o Zh`@s dZddlmZmZmZmZmZmZddlZ ddl m Z m Z m Z mZddlmZmZddlmZmZmZmZmZmZmZmZmZmZmZddlmZm Z m!Z!m"Z"m#Z#m$Z$dd l%m&Z&dd l'm(Z(e#rgddl)Z)e!rnddl*Z*e$+e,Z-e(d d Gd dde Z.dgZ/dS)zImage processor class for Beit.)AnyDictListOptionalTupleUnionN)INIT_SERVICE_KWARGSBaseImageProcessor BatchFeature get_size_dict)resizeto_channel_dimension_format) IMAGENET_STANDARD_MEANIMAGENET_STANDARD_STDChannelDimension ImageInputPILImageResamplinginfer_channel_dimension_formatis_scaled_imagemake_list_of_imagesto_numpy_array valid_imagesvalidate_preprocess_arguments) TensorTypefilter_out_non_signature_kwargsis_torch_availableis_torch_tensoris_vision_availablelogging)deprecate_kwarg)requires)Zvision)backendsc&sFeZdZdZdgZeddddeeddd ej dd d ddd d d f d e d e e e efdede de e e efdeeefde de de eeeefde eeeefde dd ffdd Zede e efffdd Zej d d fdejd e e efdede ee efde ee efdejf dd Zd!edejfd"d#Z d5dede e d e e d e e e efdede e de e e efde e de ede e de eeeefde eeeefde ee effd$d%Z d5ded e e d e e e efdede e de e e efde e de ede e de eeeefde eeeefde ee efde ee efdejfd&d'Z d6d(ed e e d e e e efdede e de e e efde e de ee effd)d*Zd7fd+d, Zedddded d d d d d d d d d d d d ej d fd-ed.e ed e e d e e e efdede e de e e efde e de ede e de eeeefde eeeefde e d/e ee e!fdede ee efde"j#j#f"d0d1Z$d7d2e ee%fd3d4Z&Z'S)8BeitImageProcessoraK Constructs a BEiT 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 the `do_resize` parameter in the `preprocess` method. size (`Dict[str, int]` *optional*, defaults to `{"height": 256, "width": 256}`): Size of the output image after resizing. Can be overridden by the `size` parameter in the `preprocess` method. resample (`PILImageResampling`, *optional*, defaults to `Resampling.BICUBIC`): Resampling filter to use if resizing the image. Can be overridden by the `resample` parameter in the `preprocess` method. do_center_crop (`bool`, *optional*, defaults to `True`): Whether to center crop the image. If the input size is smaller than `crop_size` along any edge, the image is padded with 0's and then center cropped. Can be overridden by the `do_center_crop` parameter in the `preprocess` method. crop_size (`Dict[str, int]`, *optional*, defaults to `{"height": 224, "width": 224}`): Desired output size when applying center-cropping. Only has an effect if `do_center_crop` is set to `True`. Can be overridden by the `crop_size` parameter 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 the `rescale_factor` parameter in the `preprocess` method. do_rescale (`bool`, *optional*, defaults to `True`): Whether to rescale the image by the specified scale `rescale_factor`. Can be overridden by the `do_rescale` parameter in the `preprocess` method. do_normalize (`bool`, *optional*, defaults to `True`): Whether to normalize the image. Can be overridden by the `do_normalize` parameter in the `preprocess` method. image_mean (`float` or `List[float]`, *optional*, defaults to `IMAGENET_STANDARD_MEAN`): The mean to use if normalizing the image. This is a float or list of floats of length of the number of channels of 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`): The standard deviation to use if normalizing the image. This is a float or list of floats of length of the number of channels of the image. Can be overridden by the `image_std` parameter in the `preprocess` method. do_reduce_labels (`bool`, *optional*, defaults