eTheSSKJr SSKJrJr SSKJrJrJrJ r SSK r SSK J r JrJr SSKJrJrJrJrJr SSKJrJrJrJrJrJrJrJrJrJ r J!r! SS K"J#r# SS K$J%r%J&r&J'r'J(r(J)r)J*r*J+r+ \*"5(aSS KJ,r, \'"5(aSSK-r-\("5(a SS KJ.r. \)"5(aSS K/J0r1 O SS K2J0r1 OSr.\+Rf"\45r5\"SS9SSSSSSSSSSSSS\Rl4S\\7S\\8S\\7S\\ \8\9\84S\\ \8\9\84S\\7S\\:S\\7S\\S\\7S\\S\SS\\ \;\%4S\\4Sjj5rS'\9S!S#\?\:4S(jr@S)\ \ RS!4S*\:S#\9\ \ RS!44S+jrB"S,S-\S.S/9rC\&"S0S1\ 55rD"S2S35rEg)5)Iterable) lru_cachepartial)AnyOptional TypedDictUnionN)BaseImageProcessor BatchFeature get_size_dict)convert_to_rgbget_resize_output_image_sizeget_size_with_aspect_ratiogroup_images_by_shapereorder_images) ChannelDimension ImageInput ImageTypeSizeDictget_image_size#get_image_size_for_max_height_widthget_image_typeinfer_channel_dimension_formatmake_flat_list_of_imagesvalidate_kwargsvalidate_preprocess_arguments)Unpack) TensorTypeauto_docstringis_torch_availableis_torchvision_availableis_torchvision_v2_availableis_vision_availablelogging)PILImageResampling)pil_torch_interpolation_mapping) functional maxsize do_rescalerescale_factor do_normalize image_mean image_stddo_padsize_divisibilitydo_center_crop crop_size do_resizesizeresampler&return_tensors data_formatc[UUUUUUUUUU U U S9 U bU S:wa [S5eU [R:wa [S5eg)z Checks validity of typically used arguments in an `ImageProcessorFast` `preprocess` method. Raises `ValueError` if arguments incompatibility is caught. ) r,r-r.r/r0r1r2r3r4r5r6r7Nptz6Only returning PyTorch tensors is currently supported.z6Only channel first data format is currently supported.)r ValueErrorrFIRST)r,r-r.r/r0r1r2r3r4r5r6r7r8r9s `/var/www/auris/envauris/lib/python3.13/site-packages/transformers/image_processing_utils_fast.py"validate_fast_preprocess_argumentsr?Isk*"%!+%  !n&<QRR&,,,QRR-tensor torch.TensoraxisreturnclUcUR5$URUS9$![a Us$f=f)z> Squeezes a tensor, but only if the axis specified has dim 1. )rC)squeezer<)rArCs r> safe_squeezerGts@ |~~~~4~((  s $ 33valuescP[U6Vs/sHn[U5PM sn$s snf)zG Return the maximum value across all indices of an iterable of values. )zipmax)rHvalues_is r>max_across_indicesrMs$+.v, 7,hCM, 77 7s#imagescb[UVs/sHoRPM sn5up#nX44$s snf)z@ Get the maximum height and width across all images in a batch. )rMshape)rNimg_ max_height max_widths r>get_max_height_widthrUs3 22O992OPA9  ""3Ps,image patch_sizec/n[U[RS9up4[SX15H9n[SXA5H&nUSS2XUU-2XfU-24nUR U5 M( M; U$)a Divides an image into patches of a specified size. Args: image (`Union[np.array, "torch.Tensor"]`): The input image. patch_size (`int`): The size of each patch. Returns: list: A list of Union[np.array, "torch.Tensor"] representing the patches. ) channel_dimrN)rrr=rangeappend)rVrWpatchesheightwidthijpatchs r>divide_to_patchesrbsrG"56F6L6LMMF 1f )q%,A!QZ/^1CCDE NN5 !-* Nr@cn\rSrSr%\\\S'\\\\ 4\S'\\\S'\\ S\S'\\\S'\\\\ 4\S'\\\S '\\ \ \ 4\S '\\\S '\\ \ \ \ 4\S '\\ \ \ \ 4\S '\\\S'\\ \\ 4\S'\\\S'\\ \\4\S'\S\S'Srg)DefaultFastImageProcessorKwargsr5r6default_to_squarer&F.InterpolationModer7r3r4r,r-r.r/r0do_convert_rgbr8r9input_data_format torch.devicedeviceN)__name__ __module__ __qualname____firstlineno__rbool__annotations__dictstrintr floatlistrr__static_attributes__rmr@r>rdrds~ 4S> ""~%uHIJJTN"S#X''U3:.