o i\@sdZddlmZmZddlZddlmZmZm Z ddl m Z m Z m Z mZddlmZmZmZmZmZmZmZmZmZddlmZmZmZmZmZmZdd l m!Z!erWddl"Z"er^ddl#Z#e$e%Z&e!d d Gd d d eZ'd gZ(dS)z$Image processor class for MobileViT.)OptionalUnionN)BaseImageProcessor BatchFeature get_size_dict)flip_channel_orderget_resize_output_image_sizeresizeto_channel_dimension_format) ChannelDimension ImageInputPILImageResamplinginfer_channel_dimension_formatis_scaled_imagemake_flat_list_of_imagesto_numpy_array valid_imagesvalidate_preprocess_arguments) TensorTypefilter_out_non_signature_kwargsis_torch_availableis_torch_tensoris_vision_availablelogging)requires)vision)backendsc!s|eZdZdZdgZddejddddddf dedee e e fd ed ed e e e fd ed ee e e fdededdffdd Zejddfdejde e e fd edee e efdee e efdejf ddZ  d.dejdee e efdee e efdejfddZdedejfddZd/fdd Z     d0dededed ed ededee e e fd eed ee d ee e e fdee e effdd Z          d1dedeedee e e fd eed eed ee d eed ee e e fdeedee e efdee e efdejfd!d"Z      d2d#edeedeedee e e fd eed ee e e fdee e efdejfd$d%Zedddddddddddejdf d&ed'eedeedee e e fd eed eed ee d eed ee e e fdeedeed(ee e efdedee e efdejjfd)d*Zd/d+ee e!fd,d-Z"Z#S)3MobileViTImageProcessora Constructs a MobileViT 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 `{"shortest_edge": 224}`): Controls the size of the output image after resizing. Can be overridden by the `size` parameter in the `preprocess` method. resample (`PILImageResampling`, *optional*, defaults to `Resampling.BILINEAR`): Defines the resampling filter to use if resizing the image. Can be overridden by the `resample` 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. 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_center_crop (`bool`, *optional*, defaults to `True`): Whether to crop the input at the center. 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": 256, "width": 256}`): Desired output size `(size["height"], size["width"])` when applying center-cropping. Can be overridden by the `crop_size` parameter in the `preprocess` method. do_flip_channel_order (`bool`, *optional*, defaults to `True`): Whether to flip the color channels from RGB to BGR. Can be overridden by the `do_flip_channel_order` 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_valuesTNgp?F do_resizesizeresample do_rescalerescale_factordo_center_crop crop_sizedo_flip_channel_orderdo_reduce_labelsreturnc  stjd i| |dur|nddi}t|dd}|dur|nddd}t|dd}||_||_||_||_||_||_||_ ||_ | |_ dS) N shortest_edgeFdefault_to_square)heightwidthr& param_name) super__init__rr r!r"r#r$r%r&r'r() selfr r!r"r#r$r%r&r'r(kwargs __class__r3l/home/ubuntu/.local/lib/python3.10/site-packages/transformers/models/mobilevit/image_processing_mobilevit.pyr5_s    z MobileViTImageProcessor.__init__image data_formatinput_data_formatc Ksnd}d|vr |d}d}nd|vrd|vr|d|df}ntdt||||d}t|f||||d|S) a[ Resize an image. The shortest edge of the image is resized to size["shortest_edge"], with the longest edge resized to keep the input aspect ratio. Args: image (`np.ndarray`): Image to resize. size (`dict[str, int]`): Size of the output image. resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BILINEAR`): Resampling filter to use when resiizing the image. data_format (`str` or `ChannelDimension`, *optional*): The channel dimension format of the image. If not provided, it will be the same as the input image. input_data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format of the input image. If not provided, it will be inferred. Tr*Fr/r0zASize must contain either 'shortest_edge' or 'height' and 'width'.)r!r-r=)r!r"r<r=) ValueErrorr r ) r6r;r!r"r<r=r7r- output_sizer3r3r:r }s.zMobileViTImageProcessor.resizecCst|||dS)a Flip the color channels from RGB to BGR or vice versa. Args: image (`np.ndarray`): The image, represented as a numpy array. data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format of the image. If not provided, it will be the same as the input image. input_data_format (`ChannelDimension` or `str`, *optional*): The channel dimension format of the input image. If not provided, it will be inferred. )r<r=)r)r6r;r<r=r3r3r:rsz*MobileViTImageProcessor.flip_channel_orderlabelcCs,t|}d||dk<|d}d||dk<|S)Nr)r)r6r@r3r3r: reduce_labels   z$MobileViTImageProcessor.reduce_labelc stj|fd|i|S)z Preprocesses a batch of images and optionally segmentation maps. Overrides the `__call__` method of the `Preprocessor` class so that both images and segmentation maps can be passed in as positional arguments. segmentation_maps)r4__call__)r6imagesrEr7r8r3r:rFsz MobileViTImageProcessor.