+ j|.^RIt^RIHt^RIt^RIHt^RIHt^RI H t ^RI H t H t ^RIHtHtHt^RIHtHt^R IHtHt^R IHtHt^R IHt!R R ]RR7tRRltRRlt]!RR]44t R.t!R#)N) Annotated) functional)TorchvisionBackend) BatchFeature)group_images_by_shapereorder_images) ImageInputPILImageResamplingSizeDict) ImagesKwargsUnpack) TensorTypeauto_docstring)IMAGENET_STANDARD_MEANIMAGENET_STANDARD_STD) positive_intc0a]tRt^%toRtV3RltRtVtR#)NeoMMEImageProcessorKwargsz patch_size (`int`, *optional*, defaults to `self.patch_size`): Side, in pixels, of one patch token. The image is padded to a whole multiple of it. max_side (`int`, *optional*): Longest-side cap in pixels. Unset means no longest-side resize. c<V^8dQh/S[S[S[!43,;R&S[S[R,S[!43,;R&#) patch_sizeNmax_side)rintr)format __classdict__s"s/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/neomme/image_processing_neomme.py __annotate__'NeoMMEImageProcessorKwargs.__annotate__%s;#|~-..d LN233N)__name__ __module__ __qualname____firstlineno____doc____annotate_func____static_attributes____classdictcell__)rs@rrr%srrF)totalc,V^8dQhRRR\RR/#)rimages torch.Tensorrreturnr)rs"rrr1s!  ^   rcVPwr#rEWA,pWQ,pVPW#WaWq4pVP^^^^^^4pVPW&V,R4pV#)z Convert 3D array image of shape (image_height, image_width, num_channels) into 2D array of patches of shape (num_patches_height * num_patches_width, patch_size * patch_size * num_channels). )shapereshapepermute) r+r batch_size num_channels image_height image_widthnum_patches_heightnum_patches_width patched_images && rconvert_image_to_patchesr;1sw ;A,,7Jl%3#1NN"4BSM"))!Q1a;M!))*K\6\^`aM rc V^8dQhR\R\R\R,R\R,R\\\3,/#)rheightwidthrNsizer-)rr tuple)rs"rrrAsG&)&)3&)s&)cDj&)PX[_P_&)dijmorjrds&)rcVe VPMRpVe VPMRpVe$W,V8dW@V,, R,MRpVe\Wb\W4, 4pVe \WeW,, R,4pVR8XdW3#\^\ W,44p\^\ W,44pVRJ;'d\Wx4V8;'gVRJ;'d Wx,V8p V 'd\^\ P !W,44p\^\ P !W,44pVe;Wx,V8d.Wx8d\^WX,4pWx3#\^WW,4pWx3#VeW,V8dWx,V8dWx3#\^\ P!W,44p \^\ P!W,44p VRJ;'d\W4V8;'gVRJ;'d W,V8p V 'gW3#Wx3#)zHCompute integer height and width from the configured image size targets.Ng?g?) min_pixels max_pixelsminmaxroundmathfloorceil) r=r>rr?rBrCscaleresized_height resized_width exceeds_cap ceiled_height ceiled_widths &&&& rget_resize_output_sizerPAs$($4$J$($4$J6@6LQWQ_blQlZE> *s 2ruEEc&&889E&.9cAB |}E&.12N5/0M4'YYC,NQY,Y$NN>#AJ#NQ 6> :;Atzz%-89  !n&Dz&Q.!$Q (C!D,,!$Az'C D ,,&.:"=  )Z 7!0 0Atyy89 1dii 67 t+[[M0PS[0[  d " P P}'Cj'P  . .  ((rcaa]tRt^jtoRt]t]Pt ] t ] t RtRtRtRtRt^ tRtRtRR.tV3RlV3Rllt]V3RlV3R ll4tV3R lV3R lltV3R lR ltV3RlRltV3RlRltRV3RlRlltRtVt V;t!