`i."dZddlmZddlmZmZmZddlmZddl m Z m Z m Z ddl mZmZmZmZddlmZmZdd lmZerdd lmZejeZd efd Zd ZGddedZGddedZGddeZ dgZ!dS)z Processor class for IDEFICS2. ) accumulate) TYPE_CHECKINGOptionalUnion) BatchFeature) ImageInputis_valid_image load_image) ImagesKwargsProcessingKwargsProcessorMixinUnpack) AddedToken TextInput)logging)PreTokenizedInputreturncVt|to|dS)Nhttp) isinstancestr startswith)vals /home/jaya/work/projects/VOICE-AGENT/VIET/agent-env/lib/python3.11/site-packages/transformers/models/idefics2/processing_idefics2.pyis_urlr)s# c3   :CNN6$:$::c>t|pt|SN)rr )elems ris_image_or_image_urlr!-s $<< />$///rc&eZdZUeeed<dS)Idefics2ImagesKwargs image_seq_lenN)__name__ __module__ __qualname__rint__annotations__rrr#r#1s"C=     rr#F)totalc,eZdZUeed<ddddidZdS)Idefics2ProcessorKwargs images_kwargsTF)add_special_tokenspaddingis_split_into_words) text_kwargsr.N)r%r&r'r#r) _defaultsr*rrr-r-5sB''''#'#(   IIIrr-c eZdZdZddgZdZdZ dded ee ffd Z d Z dd e e ee eee fd e edeeedfdeedefdZxZS)Idefics2Processora Constructs a IDEFICS2 processor which wraps a LLama tokenizer and IDEFICS2 image processor into a single processor. [`IdeficsProcessor`] offers all the functionalities of [`Idefics2ImageProcessor`] and [`LlamaTokenizerFast`]. See the docstring of [`~IdeficsProcessor.__call__`] and [`~IdeficsProcessor.decode`] for more information. Args: image_processor (`Idefics2ImageProcessor`): An instance of [`Idefics2ImageProcessor`]. The image processor is a required input. tokenizer (`PreTrainedTokenizerBase`, *optional*): An instance of [`PreTrainedTokenizerBase`]. This should correspond with the model's text model. The tokenizer is a required input. image_seq_len (`int`, *optional*, defaults to 64): The length of the image sequence i.e. the number of tokens per image in the input. This parameter is used to build the string from the input prompt and image tokens and should match the config.perceiver_config.resampler_n_latents value for the model used. chat_template (`str`, *optional*): A Jinja template which will be used to convert lists of messages in a chat into a tokenizable string. image_processor tokenizerIdefics2ImageProcessor AutoTokenizerN@r$ chat_templatec (t|ds}tdddj|_tdddj|_d|j|jgi}||||j|_n$|j|_|j|_|j|_tddd|_ |d|j gi||_ t ||| dS) N image_tokenzFT) normalizedspecialzadditional_special_tokensz)r;) hasattrrcontentfake_image_tokenr=r/convert_tokens_to_idsimage_token_idimage_boundary_tokenend_of_utterance_tokenr$super__init__)selfr6r7r$r;kwargs tokens_to_add __class__s rrIzIdefics2Processor.__init__Zsy-00 ;$./JW\fj$k$k$k$sD !))tTTT\D 84;PRVRb:cdM  ( ( 7 7 7"+"A"A$BR"S"SD  $-$BD !(4D "+":D &01ERWae&f&f&f#$$&ADD_C`%abbb* )=QQQQQrcg}|D]t}g}|D]X}t|r||'t|r"|t|Y||u|Sr)r appendrr )rJprompts prompt_imagespromptimagesr s r_extract_images_from_promptsz.Idefics2Processor._extract_images_from_promptsns  ) )FF 4 4!