__call__c Csb|r||}|r|j|||| d}|r|j|| | d}|r&|j|| | d}|r/|j|| d}|S)N)r;r!r"r=)r;scaler=)r;r!r=)r=)rDr rescale center_cropr) r6r;r(r r#r%r'r!r"r$r&r=r3r3r: _preprocesss z#MobileViTImageProcessor._preprocessc Cs`t|}|rt|rtd| durt|} |j|d|||||||| | d }t|| | d}|S)zPreprocesses a single image.zIt looks like you are trying to rescale already rescaled images. If the input images have pixel values between 0 and 1, set `do_rescale=False` to avoid rescaling them again.NF) r;r(r r!r"r#r$r%r&r'r=)input_channel_dim)rrlogger warning_oncerrKr ) r6r;r r!r"r#r$r%r&r'r<r=r3r3r:_preprocess_images, z)MobileViTImageProcessor._preprocess_imagesegmentation_mapc Cs|t|}|jdkrd}|d}tj}n d}|durt|dd}|j||||tjd||d|d }|r6|d }| t j }|S) zPreprocesses a single mask.T)N.FNrB) num_channels) r;r(r r!r"r#r%r&r'r=r) rndimr FIRSTrrKrNEARESTsqueezeastypenpint64) r6rPr(r r!r%r&r=added_channel_dimr3r3r:_preprocess_masks0    z(MobileViTImageProcessor._preprocess_maskrGrEreturn_tensorsc sdurn jdurn jdurn j dur! n j dur*n jdur3n j dur< n j t dd durKn jtdddurZn j t |}|durkt |dd}t |}t |swt d|durt |st d t   d  f d d |D}d |i}|dur fdd |D}||d<t|| dS)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`. segmentation_maps (`ImageInput`, *optional*): Segmentation map 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`): 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_rescale (`bool`, *optional*, defaults to `self.do_rescale`): Whether to rescale the image by rescale factor. rescale_factor (`float`, *optional*, defaults to `self.rescale_factor`): Rescale factor to rescale the image by if `do_rescale` 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 center crop if `do_center_crop` is set to `True`. do_flip_channel_order (`bool`, *optional*, defaults to `self.do_flip_channel_order`): Whether to flip the channel order of the image. 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: - `ChannelDimension.FIRST`: image in (num_channels, height, width) format. - `ChannelDimension.LAST`: image in (height, width, num_channels) 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. NFr,r&r1rQ)expected_ndimszkInvalid image type. Must be of type PIL.Image.Image, numpy.ndarray, torch.Tensor, tf.Tensor or jax.ndarray.zvInvalid segmentation map type. Must be of type PIL.Image.Image, numpy.ndarray, torch.Tensor, tf.Tensor or jax.ndarray.)r#r$r%r&r r!r"cs,g|]} j| d qS)) r;r r!r"r#r$r%r&r'r<r=)rO).0img) r&r<r%r'r#r r=r"r$r6r!r3r: s z6MobileViTImageProcessor.preprocess..rc s$g|]}j|dqS))rPr(r r!r%r&r=)r[)r^rP)r&r%r(r r=r6r!r3r:r`s labels)data tensor_type)r r"r#r$r%r'r!rr&r(rrr>rr)r6rGrEr r!r"r#r$r%r&r'r(r\r<r=rbr3) r&r<r%r'r(r#r r=r"r$r6r!r: preprocessEsVB      z"MobileViTImageProcessor.preprocess target_sizescs|j}|durHt|t|krtdt|r|}gtt|D]"}tjjj ||j dd||ddd}|dj dd} |q#S|j ddfd d tj dDS) a@ Converts the output of [`MobileViTForSemanticSegmentation`] into semantic segmentation maps. Only supports PyTorch. Args: outputs ([`MobileViTForSemanticSegmentation`]): 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)dimbilinearF)r!mode align_cornersrBcsg|]}|qSr3r3)r^isemantic_segmentationr3r:r`szNMobileViTImageProcessor.post_process_semantic_segmentation..)logitslenr>rnumpyrangetorchnn functional interpolate unsqueezeargmaxappendshape)r6outputsrermidxresized_logits semantic_mapr3rkr:"post_process_semantic_segmentations&  z:MobileViTImageProcessor.post_process_semantic_segmentation)NN)N)NNNNN) NNNNNNNNNN)NNNNNN)$__name__ __module__ __qualname____doc__model_input_namesrBILINEARboolrdictstrintrfloatr5rXndarrayr r rr rDrFrKrOr[rrTrPILImagerdlisttupler} __classcell__r3r3r8r:r7s$    "  4     "     . (       r))rtypingrrrorXimage_processing_utilsrrrimage_transformsrr r r image_utilsr r rrrrrrrutilsrrrrrrutils.import_utilsrrrq get_loggerr~rMr__all__r3r3r3r:s&,   Q