o ei@sPdZddlmZddlmZddlmZmZee Z GdddeZ dgZ dS) zPaliGemmamodel configuration)PreTrainedConfig)logging)CONFIG_MAPPING AutoConfigcsVeZdZdZdZddiZeedZdgZ    dd e dBffd d Z Z S)PaliGemmaConfiga This is the configuration class to store the configuration of a [`PaliGemmaForConditionalGeneration`]. It is used to instantiate an PaliGemmamodel according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the PaliGemma-2B. e.g. [paligemma-hf/paligemma-2b](https://huggingface.co/paligemma-hf/paligemma-2b) Configuration objects inherit from [`PreTrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PreTrainedConfig`] for more information. Args: vision_config (`PaliGemmaVisionConfig`, *optional*): Custom vision config or dict text_config (`Union[AutoConfig, dict]`, *optional*): The config object of the text backbone. Can be any of `LlamaConfig` or `MistralConfig`. image_token_index (`int`, *optional*, defaults to 256000): The image token index to encode the image prompt. vocab_size (`int`, *optional*, defaults to 257152): Vocabulary size of the PaliGemmamodel. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`~PaliGemmaForConditionalGeneration`] projection_dim (`int`, *optional*, defaults to 2048): Dimension of the multimodal projection space. hidden_size (`int`, *optional*, defaults to 2048): Dimension of the hidden layer of the Language model. tie_word_embeddings (`bool`, *optional*, defaults to `True`): Whether to tie weight embeddings Example: ```python >>> from transformers import PaliGemmaForConditionalGeneration, PaliGemmaConfig, SiglipVisionConfig, GemmaConfig >>> # Initializing a Siglip-like vision config >>> vision_config = SiglipVisionConfig() >>> # Initializing a PaliGemma config >>> text_config = GemmaConfig() >>> # Initializing a PaliGemma paligemma-3b-224 style configuration >>> configuration = PaliGemmaConfig(vision_config, text_config) >>> # Initializing a model from the paligemma-3b-224 style configuration >>> model = PaliGemmaForConditionalGeneration(configuration) >>> # Accessing the model configuration >>> configuration = model.config ``` paligemmaimage_token_idimage_token_index) text_config vision_configpast_key_valuesNTtie_word_embeddingsc s*||_||_||_||_||_d|_t|jtr-|dd|d<t |ddi||_n|dur@t ddddddd d dd |_||_ t|j tr^|dd |d<t |ddi||_ n|durpt d d ddddd|d|_ |j j durzd|j _ |jj |jj d|j _||j_tjdi|dS)NF model_typesiglip_vision_modeliir)intermediate_size hidden_size patch_size image_sizenum_hidden_layersnum_attention_heads vocab_sizevision_use_headgemmari@)rrrrnum_key_value_headsis_encoder_decoderrTr)r projection_dimrr rr% isinstancedictgetrr use_bidirectional_attentionrrnum_image_tokenssuper__init__) selfr r r rr'rrkwargs __class__r&s/home/ubuntu/transcripts/venv/lib/python3.10/site-packages/transformers/models/paligemma/configuration_paligemma.pyr.OsN    zPaliGemmaConfig.__init__)NNrrrrT) __name__ __module__ __qualname____doc__r attribute_mapr sub_configskeys_to_ignore_at_inferenceboolr. __classcell__r&r&r1r3rs"0 rN) r7configuration_utilsrutilsrautorr get_loggerr4loggerr__all__r&r&r&r3s    r