o wiT+@sdZddlZddlmZmZddlmZddlmZddl m Z m Z m Z m Z mZmZddlmZmZdd lmZd d lmZeeZGd d d eZd gZdS)zq Processor class for InstructBLIP. Largely copy of Blip2Processor with addition of a tokenizer for the Q-Former. N)OptionalUnion) BatchFeature)ProcessorMixin) AddedToken BatchEncodingPaddingStrategyPreTokenizedInput TextInputTruncationStrategy) TensorTypelogging) VideoInput) AutoTokenizerc$seZdZdZgdZdZdZdZd(fdd Z         d)d e d e e e e e e e fd ede eeefde eeefdeededeedeedededededededee eefdef"ddZddZd d!Zed"d#Zfd$d%Zefd&d'ZZS)*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: video_processor (`InstructBlipVideoVideoProcessor`): An instance of [`InstructBlipVideoVideoProcessor`]. The video 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. )video_processor tokenizerqformer_tokenizerAutoVideoProcessorrNc sPt|dstdddd|_|j|jgddn|j|_||_t|||dS)N video_tokenz