o wi5@sdZddlmZddlZddlmZddlmZm Z m Z ddl m Z m Z mZmZmZddlmZmZdd lmZmZerCd d lmZeeZGd d d e ddZdefddZddZGdddeZ dgZ!dS)z Processor class for Pixtral. )UnionN) BatchFeature) ImageInputis_valid_image load_image)MultiModalDataProcessingKwargsProcessorMixinUnpack!_validate_images_text_input_order)PreTokenizedInput TextInput)is_vision_availablelogging)get_resize_output_image_sizec@s"eZdZdddiddidZdS)PixtralProcessorKwargsF)paddingreturn_mm_token_type_idsreturn_tensorspt) text_kwargs images_kwargs common_kwargsN)__name__ __module__ __qualname__ _defaultsrrk/home/ubuntu/sommelier/.venv/lib/python3.10/site-packages/transformers/models/pixtral/processing_pixtral.pyr+s rF)totalreturncCst|to |dS)Nhttp) isinstancestr startswith)valrrr is_url9sr(cCst|pt|SN)r(r)elemrrr is_image_or_image_url>sr+c seZdZdZddgZdZdZ      dd ed effdd Z    dde de e e e e e e fdeedefddZd ddZddZddZeddZZS)!PixtralProcessorab Constructs a Pixtral processor which wraps a Pixtral image processor and a Pixtral tokenizer into a single processor. [`PixtralProcessor`] offers all the functionalities of [`CLIPImageProcessor`] and [`LlamaTokenizerFast`]. See the [`~PixtralProcessor.__call__`] and [`~PixtralProcessor.decode`] for more information. Args: image_processor ([`PixtralImageProcessor`], *optional*): The image processor is a required input. tokenizer ([`LlamaTokenizerFast`], *optional*): The tokenizer is a required input. patch_size (`int`, *optional*, defaults to 16): Patch size from the vision tower. spatial_merge_size (`int`, *optional*, defaults to 1): The downsampling factor for the spatial merge operation. chat_template (`str`, *optional*): A Jinja template which will be used to convert lists of messages in a chat into a tokenizable string. image_token (`str`, *optional*, defaults to `"[IMG]"`): Special token used to denote image location. image_break_token (`str`, *optional*, defaults to `"[IMG_BREAK]"`): Special token used to denote the end of a line of pixels in an image. image_end_token (`str`, *optional*, defaults to `"[IMG_END]"`): Special token used to denote the end of an image input. image_processor tokenizerAutoImageProcessor AutoTokenizerNr[IMG] [IMG_BREAK] [IMG_END] patch_sizespatial_merge_sizec  s~||_||_||_||j|_||_||_||j|_||j|_||j|_|j|j|jg|_ t j |||ddS)N) chat_template) r5r6 image_tokenconvert_tokens_to_idsimage_token_idimage_break_tokenimage_end_tokenimage_break_token_idimage_end_token_id image_idssuper__init__) selfr-r.r5r6r7r8r;r<kwargs __class__rr rA`s zPixtralProcessor.__init__imagestextrCr"cKst||\}}|jtfd|jji|}|j|j}|durmt|r&|g}n2t|t t fr4t|dr4n$t|t t frTt|dt t frTt|ddrTdd|D}nt ddd|D}|j |fd|i|d }ni}t|t rx|g}nt|t st|dt st d |} |d durt|d } g} g} |D]^} |j| vrt| \} }| |}||}|jg||jgg|}d d|D}|j|d<d|}| || |jdd} |j| vsd| vr| d}| d|d} d| vs| | q|ddd}|ddd}|j| fi|dddi}|j| |dgd|rDt|d}t|d}d|t||j<||d<ti|||dS)a 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 CLIPImageProcessor's [`~CLIPImageProcessor.__call__`] if `images` is not `None`. Please refer to the docstring of the above two methods for more information. 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. Both channels-first and channels-last formats are supported. 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. 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`. tokenizer_init_kwargsNrcSg|] }|D]}|qqSrr).0sublistimagerrr z-PixtralProcessor.__call__..zdInvalid input images. Please provide a single image, a list of images, or a list of lists of images.cSs"g|] }t|tr t|n|qSr)r$r%r)rJimrrr rMs"r5rzAInvalid input text. Please provide a string, or a list of strings pixel_values image_sizescSrIrr)rJrKitemrrr rMrNz rrrrFrL) modalities input_idsmm_token_type_ids)data tensor_type) r _merge_kwargsrr. init_kwargsr5r6r+r$listtuple ValueErrorr-r%getiterr8nextr;r<joinappendreplacepop_check_special_mm_tokensnparray zeros_likeisinr?tolistr)rBrFrGaudiovideosrC output_kwargsr5 image_inputsprompt_stringsrQreplace_stringssampleheightwidthnum_height_tokensnum_width_tokensreplace_tokens replace_strrr text_inputs array_idsrWrrr __call__xs)             zPixtralProcessor.__call__cKsi}|durbtjdi}|||ddp|jj}|j|j}g}|D],\}} tt || df|d|df||fd\} } | |} | |} | | d| q&dgt |}|||dt d i|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. Nrsizer longest_edge)r|r5r)num_image_tokensnum_image_patchesr)rrr_updater-r|r5r6rrgzerosrclenr)rBrQrC vision_datarr|r5r~rsrtresized_height resized_widthrurvrrrr _get_num_multimodal_tokenss&     z+PixtralProcessor._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_decoderBargsrCrrr rzPixtralProcessor.batch_decodecOr)z This method forwards all its arguments to LlamaTokenizerFast's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this method for more information. )r.decoderrrr rrzPixtralProcessor.decodecCs"|jj}|jj}tt||Sr))r.model_input_namesr-r\dictfromkeys)rBtokenizer_input_namesimage_processor_input_namesrrr r!sz"PixtralProcessor.model_input_names)NNr1rNr2r3r4)NNNNr))rrr__doc__ attributesimage_processor_classtokenizer_classintrArrrr r\r rrr{rrrpropertyr __classcell__rrrDr r,BsH u%r,)"rtypingrnumpyrgfeature_extraction_utilsr image_utilsrrrprocessing_utilsrr r r r tokenization_utils_baser rutilsrrimage_processing_pixtralr get_loggerrloggerrboolr(r+r,__all__rrrr s"     h