o ei\@sFdZddlmZddlmZddlmZeGdddeZdgZdS)z( Image/Text processor class for CLIPSeg )ProcessorMixin) BatchEncoding)auto_docstringcs,eZdZdfdd ZedddZZS)CLIPSegProcessorNc st||dS)N)super__init__)selfimage_processor tokenizerkwargs __class__l/home/ubuntu/transcripts/venv/lib/python3.10/site-packages/transformers/models/clipseg/processing_clipseg.pyrszCLIPSegProcessor.__init__c Ks2|dur|dur|durtd|dur|durtd|j|jfd|jji|}|dur;|j|fd|i|d}|durL|j|fd|i|d}|dur]|j|fd|i|d} |durn|durn| j|jd}|S|dur}|dur}| j|d <|S|dur|S|durd |ji}|Sttd i| |d S) a visual_prompt (`PIL.Image.Image`, `np.ndarray`, `torch.Tensor`, `list[PIL.Image.Image]`, `list[np.ndarray]`, `list[torch.Tensor]`): The visual prompt image or batch of images to be prepared. Each visual prompt 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. Returns: [`BatchEncoding`]: A [`BatchEncoding`] 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`. Nz9You have to specify either text, visual prompt or images.zMYou have to specify exactly one type of prompt. Either text or visual prompt.tokenizer_init_kwargsreturn_tensors text_kwargs images_kwargs) pixel_valuesconditional_pixel_valuesrr)data tensor_typer) ValueError _merge_kwargsvalid_processor_kwargsr init_kwargsr rrdict) rtextimages visual_promptrr output_kwargsencodingprompt_featuresimage_featuresrrr__call__sX zCLIPSegProcessor.__call__)NN)NNNN)__name__ __module__ __qualname__rrr$ __classcell__rrr rrsrN) __doc__processing_utilsrtokenization_utils_baserutilsrr__all__rrrrs    @