qiOdZddlZddlZddlZddlmZmZmZddl m Z m Z m Z ddl mZmZmZmZmZmZmZmZmZmZmZmZddlmZddlmZmZmZm Z e jBe"Z#erddl$Z$Gd d ed Z%Gd deZ&dgZ'y)z!Image processor class for Gemma3.N)BaseImageProcessor BatchFeature get_size_dict)convert_to_rgbresizeto_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) ImagesKwargs) TensorTypefilter_out_non_signature_kwargsis_vision_availableloggingc:eZdZUdZeed<eed<eed<eed<y)Gemma3ImageProcessorKwargsa do_pan_and_scan (`bool`, *optional*): Whether to apply `pan_and_scan` to images. pan_and_scan_min_crop_size (`int`, *optional*): Minimum size of each crop in pan and scan. pan_and_scan_max_num_crops (`int`, *optional*): Maximum number of crops per image in pan and scan. pan_and_scan_min_ratio_to_activate (`float`, *optional*): Minimum aspect ratio to activate pan and scan. do_pan_and_scanpan_and_scan_min_crop_sizepan_and_scan_max_num_crops"pan_and_scan_min_ratio_to_activateN)__name__ __module__ __qualname____doc__bool__annotations__intfloatd/opt/pipecat/venv/lib/python3.12/site-packages/transformers/models/gemma3/image_processing_gemma3.pyrr4s!  ## ##(--r*rF)totalc&eZdZdZddgZeZddejddddddddddf de de e e fdzd ed e d e e zd e d e ee zdzde ee zdzde dzde dzde dzde dzde dzddffd Z ddej"de de de de ezdzde ezdzf dZ ddeej"de de de de de ezdzde ezdzfdZedddddddddej,ddddddfdede dzde e e fdzd edzd e dzd e dzd e dzd e ee zdzde ee zdzde ezdzdedzde ezdzde dzde dzde dzde dzde dzdej4j4f$dZxZS)Gemma3ImageProcessoraI Constructs a SigLIP 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 `{"height": 224, "width": 224}`): Size of the image after resizing. 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_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 by the specified mean and standard deviation. Can be overridden by `do_normalize` in the `preprocess` method. image_mean (`float` or `list[float]`, *optional*, defaults to `[0.5, 0.5, 0.5]`): 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 `[0.5, 0.5, 0.5]`): Standard deviation 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_std` parameter in the `preprocess` method. Can be overridden by the `image_std` parameter in the `preprocess` method. do_convert_rgb (`bool`, *optional*, defaults to `True`): Whether to convert the image to RGB. do_pan_and_scan (`bool`, *optional*): Whether to apply `pan_and_scan` to images. pan_and_scan_min_crop_size (`int`, *optional*): Minimum size of each crop in pan and scan. pan_and_scan_max_num_crops (`int`, *optional*): Maximum number of crops per image in pan and scan. pan_and_scan_min_ratio_to_activate (`float`, *optional*): Minimum aspect ratio to activate pan and scan. pixel_values num_cropsTNgp? do_resizesizeresample do_rescalerescale_factor do_normalize image_mean image_stddo_convert_rgbrrrr returnc .t|di|||nddd}t|d}||nt}||nt}||_||_||_||_||_ ||_ ||_ ||_ | |_ | |_| |_| |_| |_y)N)heightwidthT)default_to_squarer))super__init__rr r r1r2r3r4r5r6r7r8r9rrrr )selfr1r2r3r4r5r6r7r8r9rrrr kwargs __class__s r+rAzGemma3ImageProcessor.__init__qs" "6"'tc-JTT:#-#9Z?U !*!6I num_crops_w num_crops_h crop_size_w crop_size_hicrop_positions_wcrop_positions_hpos_hpos_w image_cropss r+ pan_and_scanz!Gemma3ImageProcessor.pan_and_scansT6'u-  F?v~ BB djj#)=>?Kc$**U5O-O"PQS^_Ka-K8+FKK ~ BB djj%#)=>?Kc$**V6P-P"QRT_`Ka-K8+FKK$))EK$789 $))F[$89:  {K (+E EI5:;5GHK!OHH5:;5GHK!OHH  0 5 5 5%.$5$56FHX$Y E5eek1155;;N3NNOK%.