`iV0dZddlmZmZddlZddlmZmZm Z ddl m Z m Z m Z mZmZddlmZmZmZmZmZmZmZmZmZmZmZmZddlmZmZm Z dd l!m"Z"m#Z#e j$e%Z&e"rddl'Z'e#d Gd d eZ(d gZ)dS)z Image processor class for Donut.)OptionalUnionN)BaseImageProcessor BatchFeature get_size_dict)convert_to_rgbget_resize_output_image_sizepadresizeto_channel_dimension_format) IMAGENET_STANDARD_MEANIMAGENET_STANDARD_STDChannelDimension ImageInputPILImageResamplingget_image_sizeinfer_channel_dimension_formatis_scaled_imagemake_flat_list_of_imagesto_numpy_array valid_imagesvalidate_preprocess_arguments) TensorTypefilter_out_non_signature_kwargslogging)is_vision_availablerequires)vision)backendsc%eZdZdZdgZddejddddddddf dedee e e fd ed ed ed ed ede e e fdedee e ee fdee e ee fddffd Z ddejde e e fdee e efdee e efdejf dZ d dejde e e fdedee e efdee e efdejf dZejddfdejde e e fd edee e efdee e efdejf dZejddfdejde e e fd edee e efdee e efdejf dZedddddddddddddejdfdedeedee e e fd eed eed eed eeded eedee deedee e ee fdee e ee fdee e efdeedee e efdejjf"dZxZS)!DonutImageProcessora Constructs a Donut image processor. Args: do_resize (`bool`, *optional*, defaults to `True`): Whether to resize the image's (height, width) dimensions to the specified `size`. Can be overridden by `do_resize` in the `preprocess` method. size (`dict[str, int]` *optional*, defaults to `{"shortest_edge": 224}`): Size of the image after resizing. The shortest edge of the image is resized to size["shortest_edge"], with the longest edge resized to keep the input aspect ratio. Can be overridden by `size` in the `preprocess` method. resample (`PILImageResampling`, *optional*, defaults to `Resampling.BILINEAR`): Resampling filter to use if resizing the image. Can be overridden by `resample` in the `preprocess` method. do_thumbnail (`bool`, *optional*, defaults to `True`): Whether to resize the image using thumbnail method. do_align_long_axis (`bool`, *optional*, defaults to `False`): Whether to align the long axis of the image with the long axis of `size` by rotating by 90 degrees. do_pad (`bool`, *optional*, defaults to `True`): Whether to pad the image. If `random_padding` is set to `True` in `preprocess`, each image is padded with a random amount of padding on each size, up to the largest image size in the batch. Otherwise, all images are padded to the largest image size in the batch. do_rescale (`bool`, *optional*, defaults to `True`): Whether to rescale the image by the specified scale `rescale_factor`. Can be overridden by `do_rescale` in the `preprocess` method. rescale_factor (`int` or `float`, *optional*, defaults to `1/255`): Scale factor to use if rescaling the image. Can be overridden by `rescale_factor` in the `preprocess` method. do_normalize (`bool`, *optional*, defaults to `True`): Whether to normalize the image. Can be overridden by `do_normalize` in the `preprocess` method. image_mean (`float` or `list[float]`, *optional*, defaults to `IMAGENET_STANDARD_MEAN`): Mean to use if normalizing the image. This is a float or list of floats the length of the number of channels in the image. Can be overridden by the `image_mean` parameter in the `preprocess` method. image_std (`float` or `list[float]`, *optional*, defaults to `IMAGENET_STANDARD_STD`): Image standard deviation. pixel_valuesTNFgp? do_resizesizeresample do_thumbnaildo_align_long_axisdo_pad do_rescalerescale_factor do_normalize image_mean image_stdreturnc  vtjdi| ||nddd}t|ttfr |ddd}t |}||_||_||_||_ ||_ ||_ ||_ ||_ | |_| | nt|_| | nt"|_dS)Ni i)heightwidth)super__init__ isinstancetuplelistrr$r%r&r'r(r)r*r+r,rr-rr.)selfr$r%r&r'r(r)r*r+r,r-r.kwargs __class__s /home/jaya/work/projects/VOICE-AGENT/VIET/agent-env/lib/python3.11/site-packages/transformers/models/donut/image_processing_donut.pyr6zDonutImageProcessor.__init__^s ""6"""'ttt-L-L dUDM * * ":DT"""   ("4 $,((2(>**DZ&/&;AVimage data_formatinput_data_formatcft||\}}|d|d}}|t|}|tjkrd} n%|tjkrd} nt d|||kr||ks ||kr||krt j|d| }|t||| }|S) a Align the long axis of the image to the longest axis of the specified size. Args: image (`np.ndarray`): The image to be aligned. size (`dict[str, int]`): The size `{"height": h, "width": w}` to align the long axis to. data_format (`str` or `ChannelDimension`, *optional*): The data format of the output image. If unset, the same format as the input image is used. input_data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format of the input image. If not provided, it will be inferred. Returns: `np.ndarray`: The aligned image.  