sg>dZddlmZmZmZddlmZddlmZm Z ddl m Z m Z m Z mZmZmZddlmZmZmZddlmZej.eZd Zed Dcgc] }d |d d  c}edDcgc] }d|dd  c}zZGddeZGdde ZGdde dZde fdZ!dZ"dZ#dZ$deeefdZ%Gdd e Z&y!cc}wcc}w)"z Processor class for PaliGemma. )ListOptionalUnion) BatchFeature) ImageInputis_valid_image) ImagesKwargsProcessingKwargsProcessorMixin TextKwargsUnpack!_validate_images_text_input_order) AddedTokenPreTokenizedInput TextInput)loggingziz4>z3c8eZdZUeeeeeeeefed<y)PaliGemmaTextKwargssuffixN) __name__ __module__ __qualname__rrrrr__annotations__e/var/www/html/venv/lib/python3.12/site-packages/transformers/models/paligemma/processing_paligemma.pyrr-s& U9&7i$O`Jaab ccrrceZdZUeeed<y)PaliGemmaImagesKwargsdo_convert_rgbN)rrrrboolrrrrr!r!1s TN"rr!c4eZdZUeed<eed<ddiddidZy) PaliGemmaProcessorKwargs text_kwargs images_kwargspaddingF data_formatchannels_first)r&r'N)rrrrrr! _defaultsrrrr%r%5s.$$(( u  + Irr%F)totalreturncHt|txr|jdS)Nhttp) isinstancestr startswith)vals ris_urlr4Cs c3  :CNN6$::rc2t|xs t|SN)r4r elems ris_image_or_image_urlr9Hs $< />$//rc<t|txs t|Sr6)r0r1r9r7s r_is_str_or_imager;Ls dS " A&;D&AArc ||z|z||dS)aZ Builds a string from the input prompt and image tokens. For example, for the call: build_string_from_input( prompt="Prefix str" bos_token="", image_seq_len=3, image_token="", ) The output will be: "Initial str" Args: prompt (`List[Union[str, ImageInput]]`): The input prompt. bos_token (`str`): The beginning of sentence token. image_seq_len (`int`): The length of the image sequence. image_token (`str`): The image token. num_images (`int`): Number of images in the prompt.  rprompt bos_token image_seq_len image_token num_imagess rbuild_string_from_inputrDPs$&M)J6 7 {6(" MMrcDt|ttfrCt|dttfr*t|ddr|Dcgc] }|D]}| c}}St|ttfrt|dr|St|r|gSt d|cc}}w)a  Accepts images in list or nested list format, and makes a list of images for preprocessing. Args: images (`Union[List[List[ImageInput]], List[ImageInput], ImageInput]`): The input image. Returns: list: A list of images. rz"Could not make batched video from )r0listtupler ValueError)imagesimg_listimgs rmake_batched_imagesrLgs&4-(Zq D%=-QVdeklmenopeqVr$*?h?s??? FT5M *~fQi/H  x 9&B CC@sBc eZdZdZddgZdgZdZdZ dfd Z dde d e e e e e e e fd eed efd Zd ZdZedZxZS)PaliGemmaProcessora Constructs a PaliGemma processor which wraps a PaliGemma image processor and a PaliGemma tokenizer into a single processor. [`PaliGemmaProcessor`] offers all the functionalities of [`SiglipImageProcessor`] and [`LlamaTokenizerFast`]. See the [`~PaliGemmaProcessor.__call__`] and [`~PaliGemmaProcessor.decode`] for more information. Args: image_processor ([`SiglipImageProcessor`], *optional*): The image processor is a required input. tokenizer ([`LlamaTokenizerFast`], *optional*): 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. image_processor tokenizer chat_templateSiglipImageProcessor)GemmaTokenizerGemmaTokenizerFastc | td| tdt|ds td|j|_t|dsCttdd}d |gi}|j ||j t|_n|j|_|jtd|_ d|_ t|5||| y) Nz)You need to specify an `image_processor`.z"You need to specify a `tokenizer`.image_seq_lengthz;Image processor is missing an `image_seq_length` attribute.rBFT) normalizedspecialadditional_special_tokens)rQ)rHhasattrrVr IMAGE_TOKENadd_special_tokensconvert_tokens_to_idsimage_token_id add_tokens EXTRA_TOKENS add_bos_token add_eos_tokensuper__init__)selfrOrPrQkwargsrB tokens_to_add __class__s rrdzPaliGemmaProcessor.__init__s  "HI I  AB B(:;Z[ [ / @ @y-0$[UDQK8;-HM  ( ( 7"+"A"A+"ND "+":":D \*"' "'  )=QrrItextrfr-c zt||\}}|jtfd|jji|}|dj dd}|dnd}| t d|tjdd }t|r|g}nt|trt|d r |T|Qtd |Dstjd t|trKt|tr;t|t|k7r$t d t|dt|dt!|r|gg}nnt|trt!|d r|D cgc]} | g}} n?t|tr$t|d trt!|d d s t d|t|r|g}|&|D cgc]} | |jj"z}} t%||D cgc]R\} } t'| |jj(|j*t,t| tr t| ndT} } } t/|}ng}|D]}|j1t,t,|j*z}|j3t,}|dk7r|tt,znd }|d||jj(z||dz}|j5||Dcgc]}|d } }|j6|fi|dd}|dj9dd|ddxx|j*z cc<|j f||d|d}i|d|i}|r.|dj;|dd k(d}|j=d|it?|Scc} wcc} wcc} } wcc}w)ah Main method to prepare for the model one or several sequences(s) and image(s). This method forwards the `text` and `kwargs` arguments to LlamaTokenizerFast's [`~LlamaTokenizerFast.__call__`] if `text` is not `None` to encode the text. To prepare the image(s), this method forwards the `images` and `kwrags` arguments to SiglipImageProcessor's [`~SiglipImageProcessor.