$5$56FHX$Y E5a!44eek>Q6QQRK IH s H 6 H0H)Himagesc g}g} |D]H} |j| |||||} |j| g| z| jt| J|| fS)N)rErrr rFrG)r^extendappendlen) rBr_rrrr rFrGpas_images_listr0rE pas_imagess r+ _process_images_for_pan_and_scanz5Gemma3ImageProcessor._process_images_for_pan_and_scansz  .E**+E+E3U'"3 +J  " "E7Z#7 8   S_ - . ))r*return_tensorsc ||n |j}||n |j}t|dd}||n |j}||n |j}||n |j }||n |j }||n |j}| | n |j} | | n |j} ||n |j}||n |j}||n |j}||n |j}|j|}t|}t!|s t#dt%||||| |||| r|Dcgc] }t'|}}|Dcgc] }t)|}}|r#t+|drt,j/d| t1|d} |r|j3|||||| | \}}n|Dcgc]}d}}g}|D]k}|r|d |d }}t5|||f|| }|r|j7||| }|r|j9||| | }t;|| | }|j=|m||d}t?|| Scc}wcc}wcc}w)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. 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. 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. Only has an effect if `do_normalize` is set to `True`. image_std (`float` or `list[float]`, *optional*, defaults to `self.image_std`): Image standard deviation to use for normalization. Only has an effect if `do_normalize` is set to `True`. return_tensors (`str` or `TensorType`, *optional*): The type of tensors to return. Can be one of: - Unset: Return a list of `np.ndarray`. - `TensorType.PYTORCH` or `'pt'`: Return a batch of type `torch.Tensor`. - `TensorType.NUMPY` or `'np'`: Return a batch of type `np.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. do_convert_rgb (`bool`, *optional*, defaults to `self.do_convert_rgb`): Whether to convert the image to RGB. do_pan_and_scan (`bool`, *optional*, defaults to `self.do_pan_and_scan`): Whether to apply `pan_and_scan` to images. pan_and_scan_min_crop_size (`int`, *optional*, defaults to `self.pan_and_scan_min_crop_size`): Minimum size of each crop in pan and scan. pan_and_scan_max_num_crops (`int`, *optional*, defaults to `self.pan_and_scan_max_num_crops`): Maximum number of crops per image in pan and scan. pan_and_scan_min_ratio_to_activate (`float`, *optional*, defaults to `self.pan_and_scan_min_ratio_to_activate`): Minimum aspect ratio to activate pan and scan. r2F) param_namer?zSInvalid image type. Must be of type PIL.Image.Image, numpy.ndarray, or torch.Tensor)r4r5r6r7r8r1r2r3rzIt 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.)r_rrrr rFrGr=r>)rEr2r3rG)rEscalerG)rEmeanstdrG)input_channel_dim)r/r0)data tensor_type) r1r2rr3r4r5r6r7r8r9rrrr fetch_imagesrr ValueErrorrrrrlogger warning_oncerrfrrescale normalizer rbr)rBr_r1r2r3r4r5r6r7r8rgrFrGr9rrrr rEr0_processed_imagesr=r>rns r+ preprocesszGemma3ImageProcessor.preprocesssR"+!6IDNN 'tTYYTfN'38#-#9Zt +9+E4K^K^'3'?|TEVEV #-#9Zt !*!6IDNN +9+E4K^K^-<-H/dNbNb*D*P &VZVuVu #+E*P &VZVuVu # 2= /88 + ""6*)&1F#rs s%!)%!  9?@nU+@F@6<vay I   $ E E /+E+E3U'"3!F! FI%++q+I+ +E $XW vuo\m 5Zk ljiSd'0{VghE  # #E *! +$!1yI>BBcA=0,s*H<I- I)NN)r!r"r#r$model_input_namesr valid_kwargsrBILINEARr%dictstrr'r(listrAnpndarrayr r^rfrFIRSTr rPILImagerx __classcell__)rDs@r+r.r.Fs%N(5-L&*'9'B'B&-!1504&*'+1515;?#U#U38nt##U% #U  #U e #U#UDK'$.#U4;&-#Ut #U#U%($J#U%($J#U-2DL#U !#UV6:;?PzzP%(P%( P -2 P ++d2 P!11D8Pr6:;?*RZZ **%( * %( * -2 *++d2*!11D8*2%&"&&*.2"&'+$(150426/?/E/E;?&*'+1515;?%aCaC$;aC38nt# aC %t+ aC 4K aC aCTkaCDK'$.aC4;&-aCj(4/aC&,aC!11D8aCt aCaC %($J!aC"%($J#aC$-2DL%aC& 'aC'aCr*r.)(r$rRrKnumpyrimage_processing_utilsrrrimage_transformsrrr image_utilsr r r r rrrrrrrrprocessing_utilsrutilsrrrr get_loggerr!rrrrr.__all__r)r*r+rs( UU    -^^   H %.U.$]C-]C@ " "r*