to `False`): Whether or not to reduce all label values of segmentation maps by 1. Usually used for datasets where 0 is used for background, and background itself is not included in all classes of a dataset (e.g. ADE20k). The background label will be replaced by 255. Can be overridden by the `do_reduce_labels` parameter in the `preprocess` method. pixel_values reduce_labelsdo_reduce_labelsz4.41.0)new_nameversion)extraTNgp?F do_resizesizeresampledo_center_crop crop_sizerescale_factor do_rescale do_normalize image_mean image_stdreturnc  stjdi| |dur|nddd}t|}|dur|nddd}t|dd}||_||_||_||_||_||_||_ ||_ | durG| nt |_ | durP| nt |_| |_dS)N)heightwidthr.) param_name)super__init__r r*r+r,r-r.r0r/r1rr2rr3r&) selfr*r+r,r-r.r/r0r1r2r3r&kwargs __class__r:]/var/www/auris/lib/python3.10/site-packages/transformers/models/beit/image_processing_beit.pyr<hs   zBeitImageProcessor.__init__image_processor_dictc s2|}d|vr|d|d<tj|fi|S)z Overrides the `from_dict` method from the base class to save support of deprecated `reduce_labels` in old configs r%r&)copypopr; from_dict)clsrBr>r?r:rArEszBeitImageProcessor.from_dictimage data_formatinput_data_formatcKsVt|ddd}d|vsd|vrtd|t|f|d|df|||d|S)a Resize an image to (size["height"], size["width"]). Args: image (`np.ndarray`): Image to resize. size (`Dict[str, int]`): Size of the output image. resample (`PILImageResampling`, *optional*, defaults to `PIL.Image.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 (`str` or `ChannelDimension`, *optional*): The channel dimension format of the input image. If not provided, it will be inferred. Tr+Zdefault_to_squarer9r6r7z@The `size` argument must contain `height` and `width` keys. Got )r+r,rHrI)r ValueErrorkeysr )r=rGr+r,rHrIr>r:r:rAr szBeitImageProcessor.resizelabelcCs,t|}d||dk<|d}d||dk<|S)Nr)r)r=rMr:r:rA reduce_labels   zBeitImageProcessor.reduce_labelcCsf|r||}|r|j|||| d}|r|j||| d}|r&|j|| | d}| r1|j|| | | d}|S)N)rGr+r,rI)rGr+rI)rGscalerI)rGmeanZstdrI)rQr Z center_cropZrescale normalize)r=rGr&r*r+r,r-r.r0r/r1r2r3rIr:r:rA _preprocesss zBeitImageProcessor._preprocesscCslt|}|rt|rtd| durt|} |j|d|||||||| | | | d }| dur4t|| | d}|S)zPreprocesses a single image.zIt 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.NF) r&r*r+r,r-r.r0r/r1r2r3rI)Zinput_channel_dim)rrloggerZ warning_oncerrUr)r=rGr*r+r,r-r.r0r/r1r2r3rHrIr:r:rA_preprocess_images2 z$BeitImageProcessor._preprocess_imagesegmentation_mapc Cst|}|jdkr|d}d} tj}n d} |durt|dd}|j|||||||ddtjd }| r8tj|d d }|tj }|S) z'Preprocesses a single segmentation map.)N.TFNrO)Z num_channels) rGr&r*r,r+r-r.r1r0rIr)Zaxis) rndimrFIRSTrrUnpZsqueezeZastypeZint64) r=rXr*r+r,r-r.r&rIZadded_dimensionr:r:rA_preprocess_segmentation_maps0    z/BeitImageProcessor._preprocess_segmentation_mapc stj|fd|i|S)Nsegmentation_maps)r;__call__)r=imagesr^r>r?r:rAr_7szBeitImageProcessor.__call__r`r^return_tensorsc sdurn j dur n j t ddd dur n j dur(n jdur1n jtddddurAn j durJ n j durSn jdur\n j duren j durnn j t |}|durt |dd}|durt |stdt |std t  d  f d d |D}d |i}|durч fdd |D}||d<t||dS)aI 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`. segmentation_maps (`ImageInput`, *optional*) Segmentation maps 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. 