+ j"^RIt^RIHuHuHt^RIHt^RIH t ^RI H t H t ^RI HtHtHtHtHt^RIHtHt^RIHtHt!RR ]R R 7t]!R R ]44tR .tR#)N)TorchvisionBackend) BatchFeature)group_images_by_shapereorder_images)OPENAI_CLIP_MEANOPENAI_CLIP_STD ImageInputPILImageResamplingSizeDict) ImagesKwargsUnpack) TensorTypeauto_docstringc0a]tRt^$toRtV3RltRtVtR#)JanusImageProcessorKwargsz min_size (`int`, *optional*, defaults to 14): The minimum allowed size for the resized image. Ensures that neither the height nor width falls below this value after resizing. c&<V^8dQh/S[;R&#)min_sizeint)format __classdict__s"q/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/janus/image_processing_janus.py __annotate__&JanusImageProcessorKwargs.__annotate__$sMN)__name__ __module__ __qualname____firstlineno____doc____annotate_func____static_attributes____classdictcell__)rs@rrr$srrF)totalcaa]tRt^.to]P t]t] t RRRR/t ^t Rt RtRtRt]tV3RlV3RlltRV3RlV3RllltRV3R lR lltRV3R lR lltRV3R lRlltRtVtV;t#)JanusImageProcessorheightiwidthTc0<V^8dQhRS[S[,/#)rkwargs)r r)rrs"rr JanusImageProcessor.__annotate__;s88(A!B8rc <\SV`!R/VBVPR4fRpMJ\;QJd#.RVPR44FNK 5M!RVPR444p\V4VnR#) image_meanNc3F"TFp\V^,4xK R#5i)Nr).0xs& r /JanusImageProcessor.__init__..@s$T;SaSS\\;Ss!r)r6r6)super__init__gettuplebackground_color)selfr,r; __class__s&, rr8JanusImageProcessor.__init__;sg "6" ::l # +. $u$T6::l;S$Tuu$T6::l;S$TT  %&6 7rc 8<V^8dQhRRRS[RS[RRRS[RR/#) rimage torch.Tensorsizerresample7PILImageResampling | tvF.InterpolationMode | int | None antialiasreturn)r rbool)rrs"rrr-CsPiiii i L i  i irc <VPe*VPeVPVP8wd \RVR, RVR, 24hVPpVPRRwrx\ Wx4p W), p \ \ \ Wz,4V4\ \ W,4V4R7p \S V`!WWER7#)Nz5Output height and width must be the same. Got height=r)z and width=r*)r)r*)rBrCrE) r)r* ValueErrorshapemaxr roundr7resize) r<r@rBrrCrEr,r)r*max_sizedeltaoutput_size_nonpaddedr=s &&&&&&, rrNJanusImageProcessor.resizeCs ;; $**"4 tzz8QGXGWWbcghocpbqr {{ BC( v% (uV^,h7eEM*H5!  w~e(~hhrcR<V^8dQhRRRS[S[S[S[S[3,,RR/#)rimagesrAr;rF)rr:)rrs"rrr-^s9000c3m 440  0rc .VPRRwr4VP^,pVP^,pW48XdV#\W44p\V\4'dV.pM\ V4V8wd\ RV R24h\ P!WeWw3VPVPR7p\V4F wrWRV RR3&K WC8d"Ws, ^,p WRRWV,1R3&V#Wt, ^,p WRRRWV,13&V#)a Pads an image to a square based on the longest edge. Args: images (`torch.Tensor`): The images to pad. background_color (`int` or `tuple[int, int, int]`, *optional*, defaults to 0): The color to use for the padding. Can be an integer for single channel or a tuple of integers representing for multi-channel images. If passed as integer in multi-channel mode, it will default to `0` in subsequent channels. Returns: `torch.Tensor`: The padded images. Nz(background_color must have no more than z) elements to match the number of channels)dtypedevice:NNNrI) rKrL isinstancerlenrJtorchzerosrVrW enumerate) r<rTr;r)r* num_channels batch_sizemax_dim padded_imagesicolorstarts &&& r pad_to_square!JanusImageProcessor.pad_to_square^s.& RS) ||A \\!_ ?Mf$ & , , 01  ! "l 2:<.Hqr  w 8 U[UbUb ""23HA(-!Q1* %4 >%!+E=C!Q 69 : _*E!%>jr!s,: "5 )&<((>@TU0E^fv/w,,#% %3%9%9%; !E>!%!3!3NUYUjUj!3!k!77 LV_N/= $U +&<**BDXY.2B!CQ_``rc<V^8dQhRS[RS[R,RS[R,RS[R,RS[S[,R,RS[S[,R,RS[R,R R /#) rrTrhNrirjr/rkrmrFrA)r rGrprorq)rrs"rrr-s&W&W&W4K&W &W Tk &W K$& &W;%&Wd &W &Wrc zaVeTM VPpSfRVP, MSoVeTM VPpVeTM VPpVeTM VPp\ ;QJd .V3Rl\ WV44FNK 5M!V3Rl\ WV444p\ ;QJd.RV4FNK 5M !RV44pVPVVSVVVRRVR7 PpV'd>VUu.uF1qP^^4P\P4NK3 ppV'd5V'd-VR8Xd&VUu.uFp\P!V4NK ppVR8wdTMRpVR8Xd\P!V^R7MTp\!R V/VR 7#uupiuupi) Ng?c3J<"TFwrS)V,V, xK R#5i)Nr)r2meanstdris& rr42JanusImageProcessor.postprocess..s$dIcIDN?T1C77Ics #c34"TFp^V, xK R#5i)Nr)r2rs& rr4rs7Yc!c''YsF)rhrirjr/rkrgrnrmzPIL.Image.Imagept)dimrtru)rhrirjr/rkr:zip preprocessrtcliptorZuint8tvF to_pil_imagestackr) r<rTrhrirjr/rkrmr@s &&&f&&&& r postprocessJanusImageProcessor.postprocesss$.#9Zt 6D6Lt222R`'3'?|TEVEV #-#9Zt !*!6IDNN UdZIcdUUdZIcdd E7Y7EE7Y77  !)%!)!   ,  FLMfUjjC(++EKK8fFM J>=N+N;AB6%c&&u-6FB+9=N+NTX/=/EV+6.&!9~VVNCs 7F3F8rs)T)r)NNNNNN)rrr r!r BICUBICrCrr/rrkrBrrgrhrjrnr valid_kwargsr8rNrdrrr$r% __classcell__)r=rs@@rr(r(.s!))H!JI c7C (DHIJL F,L88ii600d)a)aV&W&W&Wrr()rZ$torchvision.transforms.v2.functional transformsv2 functionalrimage_processing_backendsrimage_processing_utilsrimage_transformsrr image_utilsrrr r r processing_utilsr r utilsrrrr(__all__rrrrsh  22;2E5  ErW,rWrWj ! !r