//4. ud5k1233eT%[0122TN"U3 ?344*++c+;&; <== ^ $$r@rdF)totalc^\rSrSrSrSrSrSrSrSr Sr Sr Sr Sr SrSrSr\R$rSrSrS/r\rSrS\\SS4U4SjjrS8S S S \S S S\SS 4 SjjrS S S\SS 4Sjr S S S\!\\"\4S\!\\"\4SS 4Sjr#\$"SS9S9S\%\S\%\!\\&\4S\%\!\\&\4S\%\S\%\S\%SS\'4Sjj5r(SS S\S\S\S\!\\&\4S\!\\&\4SS 4S jr)S S S \*\+\,4SS 4S!jr-S \.S\.4S"jr/S\*4S#jr0S\.S\.4S$jr1S:S \.S%\%\S&\%\!\+\4S\%SSS 4 S'jjr2S:S\.S%\%\S&\%\!\+\4S\%SS\&S 4 S(jjr3S9S \%\S)\%\S*\%\S\%\!\\&\4S\%\!\\&\4S+\%\S\*4S,jjr4S;S\%\S\%\S\%\S\%\!\\'\4S\%\!\\'\4S-\%\S \%\S.\%\S)\%\S/\%\!S0S1\%\!\+\54S+\%\4S2jjr6S\.S\\S\74S3jr8\9S\.S\\S\74S4j5r:S\&S S-\S \S \%S S.\S)\S\S\S\S\%\!\\&\4S\%\!\\&\4S1\%\!\+\54S\74S5jr;U4S6jr$)<BaseImageProcessorFastNTgp? pixel_valueskwargsrDc f>[TU]"S0UD6 URU5nURSUR5nUb#[ X!RSUR 5S9OSUlURSUR5nUb [ USS9OSUlURRR5H<nURUS5nUb[XU5 M&[X[XS55 M> [URRR55Ulg)Nr6rfr6rfr4 param_namerm)super__init__filter_out_unused_kwargspopr6r rfr4 valid_kwargsrskeyssetattrgetattrrx_valid_kwargs_names)selfrr6r4keykwarg __class__s r>rBaseImageProcessorFast.__init__s "6"..v6zz&$)), tzzBUW[WmWm7n o JJ{DNN; MVMby[Ihl$$4499;CJJsD)E 5)74d#;< <$((9(9(I(I(N(N(P#Q r@rVrBr6 interpolationrh antialiasc UbUO[RRnUR(aDUR(a3[ UR 5SSURUR5nOUR(a%[UURS[RS9nOUR(aDUR(a3[UR 5SSURUR5nOJUR(a*UR(aURUR4nO[SUS35e[R "XX4S9$)a Resize an image to `(size["height"], size["width"])`. Args: image (`torch.Tensor`): Image to resize. size (`SizeDict`): Dictionary in the format `{"height": int, "width": int}` specifying the size of the output image. interpolation (`InterpolationMode`, *optional*, defaults to `InterpolationMode.BILINEAR`): `InterpolationMode` filter to use when resizing the image e.g. `InterpolationMode.BICUBIC`. Returns: `torch.Tensor`: The resized image. NF)r6rfrjzjSize must contain 'height' and 'width' keys, or 'max_height' and 'max_width', or 'shortest_edge' key. Got .)rr)FInterpolationModeBILINEAR shortest_edge longest_edgerr6rrr=rSrTrr]r^r<resize)rrVr6rrrnew_sizes r>rBaseImageProcessorFast.resizes ,*7)B H[H[HdHd   $"3"32 RS!""!!H   3''"'"2"8"8 H __:5::<;Ldoo_c_m_mnH [[TZZ TZZ0H6 xx}ZZr@scalec X-$)z Rescale an image by a scale factor. image = image * scale. Args: image (`torch.Tensor`): Image to rescale. scale (`float`): The scaling factor to rescale pixel values by. Returns: `torch.Tensor`: The rescaled image. rm)rrVrrs r>rescaleBaseImageProcessorFast.rescales $}r@meanstdc 0[R"XU5$)a Normalize an image. image = (image - image_mean) / image_std. Args: image (`torch.Tensor`): Image to normalize. mean (`torch.Tensor`, `float` or `Iterable[float]`): Image mean to use for normalization. std (`torch.Tensor`, `float` or `Iterable[float]`): Image standard deviation to use for normalization. Returns: `torch.Tensor`: The normalized image. )r normalize)rrVrrrs r>r BaseImageProcessorFast.normalize0s*{{5,,r@r)r*r.r/r0r,r-rlrkcU(a=U(a6[R"X&S9SU- -n[R"X6S9SU- -nSnX#U4$)N)rlg?F)torchrA)rr.r/r0r,r-rls r>!