#)NeoMMEImageProcessora{ Constructs an image processor for NeoMME. The processor converts each image to RGB and splits it into row-major patches. It preserves the original size by default, but `max_side` or a pixel-area `size` can resize images. It pads the bottom and right edges to complete the final patch. A batch returns one concatenated patch tensor and one grid height and width pair per image. TN pixel_values image_grid_hwc0<V^8dQhRS[S[,/#)rkwargs)r r)rrs"rr!NeoMMEImageProcessor.__annotate__s##(B!C#rc (<\SV`!R/VBR#)Nr )super__init__)selfrV __class__s&,rrZNeoMMEImageProcessor.__init__s "6"rc<<V^8dQhRS[RS[S[,RS[/#)rr+rVr-)r r rr)rrs"rrrWs)444v>X7Y4^j4rc &<\SV`!V3/VB#)z Returns: [`BatchFeature`] with `pixel_values` of shape `(total_patches, 3 * patch_size ** 2)` and `image_grid_hw` of shape `(batch_size, 2)`. `pixel_values` concatenates patches from every image in the batch. )rY preprocess)r[r+rVr\s&&,rr`NeoMMEImageProcessor.preprocesssw!&3F33rc <V^8dQhRS[/#)rr-)r@)rrs"rrrWs==u=rc <VR,pVe'\\V44RR08wd \R4hVR,fVPRR4\SV`!R/VB#)r?NrBrCz4size must contain exactly min_pixels and max_pixels. do_resizer )setdict ValueErrorpoprY_validate_preprocess_kwargs)r[rVr?r\s&, rri0NeoMMEImageProcessor._validate_preprocess_kwargss]f~  DJL,3O OST T &> ! JJ{D )w2T)rlr?rm antialiasr0)r1rPr resize) r[rlrr?rmr=r>rKrLs &&&&& r_resize_to_budget&NeoMMEImageProcessor._resize_to_budgets\ BR(>vh(]%  *vo =L~C{{HPT{UUrc B<V^8dQhRRRS[RS[RS[S[3,/#)rrlr,rr-)rr@)rrs"rrrWs1...C.ER`begjRjLk.rcVPR,VPR,rCV)V,)V)V,)reWR,V, pWb,V, pV'g V'd\P!V^^W.^R7pWV3#)r)fillrqr0)r1tvFpad) r[rlrr=r> grid_height grid_width pad_height pad_widths &&& r_pad_to_patch_grid'NeoMMEImageProcessor._pad_to_patch_gridsw BR%+Gz$9":uf >R!%!7!7X`!a,: "5 )&<((>@TU/D^fv/w,,#%  %3%9%9%; !E>7;6M6Mn^h6i 3NK!77 LV_N/G~Wa.b $U +).z*+c..AA*  &&<&&>@TU &':]!;'   rc,<V^8dQhRS[RS[RS[/#)rr=r>r-r.)rrs"rrrWs(DD#DcDZ]Drc T;'g/pVPRVP4pVPRVP4pVPVPRVP4R7R,pVPRVP 4'd\ WWV4wrV)V,)V)V,),#)z Return the number of image patches for one image, excluding row markers. The method applies the effective resize settings, then rounds each image dimension up to a whole patch. Values in `images_kwargs` override the processor settings. rrr?)r?rd)getrr_standardize_kwargsr?rdrP)r[r=r> images_kwargsrrr?s&&&& rget_number_of_image_patches0NeoMMEImageProcessor.get_number_of_image_patchess&++ "&&|T__E  $$Z?'']->->vtyy-Q'RSYZ   [$.. 9 926(QMFJ&'eVz-A+BCCrr gp?)N)"r!r"r#r$r%r valid_kwargsr BILINEARrmrrrrdo_convert_rgbrdrrrrrr?model_input_namesrZrr`rirsr~rrr'r( __classcell__)r\rs@@rrRrRjs.L!**H'J%INIJNLJH D'9##444== V V..0 0 dDDDrrR)"rGtypingrrtorchvision.transforms.v2rrximage_processing_backendsrimage_processing_utilsrimage_transformsrr image_utilsr r r processing_utilsr r utilsrrutils.constantsrrutils.type_validatorsrrr;rPrR__all__r rrrst*  7;2ECC4/L1 4U 4  &)RJD-JDJDZ " "r