__call__`] if `images` is not `None`. Please refer to the doctsring of the above two methods for more information. The usage for PaliGemma fine-tuning preparation is slightly different than usual. suffix passed are suffixes to the prompt in `text`, and will be placed after the prompt. This is because attention is handled differently for the prefix and the suffix. For instance, ```python image = PIL_cow_image prompt = "answer en Where is the cow standing?" suffix = "on the beach" inputs = processor(text=prompt, images=image, suffix=suffix) ``` Here `inputs` will contain the `input_ids` and `token_type_ids` that follow ```python inputs["input_ids"][:, 256:] # tensor([[ 2, 6006, 603, 573, 13910, 9980, 235336, 108, 477, 573, 8318]]) inputs["token_type_ids"][:, 256:] tensor([[0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1]]) ``` Meaning the last three tokens are of "label" ("suffix") type while the other ones are of "prefix" type. Args: images (`PIL.Image.Image`, `np.ndarray`, `torch.Tensor`, `List[PIL.Image.Image]`, `List[np.ndarray]`, `List[torch.Tensor]`): The image or batch of images to be prepared. Each image can be a PIL image, NumPy array or PyTorch tensor. In case of a NumPy array/PyTorch tensor, each image should be of shape (C, H, W), where C is a number of channels, H and W are image height and width. text (`str`, `List[str]`, `List[List[str]]`): 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). return_tensors (`str` or [`~utils.TensorType`], *optional*): If set, will return tensors of a particular framework. Acceptable values are: - `'tf'`: Return TensorFlow `tf.constant` objects. - `'pt'`: Return PyTorch `torch.Tensor` objects. - `'np'`: Return NumPy `np.ndarray` objects. - `'jax'`: Return JAX `jnp.ndarray` objects. suffix (`str`, `List[str]`, `List[List[str]]`): The suffixes or batch of suffixes to be encoded. Only necessary for finetuning. See https://github.com/google-research/big_vision/blob/main/big_vision/configs/proj/paligemma/README.md for more information. If your prompt is " What is on the image", the suffix corresponds to the expected prediction "a cow sitting on a bench". 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`. If `suffix` is provided, the `input_ids` will also contain the suffix input ids. - **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`. - **labels** -- Labels compatible with training if `suffix` is not None tokenizer_init_kwargsr&rNTFzF`images` are expected as arguments to a `PaliGemmaProcessor` instance.z]You are using PaliGemma without a text prefix. It will perform as a picture-captioning model.rc3,K|] }t|vywr6)r[).0samples r z.PaliGemmaProcessor.__call__.. s@{f,@saLYou are passing both `text` and `images` to `PaliGemmaProcessor`. The processor expects special image tokens in the text, as many tokens as there are images per each text. It is recommended to add `` tokens in the very beginning of your text. For this call, we will infer how many images each text has and add special tokens.z Received z images for zK prompts. Each prompt should be associated with an image or list of images.zAimages must be an image, list of images or list of list of imagesr>r=r' pixel_values max_length) text_pairreturn_token_type_ids input_idstoken_type_idsilabels)data) r _merge_kwargsr%rP init_kwargspoprHlogger warning_oncer;r0rFanywarningrlenr eos_tokenziprDr@rVr[rLreplacerfindappendrOget masked_fillupdater)rerIriaudiovideosrf output_kwargsrrvimagesfxr? image_list input_stringsexpanded_samplesroexpanded_samplebos_rfind_index bos_indexrsinputs return_datarys r__call__zPaliGemmaProcessor.__call__sD9F *** $ "&.."<"<  }-11(DA(.(: >ef f <   o D D !6D d #(8a(A    2@4@@<dD)j.F6{c$i/('F }LT LWX "&)%hZF-.2K39:%ug:F:$VT2z&)T7RWeflmnfopqfrWs$%hii%*:6*B$XF%HNOcDNN$<$<<OFO/2$.? !+ ,%"&..":":&*&;&;$/6@T6R3z?XY  ! !-V4#% "=F&,nn[+PTPePeBe&fO&5&;&;K&HOFUY[F[#k2B BabI' 3dnn6N6NNQ`ajakQll$%++O<=>N N6F82 N N+t++FUmO6TUVde   ' + +L$ ? K - ( 6$:O:O O 6  "7 M*  ?>> K(44VO R@"^b ^.^.I0$y/4HYCZZ[^. 12 ^. ^.B<6XXrrNN)'rtypingrrrfeature_extraction_utilsr image_utilsrr processing_utilsr r r r rrtokenization_utils_baserrrutilsr get_loggerrr~r[ranger`rr!r%r#r4r9r;rDrLrN)is0rrs)(45    H % ).t5A$qgQ5RWX[R\8]Q4#wa8]] d*d#L# /u ;4; 0BN.D4Z(8#9D.hXhXi68]s C3C