o wi&@sddlmZmZddlZddlmZddlmZddl m Z m Z m Z m Z ddlmZmZddlmZd d lmZGd d d e d dZGddde ZdgZdS))OptionalUnionN) BatchFeature) ImageInput)MultiModalDataProcessingKwargsProcessorMixinUnpack)PreTokenizedInput TextInput) TensorType) AutoTokenizerc@s&eZdZddddddejdZdS)AriaProcessorKwargsF)paddingreturn_mm_token_type_ids)max_image_size split_image) text_kwargs images_kwargsreturn_tensorsN)__name__ __module__ __qualname__r PYTORCH _defaultsrre/home/ubuntu/sommelier/.venv/lib/python3.10/site-packages/transformers/models/aria/processing_aria.pyr!s rF)totalc seZdZdZddgZdZdZ    ddeee fde e de e ee e fe fffd d Z   dd eeeeeeefd e ed eedefddZdddZddZddZeddZZS) AriaProcessora AriaProcessor is a processor for the Aria model which wraps the Aria image preprocessor and the LLama slow tokenizer. Args: image_processor (`AriaImageProcessor`, *optional*): The AriaImageProcessor to use for image preprocessing. tokenizer (`PreTrainedTokenizerBase`, *optional*): An instance of [`PreTrainedTokenizerBase`]. This should correspond with the model's text model. The tokenizer is a required input. chat_template (`str`, *optional*): A Jinja template which will be used to convert lists of messages in a chat into a tokenizable string. size_conversion (`Dict`, *optional*): A dictionary indicating size conversions for images. image_processor tokenizerAriaImageProcessorrN chat_templatesize_conversioncsf|dur ddd}dd|D|_|j|_|j|_|dur(|jdur(|j|_tj|||ddS)N)ircSsi|] \}}t||qSr)int).0kvrrr Kz*AriaProcessor.__init__..)r%)itemsr& image_tokenimage_token_id pad_token unk_tokensuper__init__)selfr"r#r%r& __class__rrr5Bs zAriaProcessor.__init__textimageskwargsreturncKsb|jtfd|jji|}t|tr|g}nt|ts&t|dts&td|dur^|j|fi|d}|j |j j d}g} | d|} |D]} | |jj|jj| } | | qIni}|} |d d d} |d d d } |j| fi|dd di}|j| |d gd | rt|d}t|d}d|||jk<||d<ti||| dS)a Main method to prepare for the model one or several sequences(s) and image(s). Args: text (`TextInput`, `PreTokenizedInput`, `list[TextInput]`, `list[PreTokenizedInput]`): The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings (pretokenized string). If the sequences are provided as list of strings (pretokenized), you must set `is_split_into_words=True` (to lift the ambiguity with a batch of sequences). images (`ImageInput`): The image or batch of images to be prepared. Each image can be a PIL image, NumPy array or PyTorch tensor. Both channels-first and channels-last formats are supported. 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`. - **pixel_mask** -- Pixel mask to be fed to a model. Returned when `images` is not `None`. tokenizer_init_kwargsrzAInvalid input text. Please provide a string, or a list of stringsNrr num_cropsrrrFimage) modalities input_idsmm_token_type_ids)data tensor_type) _merge_kwargsrr# init_kwargs isinstancestrlist ValueErrorr"r& pixel_valuesshapepopreplacer0append_check_special_mm_tokensnparray zeros_liker1tolistr)r6r9r:audiovideosr; output_kwargs image_inputstokens_per_imageprompt_stringsr>samplerr text_inputs array_idsrCrrr__call__Ts@   zAriaProcessor.__call__c s~i}|dur8tjdi|ddpjjfdd|D}fdd|D}|||dtdi|S) a Computes the number of placeholder tokens needed for multimodal inputs with the given sizes. Args: image_sizes (`list[list[int]]`, *optional*): The input sizes formatted as (height, width) per each image. Returns: `MultiModalData`: A `MultiModalData` object holding number of tokens per each of the provided input modalities, along with other useful data. Nrrcs"g|] }jjg|RqSr)r"get_number_of_image_patches)r* image_size)rr6rr sz.csg|] }j|qSr)r&)r* num_patches)max_sizer6rrrbr.)num_image_tokensnum_image_patchesr)rrgetupdater"rr)r6 image_sizesr; vision_datarfrer)rrdr6r_get_num_multimodal_tokenss   z(AriaProcessor._get_num_multimodal_tokenscO|jj|i|S)z This method forwards all its arguments to LlamaTokenizerFast's [`~PreTrainedTokenizer.batch_decode`]. Please refer to the docstring of this method for more information. )r# batch_decoder6argsr;rrrrmzAriaProcessor.batch_decodecOrl)z This method forwards all its arguments to LlamaTokenizerFast's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this method for more information. )r#decodernrrrrqrpzAriaProcessor.decodecCs0|jj}|jj}dd|D}tt||S)NcSsg|]}|dkr|qS)r>r)r*namerrrrbsz3AriaProcessor.model_input_names..)r#model_input_namesr"rJdictfromkeys)r6tokenizer_input_namesimage_processor_input_namesrrrrsszAriaProcessor.model_input_names)NNNN)NNN)N)rrr__doc__ attributesimage_processor_classtokenizer_classrrrIrrtfloatr)r5r r rJrr rrr_rkrmrqpropertyrs __classcell__rrr7rr!/sB  Dr!)typingrrnumpyrRimage_processing_utilsr image_utilsrprocessing_utilsrrr r tokenization_utilsr r utilsr autorrr!__all__rrrrs