$''4MM$''''D\\4MM*T"2"2333   ( ( ( (rrStextrrKrc |td|jtfd|jji|}|ddd}||n|j}|ddd}g} i} |>t|tr|g}n?t|ts*t|dtstd |j } |j } | | |z| } |j j r | d z} |d z}g}|D]p}| || || | }|| | | }||q|j|fi|d}|||d g | |t'rggn?ttt(frt'dr|t+| t-kr6td | dt+| d| dt-d dgtt/| zfdt1t-| Dnlgnhttt(fsLtdtt(fs*t'ddstddD}||| kstd| d|ddD|j fi|d}| |t3| |S)a Processes the input prompts and returns a BatchEncoding. Example: ```python >>> import requests >>> from transformers import Idefics2Processor >>> from transformers.image_utils import load_image >>> processor = Idefics2Processor.from_pretrained("HuggingFaceM4/idefics2-8b", image_seq_len=2) >>> processor.image_processor.do_image_splitting = False # Force as False to simplify the example >>> url1 = "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" >>> url2 = "https://cdn.britannica.com/59/94459-050-DBA42467/Skyline-Chicago.jpg" >>> image1, image2 = load_image(url1), load_image(url2) >>> images = [[image1], [image2]] >>> text = [ ... "In this image, we see", ... "bla bla bla", ... ] >>> outputs = processor(images=images, text=text, return_tensors="pt", padding=True) >>> input_ids = outputs.input_ids >>> input_tokens = processor.tokenizer.batch_decode(input_ids) >>> print(input_tokens) [' In this image, we see', ' bla bla bla'] ``` Args: images (`PIL.Image.Image`, `np.ndarray`, `torch.Tensor`, `list[PIL.Image.Image]`, `list[np.ndarray]`, `list[torch.Tensor]`, *optional*): The image or batch of images to be prepared. Each image can be a PIL image, NumPy array or PyTorch tensor. If is of type `list[ImageInput]`, it's assumed that this is for a single prompt i.e. of batch size 1. text (`Union[TextInput, PreTokenizedInput, list[TextInput], list[PreTokenizedInput]]`, *optional*): 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). Wherever an image token, `` is encountered it is expanded to `` + `` * `image_seq_len` * `. return_tensors (`Union[str, TensorType]`, *optional*): If set, will return tensors of a particular framework. See [`PreTrainedTokenizerFast.__call__`] for more information. Nz+You must provide either `text` or `images`.tokenizer_init_kwargsr.r$r2return_tensorsrzAInvalid input text. Please provide a string, or a list of stringsimage) modalitieszThe total number of zP tokens in the prompts should be the same as the number of images passed. Found  z tokens and z images.cBg|]}||dzS)r*).0icumsum_images_in_textrSs r z.Idefics2Processor.__call__..sF4Q7:OPQTUPU:VVWrzdInvalid input images. Please provide a single image or a list of images or a list of list of images.c,g|]}t|Sr*)lenr_samples rrbz.Idefics2Processor.__call__..s!C!C!C&#f++!C!C!Crz!The number of images in the text z and images z should be the same.c&g|]}d|DS)c,g|]}t|Sr*)r )r_ims rrbz9Idefics2Processor.__call__...s777"z"~~777rr*res rrbz.Idefics2Processor.__call__..s'MMMF77777MMMr) tensor_type) ValueError _merge_kwargsr-r7 init_kwargspopr$rrlistrCr=r6do_image_splittingrOcountreplace_check_special_mm_tokensupdater!tuplesumrdrranger)rJrSrUaudiovideosrK output_kwargsr$rXn_images_in_textinputsrCr= image_strprompt_stringsrf text_inputsn_images_in_images image_inputsras ` @r__call__zIdefics2Processor.__call__zsql ImmWZ[aWbWbmmm ./C$zBR7S7S2T2T,T)!&s+;'<'N(N(N 8HWi NMfMMMF/4/YY-:XYYL MM, ' ' 'F????r)Nr:N)NNNN)r%r&r'__doc__ attributesimage_processor_classtokenizer_classr(rrrIrTrr rorrr-rr __classcell__)rMs@rr5r5Bs;&$[1J4%OhlRR>ARW_`cWdRRRRRR(   OSbf H@H@j$z"2Dj9I4JJKH@I2DOTJ]E^^_H@ 01 H@ H@H@H@H@H@H@H@H@rr5N)"r itertoolsrtypingrrrfeature_extraction_utilsr image_utilsr r r processing_utilsr r rrtokenization_utils_baserrutilsrr get_loggerr%loggerboolrr!r#r-r5__all__r*rrrs! 1111111111444444AAAAAAAAAA =<<<<<<<=<<<<<<  H % %;4;;;;000!!!!!