o i95 @s$dZddlmZmZddlZddlmZmZm Z ddl m Z m Z m Z mZddlmZmZmZmZmZmZmZmZddlmZmZmZmZmZmZdd lm Z erUddl!Z!er\ddl"Z"e#e$Z%d d Z&  dd ej'd ee(dee(deee(effddZ)e ddGdddeZ*dgZ+dS)z%Image processor class for LayoutLMv2.)OptionalUnionN)BaseImageProcessor BatchFeature get_size_dict)flip_channel_orderresizeto_channel_dimension_format to_pil_image)ChannelDimension ImageInputPILImageResamplinginfer_channel_dimension_formatmake_flat_list_of_imagesto_numpy_array valid_imagesvalidate_preprocess_arguments) TensorTypefilter_out_non_signature_kwargsis_pytesseract_availableis_vision_availableloggingrequires_backends)requirescCsLtd|d|td|d|td|d|td|d|gS)Nirr)int)boxwidthheightr!n/home/ubuntu/.local/lib/python3.10/site-packages/transformers/models/layoutlmv2/image_processing_layoutlmv2.py normalize_box6s r#imagelangtesseract_configinput_data_formatcsb|dur|nd}t||d}|j\}}tj||d|d}|d|d|d|d |d f\}} } } } d d t|Dfd d t|D}fdd t| D} fdd t| D} fdd t| D} fdd t| D} g} t| | | | D]\}}}}||||||g}| |q{g}| D] }|t|||qt|t|ksJd||fS)zdApplies Tesseract OCR on a document image, and returns recognized words + normalized bounding boxes.Nr'dict)r% output_typeconfigtextlefttoprr cSsg|] \}}|s|qSr!)strip.0idxwordr!r!r" Oz#apply_tesseract..cg|] \}}|vr|qSr!r!r1irrelevant_indicesr!r"r5Pr6cr7r!r!r2r3coordr8r!r"r5Qr6cr7r!r!r:r8r!r"r5Rr6cr7r!r!r:r8r!r"r5Sr6cr7r!r!r:r8r!r"r5Tr6z-Not as many words as there are bounding boxes) r size pytesseract image_to_data enumeratezipappendr#len)r$r%r&r' pil_image image_width image_heightdatawordsr.r/rr actual_boxesxywh actual_boxnormalized_boxesrr!r8r"apply_tesseract?s(  , rO)vision)backendscsFeZdZdZdgZddejdddfdedee e e fded ed ee d ee d dffd d Z ejddfde jde e e fdedeee efdeee efd e jf ddZedddddddejdf dedeedee e e fdeed eed ee d ee deee efdedeee efd ejjfddZZS)LayoutLMv2ImageProcessora Constructs a LayoutLMv2 image processor. Args: do_resize (`bool`, *optional*, defaults to `True`): Whether to resize the image's (height, width) dimensions to `(size["height"], size["width"])`. Can be overridden by `do_resize` in `preprocess`. size (`dict[str, int]` *optional*, defaults to `{"height": 224, "width": 224}`): Size of the image after resizing. Can be overridden by `size` in `preprocess`. resample (`PILImageResampling`, *optional*, defaults to `Resampling.BILINEAR`): Resampling filter to use if resizing the image. Can be overridden by the `resample` parameter in the `preprocess` method. apply_ocr (`bool`, *optional*, defaults to `True`): Whether to apply the Tesseract OCR engine to get words + normalized bounding boxes. Can be overridden by `apply_ocr` in `preprocess`. ocr_lang (`str`, *optional*): The language, specified by its ISO code, to be used by the Tesseract OCR engine. By default, English is used. Can be overridden by `ocr_lang` in `preprocess`. tesseract_config (`str`, *optional*, defaults to `""`): Any additional custom configuration flags that are forwarded to the `config` parameter when calling Tesseract. For example: '--psm 6'. Can be overridden by `tesseract_config` in `preprocess`. pixel_valuesTNr( do_resizer<resample apply_ocrocr_langr&returnc sXtjdi||dur|nddd}t|}||_||_||_||_||_||_dS)N)r rr!) super__init__rrTr<rUrVrWr&)selfrTr<rUrVrWr&kwargs __class__r!r"r[s  z!LayoutLMv2ImageProcessor.