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_center_crop (`bool`, *optional*, defaults to `self.do_center_crop`): Whether to center crop the image. crop_size (`Dict[str, int]`, *optional*, defaults to `self.crop_size`): Size of the image after center crop. If one edge the image is smaller than `crop_size`, it will be padded with zeros and then cropped do_rescale (`bool`, *optional*, defaults to `self.do_rescale`): Whether to rescale the image values between [0 - 1]. 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. image_std (`float` or `List[float]`, *optional*, defaults to `self.image_std`): Image standard deviation. do_reduce_labels (`bool`, *optional*, defaults to `self.do_reduce_labels`): Whether or not to reduce all label values of segmentation maps by 1. Usually used for datasets where 0 is used for background, and background itself is not included in all classes of a dataset (e.g. ADE20k). The background label will be replaced by 255. 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: - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format. - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format. - Unset: Use 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. NTr+rJr.rY)Zexpected_ndimszwInvalid segmentation_maps type. Must be of type PIL.Image.Image, numpy.ndarray, torch.Tensor, tf.Tensor or jax.ndarray.zkInvalid image type. Must be of type PIL.Image.Image, numpy.ndarray, torch.Tensor, tf.Tensor or jax.ndarray.) r0r/r1r2r3r-r.r*r+r,cs0g|]} j| d qS)) rGr*r-r0r1r,r+r/r.r2r3rHrI)rW).0Zimg) r.rHr-r1r0r*r2r3rIr,r/r=r+r:rA s$z1BeitImageProcessor.preprocess..r$c s$g|]}j|dqS))rXr&r*r,r+r-r.)r])rbrX)r.r-r&r*r,r=r+r:rArcs labels)dataZ tensor_type)r*r+r r,r-r.r0r/r1r2r3r&rrrKrr )r=r`r^r*r+r,r-r.r0r/r1r2r3r&rarHrIrer:)r.rHr-r1r&r0r*r2r3rIr,r/r=r+rA preprocess<s\L "   zBeitImageProcessor.preprocess target_sizescs|j}|durHt|t|krtdt|r|}gtt|D]"}tjjj ||j dd||ddd}|dj dd} |q#S|j ddfd d tj dDS) a6 Converts the output of [`BeitForSemanticSegmentation`] into semantic segmentation maps. Only supports PyTorch. Args: outputs ([`BeitForSemanticSegmentation`]): Raw outputs of the model. target_sizes (`List[Tuple]` of length `batch_size`, *optional*): List of tuples corresponding to the requested final size (height, width) of each prediction. If unset, predictions will not be resized. Returns: semantic_segmentation: `List[torch.Tensor]` of length `batch_size`, where each item is a semantic segmentation map of shape (height, width) corresponding to the target_sizes entry (if `target_sizes` is specified). Each entry of each `torch.Tensor` correspond to a semantic class id. NzTMake sure that you pass in as many target sizes as the batch dimension of the logitsr)dimZbilinearF)r+modeZ align_cornersrOcsg|]}|qSr:r:)rbiZsemantic_segmentationr:rArcszIBeitImageProcessor.post_process_semantic_segmentation..)logitslenrKrnumpyrangetorchnnZ functionalZ interpolateZ unsqueezeZargmaxappendshape)r=ZoutputsrgrlidxZresized_logitsZ semantic_mapr:rkrA"post_process_semantic_segmentations&  z5BeitImageProcessor.post_process_semantic_segmentation) NNNNNNNNNNNN)NNNNNNN)N)(__name__ __module__ __qualname____doc__Zmodel_input_namesr rr rZBICUBICboolrrstrintrfloatrr< classmethodrrEr\Zndarrayrr rrQrUrWr]r_r[rPILZImagerfrru __classcell__r:r:r?rAr#:s*          $       $      0 )       r#)0rytypingrrrrrrrnr\Zimage_processing_utilsr r r r Zimage_transformsr rZ image_utilsrrrrrrrrrrrutilsrrrrrrZutils.deprecationr Zutils.import_utilsr!rrpZ get_loggerrvrVr#__all__r:r:r:rAs( 4    M