channel_dimr1r2N)r)rEzUnsupported data format: r)axesinput_channel_dim) rrrLASTFIRST ValueErrornprot90r ) r:r?r%r@rA input_height input_width output_height output_widthrot_axess r=align_long_axisz#DonutImageProcessor.align_long_axiss.%35FW$X$X$X! k&*8nd7m|  $ >u E E   0 5 5 5HH "2"8 8 8HHL9JLLMM M = ( ([<-G-G = ( ([<-G-GHUAH555E  "/{VghhhE r>random_paddingcR|d|d}}t||\}} || z } ||z } |rItjd| dz} tjd| dz} n | dz} | dz} | | z }| | z }| |f| |ff}t ||||S) a Pad the image to the specified size. Args: image (`np.ndarray`): The image to be padded. size (`dict[str, int]`): The size `{"height": h, "width": w}` to pad the image to. random_padding (`bool`, *optional*, defaults to `False`): Whether to use random padding or not. data_format (`str` or `ChannelDimension`, *optional*): The data format of the output image. If unset, the same format as the input image is used. input_data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format of the input image. If not provided, it will be inferred. r1r2rCrrE)lowhighrF)r@rA)rrMrandomrandintr )r:r?r%rUr@rArQrRrOrP delta_width delta_heightpad_toppad_left pad_bottom pad_rightpaddings r= pad_imagezDonutImageProcessor.pad_images.'+8nd7m| $25FW$X$X$X! k"[0 $|3  (i''AL14D'EEGy((Q[1_(EEHH"a'G"a'H!G+ (* Z(8Y*?@5'{N_````r>c >t||\}}|d|d} } t|| } t|| } | |kr| |kr|S||krt|| z|z } n||krt|| z|z } t|f| | f|d||d|S)as Resize the image to make a thumbnail. The image is resized so that no dimension is larger than any corresponding dimension of the specified size. Args: image (`np.ndarray`): The image to be resized. size (`dict[str, int]`): The size `{"height": h, "width": w}` to resize the image to. resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BICUBIC`): The resampling filter to use. data_format (`Optional[Union[str, ChannelDimension]]`, *optional*): The data format of the output image. If unset, the same format as the input image is used. input_data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format of the input image. If not provided, it will be inferred. rCr1r2g@)r%r& reducing_gapr@rA)rminintr ) r:r?r%r&r@rAr;rOrPrQrRr1r2s r= thumbnailzDonutImageProcessor.thumbnails2%35FW$X$X$X! k&*8nd7m| \=11K.. \ ! !e{&:&:L + % % f,|;<c t|}t|d|d}t||d|}t|f||||d|} | S)a Resizes `image` to `(height, width)` specified by `size` using the PIL library. Args: image (`np.ndarray`): Image to resize. size (`dict[str, int]`): Size of the output image. resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BICUBIC`): Resampling filter to use when resiizing the image. data_format (`str` or `ChannelDimension`, *optional*): The channel dimension format of the image. If not provided, it will be the same as the input image. input_data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format of the input image. If not provided, it will be inferred. r1r2F)r%default_to_squarerA)r%r&r@rA)rrer r ) r:r?r%r&r@rAr; shortest_edge output_size resized_images r=r zDonutImageProcessor.resize s0T""DNDM:: 2  Rc     #/      r>imagesreturn_tensorsc  ||nj}njtttfr dddt nj||nj}||nj}||nj }| | nj }  nj | | nj }  nj  nj t|}t!|st#dt%| | |d|D}d|D}| r/t'|drt(dt-|d|rfd |D}|rfd |D}|rfd |D}|rfd |D}| r fd |D}| r fd|D}fd|D}d|i}t/||S)a Preprocess an image or batch of images. Args: images (`ImageInput`): Image to preprocess. Expects a single or batch of images with pixel values ranging from 0 to 255. If passing in images with pixel values between 0 and 1, set `do_rescale=False`. do_resize (`bool`, *optional*, defaults to `self.do_resize`): Whether to resize the image. size (`dict[str, int]`, *optional*, defaults to `self.size`): Size of the image after resizing. Shortest edge of the image is resized to min(size["height"], size["width"]) with the longest edge resized to keep the input aspect ratio. resample (`int`, *optional*, defaults