__init__r$ data_formatr'cKsTt|}d|vs d|vrtd||d|df}t|f||||d|S)a Resize an image to `(size["height"], size["width"])`. Args: image (`np.ndarray`): Image to resize. size (`dict[str, int]`): Dictionary in the format `{"height": int, "width": int}` specifying the size of the output image. resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BILINEAR`): `PILImageResampling` filter to use when resizing the image e.g. `PILImageResampling.BILINEAR`. data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format for the output image. If unset, the channel dimension format of the input image is used. Can be one of: - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format. - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format. - `"none"` or `ChannelDimension.NONE`: image in (height, width) format. input_data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format for the input image. If unset, the channel dimension format is inferred from the input image. Can be one of: - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format. - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format. - `"none"` or `ChannelDimension.NONE`: image in (height, width) format. Returns: `np.ndarray`: The resized image. r rzFThe `size` dictionary must contain the keys `height` and `width`. Got )r<rUr`r')r ValueErrorkeysr )r\r$r<rUr`r'r] output_sizer!r!r"r s#zLayoutLMv2ImageProcessor.resizeimagesreturn_tensorsc sn|dur|nj}durnjtdurnj|dur%|nj}|dur.|nj}|dur7|nj}t|}t|sFt dt |ddd|D}dur^t |d|rt dg} g} |D]} t | ||d\}}| || |qk|rfd d|D}fd d|D}fd d|D}td |i|d }|r| |d<| |d<|S)a Preprocess an image or batch of images. Args: images (`ImageInput`): Image to preprocess. do_resize (`bool`, *optional*, defaults to `self.do_resize`): Whether to resize the image. size (`dict[str, int]`, *optional*, defaults to `self.size`): Desired size of the output image after resizing. resample (`PILImageResampling`, *optional*, defaults to `self.resample`): Resampling filter to use if resizing the image. This can be one of the enum `PIL.Image` resampling filter. Only has an effect if `do_resize` is set to `True`. apply_ocr (`bool`, *optional*, defaults to `self.apply_ocr`): Whether to apply the Tesseract OCR engine to get words + normalized bounding boxes. ocr_lang (`str`, *optional*, defaults to `self.ocr_lang`): The language, specified by its ISO code, to be used by the Tesseract OCR engine. By default, English is used. tesseract_config (`str`, *optional*, defaults to `self.tesseract_config`): Any additional custom configuration flags that are forwarded to the `config` parameter when calling Tesseract. return_tensors (`str` or `TensorType`, *optional*): The type of tensors to return. Can be one of: - Unset: Return a list of `np.ndarray`. - `TensorType.TENSORFLOW` or `'tf'`: Return a batch of type `tf.Tensor`. - `TensorType.PYTORCH` or `'pt'`: Return a batch of type `torch.Tensor`. - `TensorType.NUMPY` or `'np'`: Return a batch of type `np.ndarray`. - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`. data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`): The channel dimension format for the output image. Can be one of: - `ChannelDimension.FIRST`: image in (num_channels, height, width) format. - `ChannelDimension.LAST`: image in (height, width, num_channels) format. NzkInvalid image type. Must be of type PIL.Image.Image, numpy.ndarray, torch.Tensor, tf.Tensor or jax.ndarray.)rTr<rUcSsg|]}t|qSr!)rr2r$r!r!r"r5 sz7LayoutLMv2ImageProcessor.preprocess..rr=r)csg|] }j|dqS))r$r<rUr')r rf)r'rUr\r<r!r"r5scsg|]}t|dqS)r))rrfr)r!r"r5"scsg|] }t|dqS))input_channel_dim)r rf)r`r'r!r"r5#srS)rF tensor_typerGboxes)rTr<rrUrVrWr&rrrarrrrOrAr)r\rdrTr<rUrVrWr&rer`r' words_batch boxes_batchr$rGrirFr!)r`r'rUr\r<r" preprocesssR/     z#LayoutLMv2ImageProcessor.preprocess)__name__ __module__ __qualname____doc__model_input_namesrBILINEARboolrr*strrr[npndarrayrr r rFIRSTr rPILImagerl __classcell__r!r!r^r"rRfs   0    rR)NN),rptypingrrnumpyruimage_processing_utilsrrrimage_transformsrr r r image_utilsr r rrrrrrutilsrrrrrrutils.import_utilsrrxr= get_loggerrmloggerr#rvrtrOrR__all__r!r!r!r"s:(    ' I