/codegen-2B-mono) 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 50400): Vocabulary size of the CodeGen model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`CodeGenModel`]. n_positions (`int`, *optional*, defaults to 2048): 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_ctx (`int`, *optional*, defaults to 2048): This attribute is used in `CodeGenModel.__init__` without any real effect. n_embd (`int`, *optional*, defaults to 4096): Dimensionality of the embeddings and hidden states. n_layer (`int`, *optional*, defaults to 28): Number of hidden layers in the Transformer encoder. n_head (`int`, *optional*, defaults to 16): Number of attention heads for each attention layer in the Transformer encoder. rotary_dim (`int`, *optional*, defaults to 64): Number of dimensions in the embedding that Rotary Position Embedding is applied to. 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 0.0): The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. embd_pdrop (`int`, *optional*, defaults to 0.0): The dropout ratio for the embeddings. attn_pdrop (`float`, *optional*, defaults to 0.0): 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. 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): Beginning of stream token id. eos_token_id (`int`, *optional*, defaults to 50256): End of stream token id. tie_word_embeddings (`bool`, *optional*, defaults to `False`): Whether the model's input and output word embeddings should be tied. Note that this is only relevant if the model has a output word embedding layer. Example: ```python >>> from transformers import CodeGenConfig, CodeGenModel >>> # Initializing a CodeGen 6B configuration >>> configuration = CodeGenConfig() >>> # Initializing a model (with random weights) from the configuration >>> model = CodeGenModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```codegen n_positionsn_embdn_headn_layer)max_position_embeddings hidden_sizenum_attention_headsnum_hidden_layersc ||_||_||_||_||_||_||_||_| |_| |_ | |_ | |_ | |_ ||_ ||_||_||_t#|Hd|||d|y)N) bos_token_id eos_token_idtie_word_embeddings) vocab_sizen_ctxrrrrn_inner rotary_dimactivation_function resid_pdrop embd_pdrop attn_pdroplayer_norm_epsiloninitializer_range use_cacherrsuper__init__)selfr rr!rrrr#r"r$r%r&r'r(r)r*rrrkwargs __class__s d/var/www/html/venv/lib/python3.12/site-packages/transformers/models/codegen/configuration_codegen.pyr,zCodeGenConfig.__init__fs,% &    $#6 &$$"4!2"((  %LVi ms )ir2i@Ngelu_newr7r7gh㈵>g{Gz?TPr8F)__name__ __module__ __qualname____doc__ model_type attribute_mapr, __classcell__r/s@r0rrse> @J#0'& M&!'+ + r1rceZdZ 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)CodeGenOnnxConfigconfigtaskpatching_specsuse_pastc~t|||||t|jddsd|j_yy)N)rDrErF pad_token_idr)r+r,getattr_configrH)r-rCrDrErFr/s r0r,zCodeGenOnnxConfig.__init__s= d>T\]t||^T:()DLL %;r1returnctddddi}|jr|j|dddd|d<|Sddd|d<|S) N input_idsbatchsequence)rinputs) directionzpast_sequence + sequenceattention_mask)rrFfill_with_past_key_values_)r- common_inputss r0rQzCodeGenOnnxConfig.inputssa#[g*2M$NO ==  + +MX + N29>X.YM* +3:j.IM* +r1c.|jjSN)rJrr-s r0 num_layerszCodeGenOnnxConfig.num_layerss||###r1c.|jjSrW)rJrrXs r0rz%CodeGenOnnxConfig.num_attention_headss||"""r1 tokenizer batch_size seq_lengthis_pair frameworkchtt| |||||}td|di}|jrt s t dddl}|dj\} } | dz} | |j| |jj|jzf} t|jD cgc]$} |j| |j| f&c} |d<|d|d<|jr<|dj}j!|d|j#  |gd |d<|Scc} w) N)r\r]r^r_rMzACannot generate dummy past_keys inputs without PyTorch installed.rpast_key_valuesrS)dtyperP)dim)r+r generate_dummy_inputsrrFr ValueErrortorchshaperrJrrangerYzerosrccatones)r-r[r\r]r^r_rUordered_inputsrgrNseqlenpast_key_values_length past_shape_ mask_dtyper/s r0rez'CodeGenOnnxConfig.generate_dummy_inputss^0$M *W`iN %k=3M%NO ==%' !dee -k : @ @ v)/!&,,*LL,,0H0HH  QVVZVeVePf5KLU[[,ekk*.EF501,99I+J'( =='(89??J/4yy 015::eE[cm:3nouv090N+ ,5s.)D/cy)N rrXs r0default_onnx_opsetz$CodeGenOnnxConfig.default_onnx_opsetsr1)defaultNF)rwFN)r9r:r;r strrrboolr,propertyrintrQrYrr rr rrerur?r@s@r0rBrBs-1 *  * *\* *  *WS#X%6 67$C$$#S## *. *&** *  * J' * c *XCr1rBN)r< collectionsrtypingrrrrr r r configuration_utilsr onnxr rutilsr get_loggerr9loggerrrBrr1r0rsU"#//CC34   H %t $t pN*Nr1