to `self.resample`): Resampling filter to use if resizing the image. This can be one of the enum `PILImageResampling`. Only has an effect if `do_resize` is set to `True`. do_thumbnail (`bool`, *optional*, defaults to `self.do_thumbnail`): Whether to resize the image using thumbnail method. do_align_long_axis (`bool`, *optional*, defaults to `self.do_align_long_axis`): Whether to align the long axis of the image with the long axis of `size` by rotating by 90 degrees. do_pad (`bool`, *optional*, defaults to `self.do_pad`): Whether to pad the image. If `random_padding` is set to `True`, each image is padded with a random amount of padding on each size, up to the largest image size in the batch. Otherwise, all images are padded to the largest image size in the batch. random_padding (`bool`, *optional*, defaults to `self.random_padding`): Whether to use random padding when padding the image. If `True`, each image in the batch with be padded with a random amount of padding on each side up to the size of the largest image in the batch. do_rescale (`bool`, *optional*, defaults to `self.do_rescale`): Whether to rescale the image pixel values. rescale_factor (`float`, *optional*, defaults to `self.rescale_factor`): Rescale factor to rescale the image by if `do_rescale` is set to `True`. do_normalize (`bool`, *optional*, defaults to `self.do_normalize`): Whether to normalize the image. image_mean (`float` or `list[float]`, *optional*, defaults to `self.image_mean`): Image mean to use for normalization. image_std (`float` or `list[float]`, *optional*, defaults to `self.image_std`): Image standard deviation to use for normalization. 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. - Unset: defaults to the channel dimension format of the input image. 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. Nr3zkInvalid image type. Must be of type PIL.Image.Image, numpy.ndarray, torch.Tensor, tf.Tensor or jax.ndarray.)r*r+r,r-r.r$r%r&c,g|]}t|Sr4)r .0r?s r= z2DonutImageProcessor.preprocess.. <<c,g|]}t|Sr4)rrqs r=rsz2DonutImageProcessor.preprocess..rtr>rzIt looks like you are trying to rescale already rescaled images. If the input images have pixel values between 0 and 1, set `do_rescale=False` to avoid rescaling them again.c@g|]}|S))r%rA)rTrrr?rAr:r%s r=rsz2DonutImageProcessor.preprocess..s0vvvfkd**5tO`*aavvvr>cBg|]}|S))r?r%r&rA)r )rrr?rAr&r:r%s r=rsz2DonutImageProcessor.preprocess..s> %dXYj kkr>c@g|]}|S))r?r%rA)rgrws r=rsz2DonutImageProcessor.preprocess..s-vvvfkdnn5tO`naavvvr>cBg|]}|S))r?r%rUrA)rb)rrr?rArUr:r%s r=rsz2DonutImageProcessor.preprocess..sHd>]nr>c@g|]}|S))r?scalerA)rescale)rrr?rAr+r:s r=rsz2DonutImageProcessor.preprocess..s< 5Rc ddr>cBg|]}|S))r?meanstdrA) normalize)rrr?r-r.rAr:s r=rsz2DonutImageProcessor.preprocess..s>U^oppr>c4g|]}t|S)rH)r )rrr?r@rAs r=rsz2DonutImageProcessor.preprocess..s7   ej '{N_ ` ` `   r>r#)data tensor_type)r$r%r7r8r9rr&r'r(r)r*r+r,r-r.rrrLrrlogger warning_oncerr)r:rmr$r%r&r'r(r)rUr*r+r,r-r.rnr@rArs` `` ` ` `` `` r= preprocesszDonutImageProcessor.preprocess3swV"+!6IIDN 'ttTY dUDM * * ":DT""'388'3'?||TEV 3E3Q//W[Wn!-4;#-#9ZZt +9+E4K^'3'?||TEV #-#9ZZt !*!6IIDN )&11F## :  &!)%! =vay I I   wvvvvvvouvvvF  #F  wvvvvvvouvvvF  $ F  #F  #F      nt   '>BBBBr>)NN)FNN) __name__ __module__ __qualname____doc__model_input_namesrBILINEARboolrdictstrrfrfloatr9r6rMndarrayrrTrbBICUBICrgr rrKrrPILImager __classcell__)r<s@r=r"r"6s/""H(()-'9'B!#(,3!:>9=!W!W!WtCH~&!W% !W  !W ! !W!W!Wc5j)!W!WU5$u+#567!WE%e"456!W !W!W!W!W!W!WN?CDH --z-38n-eC)9$9:; - $E#/?*?$@A -  ----f %>BDH (a(az(a38n(a (a eC)9$9:; (a $E#/?*?$@A (a (a(a(a(a\(:'A>BDH 0 0 z0 38n0 % 0 eC)9$9:; 0 $E#/?*?$@A 0  0 0 0 0 l(:'A>BDH %%z%38n%% % eC)9$9:; % $E#/?*?$@A % %%%%N%$&&%))-15'+-1!%$%)*.'+:>9=;?2B2HDH#`C`C`CD>`CtCH~& `C -. `C tn `C%TN`C`C`CTN`C!`Ctn`CU5$u+#567`CE%e"456`C!sJ!78`C ./!`C"$E#/?*?$@A#`C$ %`C`C`C'&`C`C`C`C`Cr>r")*rtypingrrnumpyrMimage_processing_utilsrrrimage_transformsr r r r r image_utilsrrrrrrrrrrrrutilsrrrutils.import_utilsrr get_loggerrrrr"__all__r4r>r=rs'&""""""""UUUUUUUUUU                            JIIIIIIIII????????  H % %JJJ ;]C]C]C]C]C,]C]C ]C@ ! !r>