+hn'bSSKJrJrJrJr SSKrSSKrSSK r SSK J r SSK J r SSKJrJrJr SSKJr SS KJrJr \R."\5rS \\\R4R4\ R644S jrS \\\R4R4\ R644S jr"SS\5rg))ListOptionalTupleUnionN) UNet2DModel)RePaintScheduler)PIL_INTERPOLATION deprecatelogging) randn_tensor)DiffusionPipelineImagePipelineOutputimagec JSn[SSUSS9 [U[R5(aU$[U[R R 5(aU/n[US[R R 5(aUSR up#SX#45up#UVs/sH6n[R"URX#4[SS 95SSS24PM8 nn[R"USS 9n[R"U5R[R5S - nURSS S S5nSU-S- n[R"U5nU$[US[R5(a[R "USS9nU$s snf)NzThe preprocess method is deprecated and will be removed in diffusers 1.0.0. Please use VaeImageProcessor.preprocess(...) instead preprocessz1.0.0F) standard_warnrc30# UH oUS-- v M g7f)N.0xs q/home/james-whalen/.local/lib/python3.13/site-packages/diffusers/pipelines/deprecated/repaint/pipeline_repaint.py $_preprocess_image..+s*6aAE 6lanczosresampleaxiso@rg@g?dim)r isinstancetorchTensorPILImagesizenparrayresizer concatenateastypefloat32 transpose from_numpycat)rdeprecation_messagewhis r_preprocess_imager<!s_] lG%8N%&& E399?? + +%(CIIOO,,Qx}}*A6*ejkej`a!((A64Ei4P(QRSWYZSZ[ejku1-&&rzz2U:1a+e c!  ' L E!Hell + + %Q' Lls)=F maskc [U[R5(aU$[U[RR5(aU/n[US[RR5(aUSR upSX45upUVs/sHEn[ R"URS5RX4[SS95SSS24PMG nn[ R"USS9nUR[ R5S- nSXS:'S XS:'[R"U5nU$[US[R5(a[R"USS 9nU$s snf) Nrc30# UH oUS-- v M g7f) Nrrs rr#_preprocess_mask..@s+FqAF FrLnearestr r"r$?r%r')r)r*r+r,r-r.r/r0convertr1r r2r3r4r6r7)r=r9r:ms r_preprocess_maskrG8s?$ %% D#))// * *v$q'399??++Aw||+QF+quvqulm3..v@QR[@\.]^_cef_fgquv~~d+{{2::&.CZS[% K DGU\\ * *yy1% KwsA E?c^\rSrSr%Sr\\S'\\S'SrS\S\4U4Sjjr \ R"5SS\ \ R\RR4S\ \ R\RR4S\S \S \S \S \\ \ R&\\ R&4S \\S\S\ \\44Sjj5rSrU=r$)RePaintPipelineLa Pipeline for image inpainting using RePaint. This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods implemented for all pipelines (downloading, saving, running on a particular device, etc.). Parameters: unet ([`UNet2DModel`]): A `UNet2DModel` to denoise the encoded image latents. scheduler ([`RePaintScheduler`]): A `RePaintScheduler` to be used in combination with `unet` to denoise the encoded image. unet schedulerc@>[TU]5 URXS9 g)N)rKrL)super__init__register_modules)selfrKrL __class__s rrORePaintPipeline.__init__^s  4=r mask_imagenum_inference_stepseta jump_length jump_n_sample generator output_type return_dictreturnc Un [U 5n U RURURRS9n [ U5nURURURRS9nU R Sn [U[5(a*[U5U :wa[S[U5SU S35eU R n [XURURRS9nURRX5X`R5 X@RlURRSS-n [U[5(aUSOUn[!UR#URR55HlupX:aDURX5R$nURR'UXXU5R(nOURR+XU5nUn Mn US- S -R-SS5nUR/5R1SSS S5R35nUS :XaUR5U5nU (dU4$[7US 9$) a The call function to the pipeline for generation. Args: image (`torch.Tensor` or `PIL.Image.Image`): The original image to inpaint on. mask_image (`torch.Tensor` or `PIL.Image.Image`): The mask_image where 0.0 define which part of the original image to inpaint. num_inference_steps (`int`, *optional*, defaults to 1000): The number of denoising steps. More denoising steps usually lead to a higher quality image at the expense of slower inference. eta (`float`): The weight of the added noise in a diffusion step. Its value is between 0.0 and 1.0; 0.0 corresponds to DDIM and 1.0 is the DDPM scheduler. jump_length (`int`, *optional*, defaults to 10): The number of steps taken forward in time before going backward in time for a single jump ("j" in RePaint paper). Take a look at Figure 9 and 10 in the [paper](https://huggingface.co/papers/2201.09865). jump_n_sample (`int`, *optional*, defaults to 10): The number of times to make a forward time jump for a given chosen time sample. Take a look at Figure 9 and 10 in the [paper](https://huggingface.co/papers/2201.09865). generator (`torch.Generator`, *optional*): A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make generation deterministic. output_type (`str`, `optional`, defaults to `"pil"`): The output format of the generated image. Choose between `PIL.Image` or `np.array`. return_dict (`bool`, *optional*, defaults to `True`): Whether or not to return a [`ImagePipelineOutput`] instead of a plain tuple. Example: ```py >>> from io import BytesIO >>> import torch >>> import PIL >>> import requests >>> from diffusers import RePaintPipeline, RePaintScheduler >>> def download_image(url): ... response = requests.get(url) ... return PIL.Image.open(BytesIO(response.content)).convert("RGB") >>> img_url = "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/repaint/celeba_hq_256.png" >>> mask_url = "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/repaint/mask_256.png" >>> # Load the original image and the mask as PIL images >>> original_image = download_image(img_url).resize((256, 256)) >>> mask_image = download_image(mask_url).resize((256, 256)) >>> # Load the RePaint scheduler and pipeline based on a pretrained DDPM model >>> scheduler = RePaintScheduler.from_pretrained("google/ddpm-ema-celebahq-256") >>> pipe = RePaintPipeline.from_pretrained("google/ddpm-ema-celebahq-256", scheduler=scheduler) >>> pipe = pipe.to("cuda") >>> generator = torch.Generator(device="cuda").manual_seed(0) >>> output = pipe( ... image=original_image, ... mask_image=mask_image, ... num_inference_steps=250, ... eta=0.0, ... jump_length=10, ... jump_n_sample=10, ... generator=generator, ... ) >>> inpainted_image = output.images[0] ``` Returns: [`~pipelines.ImagePipelineOutput`] or `tuple`: If `return_dict` is `True`, [`~pipelines.ImagePipelineOutput`] is returned, otherwise a `tuple` is returned where the first element is a list with the generated images. )devicedtyperz/You have passed a list of generators of length z+, but requested an effective batch size of z@. Make sure the batch size matches the length of the generators.)rZr_r`r%r&rDrpil)images)r<to_execution_devicerKr`rGshaper)listlen ValueErrorr rL set_timestepsrW timesteps enumerate progress_barsamplestep prev_sample undo_stepclampcpupermutenumpy numpy_to_pilr)rQrrUrVrWrXrYrZr[r\original_image batch_size image_shapet_lastr;t model_outputs r__call__RePaintPipeline.__call__bs.p*>:'**$2H2HPTPYPYP_P_*`%j1 ]]$*@*@ ]X #))!,  i & &3y>Z+GA#i.AQR&<'gi  %** [dF\F\dhdmdmdsdst $$%8}VlVlm ))!,q0$.y$$?$?IaLY d//0H0HIJDAz#yy299 ++L!N`ijvv00 JFKS''1- ##Aq!Q/557 % %%e,E8O"%00rTr)g rNraT)__name__ __module__ __qualname____firstlineno____doc__r__annotations__r model_cpu_offload_seqrOr*no_gradrr+r,r-intfloatr Generatorrstrboolrrr|__static_attributes__ __classcell__)rRs@rrIrILs&  ">[>5E> ]]_ $'MQ%* D1U\\399??23D1%,, 78D1! D1  D1  D1D1E%//43H"HIJD1c]D1D1 "E) *D1D1rTrI)typingrrrrrtr/ PIL.Imager,r*modelsr schedulersr utilsr r r utils.torch_utilsr pipeline_utilsrr get_loggerrloggerr-r+r<rGrIrrTrrs 0/ "+;;.D   H %U4%,,#FG.5syy !DE([1'[1rT