sgd+dZddlZddlmZmZmZddlmZddlm Z ddl m Z ddl m Z mZmZmZmZmZdd lmZmZd d lmZej0eZGd d e Zy)zq Processor class for InstructBLIP. Largely copy of Blip2Processor with addition of a tokenizer for the Q-Former. N)ListOptionalUnion) BatchFeature) VideoInput)ProcessorMixin) AddedToken BatchEncodingPaddingStrategyPreTokenizedInput TextInputTruncationStrategy) TensorTypelogging) AutoTokenizerc$:eZdZdZgdZdgZdZdZdZdfd Z dde de e e ee ee fd ed e eeefd e eeefd eed edeedeedededededededee eefdef"dZdZdZedZfdZefdZxZS) InstructBlipVideoProcessora Constructs an InstructBLIPVideo processor which wraps a InstructBLIP image processor and a LLaMa/T5 tokenizer into a single processor. [`InstructBlipVideoProcessor`] offers all the functionalities of [`InstructBlipVideoImageProcessor`] and [`AutoTokenizer`]. See the docstring of [`~InstructBlipVideoProcessor.__call__`] and [`~InstructBlipVideoProcessor.decode`] for more information. Args: image_processor (`InstructBlipVideoImageProcessor`): An instance of [`InstructBlipVideoImageProcessor`]. The image processor is a required input. tokenizer (`AutoTokenizer`): An instance of ['PreTrainedTokenizer`]. The tokenizer is a required input. qformer_tokenizer (`AutoTokenizer`): An instance of ['PreTrainedTokenizer`]. The Q-Former tokenizer is a required input. num_query_tokens (`int`, *optional*): Number of tokens used by the Qformer as queries, should be same as in model's config. )image_processor tokenizerqformer_tokenizernum_query_tokensInstructBlipVideoImageProcessorrc t|ds2tddd|_|j|jgdn|j|_||_t ||||y)N video_tokenz