o ei@sddlmZddlmZddlmZmZmZddlm Z m Z ddl m Z m Z e eZGdddedd Ze Gd d d eZd gZd S) ) BatchFeature) ImageInput)ProcessingKwargsProcessorMixinUnpack)PreTokenizedInput TextInput)auto_docstringloggingc@seZdZddiidZdS)Ovis2ProcessorKwargspaddingF) text_kwargs image_kwargsN)__name__ __module__ __qualname__ _defaultsrrh/home/ubuntu/transcripts/venv/lib/python3.10/site-packages/transformers/models/ovis2/processing_ovis2.pyr s  r F)totalc seZdZ     dfdd Ze  ddedBdeeBeeBeeBde e d e fd d Z deed eee fd dZddZddZeddZZS)Ovis2ProcessorNc sZ||_t|dr |jn||_t|ddr|jn||j|_tj||fd|i|dS)a image_token (`str`, *optional*, defaults to `""`): Special token used to denote image location. image_seq_length (`int`, *optional*, defaults to 256): The number of image tokens to be used for each image in the input. image_tokenimage_token_idN chat_template)image_seq_lengthhasattrrgetattrrconvert_tokens_to_idssuper__init__)selfimage_processor tokenizerrrrkwargs __class__rrr!%s  zOvis2Processor.__init__imagestextr%returncKs|jtfd|jji|}t|tr|g}nt|ts&t|dts&tdi}|durD|j|fi|d}| d }| ||}|j|fi|d}t i||dS) a Returns: [`BatchFeature`]: A [`BatchFeature`] with the following fields: - **input_ids** -- List of token ids to be fed to a model. Returned when `text` is not `None`. - **attention_mask** -- List of indices specifying which tokens should be attended to by the model (when `return_attention_mask=True` or if *"attention_mask"* is in `self.model_input_names` and if `text` is not `None`). - **pixel_values** -- Pixel values to be fed to a model. Returned when `images` is not `None`. - **image_sizes** -- Size of each image that will be used to unpad an image. Returned when `images` is not `None`. tokenizer_init_kwargszAInvalid input text. Please provide a string, or a list of stringsN images_kwargsgridsr )data) _merge_kwargsr r$ init_kwargs isinstancestrlist TypeErrorr#poptolist_expand_image_tokensr)r"r(r)r% output_kwargs image_inputs image_grids text_inputsrrr__call__=s$  zOvis2Processor.__call__r.c Csg}d}|D]d}d|vre||}|d|d}}dd|jd} ||dkrRt|D]%} t|D]} | d|j7} | |dkrF| d7} q2| |dkrQ| d7} q,| d 7} |d| d}|d7}d|vs ||q|S) Nr,rz z z z z z )rrangereplaceappend) r"r)r.processed_text grid_indexsamplegridrowcol placeholderrcrrrr8es.      z#Ovis2Processor._expand_image_tokenscO|jj|i|S)z This method forwards all its arguments to Qwen2TokenizerFast's [`~PreTrainedTokenizer.batch_decode`]. Please refer to the docstring of this method for more information. )r$ batch_decoder"argsr%rrrrLzOvis2Processor.batch_decodecOrK)z This method forwards all its arguments to Qwen2TokenizerFast's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this method for more information. )r$decoderMrrrrPrOzOvis2Processor.decodecCs |jj}|jj}t|t|S)N)r$model_input_namesr#r4)r"tokenizer_input_namesimage_processor_input_namesrrrrQsz Ovis2Processor.model_input_names)NNNrr)NN)rrrr!r rrrr4rr rr=intr8rLrPpropertyrQ __classcell__rrr&rr#s8 '  rN)feature_extraction_utilsr image_utilsrprocessing_utilsrrrtokenization_utils_baserrutilsr r get_loggerrloggerr r__all__rrrrs     q