sgjdZddlmZddlmZddlmZmZeje Z GddeZ y) zLlava model configuration)PretrainedConfig)logging)CONFIG_MAPPING AutoConfigcBeZdZdZdZeedZ dfd ZxZS) LlavaConfiga This is the configuration class to store the configuration of a [`LlavaForConditionalGeneration`]. It is used to instantiate an Llava 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 Llava-9B. e.g. [llava-hf/llava-9b](https://huggingface.co/llava-hf/llava-9b) 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 (`Union[AutoConfig, dict]`, *optional*, defaults to `CLIPVisionConfig`): The config object or dictionary of the vision backbone. text_config (`Union[AutoConfig, dict]`, *optional*, defaults to `LlamaConfig`): The config object or dictionary of the text backbone. ignore_index (`int`, *optional*, defaults to -100): The ignore index for the loss function. image_token_index (`int`, *optional*, defaults to 32000): The image token index to encode the image prompt. projector_hidden_act (`str`, *optional*, defaults to `"gelu"`): The activation function used by the multimodal projector. vision_feature_select_strategy (`str`, *optional*, defaults to `"default"`): The feature selection strategy used to select the vision feature from the vision backbone. Can be one of `"default"` or `"full"`. vision_feature_layer (`int`, *optional*, defaults to -2): The index of the layer to select the vision feature. image_seq_length (`int`, *optional*, defaults to 576): Sequence length of one image embedding. Example: ```python >>> from transformers import LlavaForConditionalGeneration, LlavaConfig, CLIPVisionConfig, LlamaConfig >>> # Initializing a CLIP-vision config >>> vision_config = CLIPVisionConfig() >>> # Initializing a Llama config >>> text_config = LlamaConfig() >>> # Initializing a Llava llava-1.5-7b style configuration >>> configuration = LlavaConfig(vision_config, text_config) >>> # Initializing a model from the llava-1.5-7b style configuration >>> model = LlavaForConditionalGeneration(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```llava) text_config vision_configc ||_||_||_||_|dvrt d|||_||_t|tr d|vr|dnd|d<t|ddi|}n|tdddddd d d d }||_ t|tr d|vr|dnd|d<t|ddi|}n| td}||_ t |4di| y)N)defaultfullzGvision_feature_select_strategy should be one of 'default', 'full'.Got: model_typeclip_vision_modeliiiP}i)intermediate_size hidden_size patch_size image_sizenum_hidden_layersnum_attention_heads vocab_sizeprojection_dimllama) ignore_indeximage_token_indexprojector_hidden_actimage_seq_length ValueErrorvision_feature_select_strategyvision_feature_layer isinstancedictrr r super__init__) selfr r r r!r"r%r&r#kwargs __class__s `/var/www/html/venv/lib/python3.12/site-packages/transformers/models/llava/configuration_llava.pyr*zLlavaConfig.__init__Ns;)!2$8! 0 )1D D679  /M+$8! mT */;}/L l+Re , '+=+FGX-XM  "*+>?"& "$$& " M+ k4 (EQU`E` L(AfmK %(\)BCRkRK  (13K& "6")NNirgeluri@) __name__ __module__ __qualname____doc__rr sub_configsr* __classcell__)r-s@r.r r s>0 dJ",zJK#'05#5#r/r N) r5configuration_utilsrutilsrautorr get_loggerr2loggerr rr/r.r=s6 3-   H %k#"k#r/