_fuse_mean_std_and_rescale_factor8BaseImageProcessorFast._fuse_mean_std_and_rescale_factorGsI ,j@C.DXYJ Y>#BVWIJj00r@rNc URUUUUUURS9upVnU(a/URUR[R S9XV5nU$U(aUR X5nU$)z Rescale and normalize images. )r.r/r0r,r-rl)dtype)rrlrtorfloat32r)rrNr,r-r.r/r0s r>rescale_and_normalize,BaseImageProcessorFast.rescale_and_normalizeXsx-1,R,R%!!)== -S- ) z ^^FIIEMMI$BJZF \\&9F r@c URb URc[SUR535e[R "XSUS45$)aj Center crop an image to `(size["height"], size["width"])`. If the input size is smaller than `crop_size` along any edge, the image is padded with 0's and then center cropped. Args: image (`"torch.Tensor"`): Image to center crop. size (`Dict[str, int]`): Size of the output image. Returns: `torch.Tensor`: The center cropped image. z=The size dictionary must have keys 'height' and 'width'. Got r]r^)r]r^r<rr center_crop)rrVr6rs r>r"BaseImageProcessorFast.center_croptsQ& ;; $**"4\]a]f]f]h\ijk k}}U(^T']$CDDr@c[U5$)z Converts an image to RGB format. Only converts if the image is of type PIL.Image.Image, otherwise returns the image as is. Args: image (ImageInput): The image to convert. Returns: ImageInput: The converted image. )r)rrVs r>r%BaseImageProcessorFast.convert_to_rgbse$$r@cURcU$URH4nX!;dM [RSUS35 URU5 M6 U$)z: Filter out the unused kwargs from the kwargs dictionary. z!This processor does not use the `z ` parameter. It will be ignored.) unused_kwargslogger warning_oncer)rr kwarg_names r>r/BaseImageProcessorFast.filter_out_unused_kwargssW    %M,,J###&G |Ss$tu :&- r@c[U5$)z Prepare the images structure for processing. Args: images (`ImageInput`): The input images to process. Returns: `ImageInput`: The images with a valid nesting. )r)rrNs r>_prepare_images_structure0BaseImageProcessorFast._prepare_images_structures(//r@rirjcB[U5nU[R[R[R4;a[ SU35eU(aUR U5nU[R:Xa[R"U5nO8U[R:Xa$[R"U5R5nUc [U5nU[R:Xa!URSSS5R5nUbUR!U5nU$)NzUnsupported input image type rr )rrPILTORCHNUMPYr<rr pil_to_tensorr from_numpy contiguousrrLASTpermuter)rrVrirjrl image_types r>_process_image%BaseImageProcessorFast._process_images$E* immY__iooN Nu E   0 5 5 5MM!Q*557E  HHV$E r@cURU5n[URUUUS9n/nUHnURU"U55 M U$)z* Prepare the input images for processing. )rirjrl)rrrr[)rrNrirjrlprocess_image_fnprocessed_imagesrVs r>_prepare_input_images,BaseImageProcessorFast._prepare_input_imagess_//7"   )/  E  # #$4U$; < r@r4rfr9c <Uc0nUb[S 0[XS9D6nUb[S 0[USS9D6n[U[5(a [ U5n[U[5(a [ U5nUc[ R nXS'X'S'X7S'XGS'XWS'XgS'U$) z Update kwargs that need further processing before being validated Can be overridden by subclasses to customize the processing of kwargs. rr4rr6rfr/r0r9rm)rr isinstancerxtuplerr=)rr6r4rfr/r0r9rs r>_further_process_kwargs.BaseImageProcessorFast._further_process_kwargss >F  \m[\D   T={#STI j$ ' 'z*J i & &i(I  *00Kv'{&7"#)|'{ +} r@r5r3r7rgr8c  ,[UUUUUUUUU U U U S9 g)z0 validate the kwargs for the preprocess method. ) r,r-r.r/r0r5r6r3r4r7r8r9N)r?)rr,r-r.r/r0r5r6r3r4r7r8r9rs r>_validate_preprocess_kwargs2BaseImageProcessorFast._validate_preprocess_kwargss0& +!)%!))# r@c.UR"U/UQ70UD6$N) preprocess)rrNargsrs r>__call__BaseImageProcessorFast.__call__8sv7777r@c V[UR5URS9 URHnURU[ XS55 M! UR S5nUR S5nUR S5nUR XXgS9nUR"S 0UD6nUR"S 0UD6 UR S5n[U[[45(a [UOUUS'UR S5 UR S 5 UR"U/UQ70UD6$) N)captured_kwargsvalid_processor_keysrirjrl)rNrirjrlr7rrfr9rm)rrr setdefaultrrrrrrrvr&r' _preprocess) rrNrrrrirjrlr7s r>r!BaseImageProcessorFast.preprocess;s&  DLdLde22J   j'$D*I J3 $45"JJ':;H%++L],  --77 ((262::j) :DHsTfNg9h9h +H 5nv   &' =!8888r@c [U5up0nUR5H"unnU(aURUX4S9nUUU'M$ [UU5n[U5up0nUR5H8unnU(aUR UU5nUR UXxXU 5nUUU'M: [UU5nU (a[ R"USS9OUn[SU0U S9$)N)rVr6rrdimr~)data tensor_type) ritemsrrrrrstackr )rrNr5r6rr3r4r,r-r.r/r0r8rgrouped_imagesgrouped_images_indexresized_images_groupedrPstacked_imagesresized_imagesprocessed_images_groupedrs r>r"BaseImageProcessorFast._preprocesscs"0EV/L,!#%3%9%9%; !E>!%>!j,: "5 )&<((>@TU0E^/T,#% %3%9%9%; !E>!%!1!1.)!L!77 LV_N/= $U +&<**BDXYCQ5;;'7Q?Wg.2B!CQ_``r@cl>[TU]5nURSS5 URSS5 U$)N_valid_processor_keysr)rto_dictr)r encoder_dictrs r>rBaseImageProcessorFast.to_dicts7w( 0$7.5r@)rr4r6)NT)NNNNNN)NNN) NNNNNNNNNNNN)?rnrorprqr7r/r0r6rfr4r5r3r,r-r.rir8rr=r9rjrlmodel_input_namesrdrrrrrrrrrwrr rrrrrxrrrrtrurvrrrrrrrrrrr rr rrrry __classcell__)rs@r>r|r|s2HJI DIINJNLNN"((K F'(2LMR89R R804 /[/[/[- /[  /[ /[b  (--E8E?*+-5(5/) * -  -.r(,:>9=%)*.+/1tn1U5$u+#5671E%e"456 1 TN 1 ! 1(1 11     %e,- U +, 8EE38nE  E.%% % t 00 0&*.DH+/  ! $E#/?*?$@A  (   J*.DH+/   ! $E#/?*?$@A  (  n   4$((,,0:>9=26"x "H%"$D> " U5$u+#567 " E%e"456 "./" "L&**.'+;?:>$(#')-(,QU;?26  TN  !  tn  U5%,#678  E%u"567  D>  x   !  H%  5!LMN  !sJ!78  ./  D8z8FCbe%9jv%9%9N)a^$)a)a )a  56 )a  )a)a)a)a)aU5$u+#567)aE%e"456)a!sJ!78)a )aVr@r|c2\rSrSrSS\\\4SjjrSrg)SemanticSegmentationMixiniN target_sizescURnUb[U5[U5:wa [S5e/n[[U55Han[R R RX5RSS9X%SSS9nUSRSS9nURU5 Mc U$URSS9n[URS5Vs/sHoUPM nnU$s snf)a Converts the output of [`MobileNetV2ForSemanticSegmentation`] into semantic segmentation maps. Only supports PyTorch. Args: outputs ([`MobileNetV2ForSemanticSegmentation`]): 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. zTMake sure that you pass in as many target sizes as the batch dimension of the logitsrrbilinearF)r6mode align_cornersr ) logitslenr<rZrnnr( interpolate unsqueezeargmaxr[rP) routputsrrsemantic_segmentationidxresized_logits semantic_mapr_s r>"post_process_semantic_segmentation %,,\: *%$%+MMaM$8 !GLMbMhMhijMkGl$mGl!1%=Gl !$m$$%nsC+rmr) rnrorprqrrxrr ryrmr@r>rrs(%QUV[Q\H](%(%r@rr)Fcollections.abcr functoolsrrtypingrrrr numpynpimage_processing_utilsr r r image_transformsrrrrr image_utilsrrrrrrrrrrrprocessing_utilsrutilsrr r!r"r#r$r%r&rr'torchvision.transforms.v2r(rtorchvision.transforms get_loggerrnrr=rrrwrxrvrur?rGrMrrUarrayrbrdr|rrmr@r>rs%(22SS    %/<"$$=:&*#   H % 2!%&*#'6:59!'+%)$( $#/37;.>.D.D'S'SUO'S4.'Sud5k123 'S eT%[012 'S TN 'S }'STN'S!'S~'S 8 'S+,'SU3 ?34'S*+'S'ST  x}  8x}8c8#n!5#%*# >) *8; %.( )*0%iu%&T/TTn)%)%r@