+hUSSKJrJrJrJr SSKrSSKrSSKJ r SSK J r SSK J r Jr SSKJr SSKJrJrJr SS KJr S S KJrJr S S KJr \"5(a SSKJs Jr Sr OSr \RB"\"5r#Sr$SSjr%"SS\5r&g))CallableListOptionalUnionN)XLMRobertaTokenizer)VaeImageProcessor)UNet2DConditionModelVQModel) DDIMScheduler)is_torch_xla_availableloggingreplace_example_docstring) randn_tensor)DiffusionPipelineImagePipelineOutput)MultilingualCLIPTFa Examples: ```py >>> from diffusers import KandinskyImg2ImgPipeline, KandinskyPriorPipeline >>> from diffusers.utils import load_image >>> import torch >>> pipe_prior = KandinskyPriorPipeline.from_pretrained( ... "kandinsky-community/kandinsky-2-1-prior", torch_dtype=torch.float16 ... ) >>> pipe_prior.to("cuda") >>> prompt = "A red cartoon frog, 4k" >>> image_emb, zero_image_emb = pipe_prior(prompt, return_dict=False) >>> pipe = KandinskyImg2ImgPipeline.from_pretrained( ... "kandinsky-community/kandinsky-2-1", torch_dtype=torch.float16 ... ) >>> pipe.to("cuda") >>> init_image = load_image( ... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main" ... "/kandinsky/frog.png" ... ) >>> image = pipe( ... prompt, ... image=init_image, ... image_embeds=image_emb, ... negative_image_embeds=zero_image_emb, ... height=768, ... width=768, ... num_inference_steps=100, ... strength=0.2, ... ).images >>> image[0].save("red_frog.png") ``` cnXS--nXS--S:waUS- nXS--nXS--S:waUS- nX2-XB-4$)Nrrr)hw scale_factornew_hnew_ws r/home/james-whalen/.local/lib/python3.13/site-packages/diffusers/pipelines/kandinsky/pipeline_kandinsky_img2img.py get_new_h_wrWs[ q E?a   q E?a    !5 55c#^\rSrSrSrSrS\S\S\S\ S\ 4 U4S jjr S r S r S%S jrS\R S\R S\R"S\R 4Sjr\R&"5\"\5S&S\\\\4S\\R \R4R4\\R \\R4R44S\R S\R S\\\\\4S\S\S\S\S\S\S\\\R<\\R<4S\\S \\\\\R /S 4S!\S"\ 4 S#jj55r!S$r"U=r#$)'KandinskyImg2ImgPipelineaa Pipeline for image-to-image generation using Kandinsky This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods the library implements for all the pipelines (such as downloading or saving, running on a particular device, etc.) Args: text_encoder ([`MultilingualCLIP`]): Frozen text-encoder. tokenizer ([`XLMRobertaTokenizer`]): Tokenizer of class scheduler ([`DDIMScheduler`]): A scheduler to be used in combination with `unet` to generate image latents. unet ([`UNet2DConditionModel`]): Conditional U-Net architecture to denoise the image embedding. movq ([`VQModel`]): MoVQ image encoder and decoder ztext_encoder->unet->movq text_encodermovq tokenizerunet schedulercv>[TU]5 URUUUUUS9 [USS5(a/S[ UR R R5S- -OSUl[USS5(a UR R ROSn[URUSSS9Ul g) N)r#r%r&r'r$r$rrbicubic)vae_scale_factorvae_latent_channelsresample reducing_gap) super__init__register_modulesgetattrlenr$configblock_out_channelsmovq_scale_factorlatent_channelsr image_processor)selfr#r$r%r&r'movq_latent_channels __class__s rr1!KandinskyImg2ImgPipeline.__init__ws  %  DK4QWY]C^C^A#dii&&99:Q> ?de DK4QWY]C^C^tyy//??de0!33 4  rc[[X-5U5n[X- S5nURRUSnXaU- 4$)Nr)minintmaxr' timesteps)r:num_inference_stepsstrengthdevice init_timestept_startrBs r get_timesteps&KandinskyImg2ImgPipeline.get_timestepssLC 3 >?ATU )91=NN,,WX6 777rcUc [X6XTS9nO>66q$..JiJilmJmprJrGr7stL NNNN334Il^M  (**62..11&9 484E4E$5F5 1 &778MST7U %?%Q%QRgmn%Q%o"//0E1/M &&!#z 1 fT/%::UVZ[jVkUlmV ~Q(OS11!0 1 s?33 )/)::J3K_J`ax/ |<33 !0 >>$&*#'#*L%1$:$:$=$=f$E !+::==fE HLHYHY/@PIZI E "$E -2215G%;%B%B1F[%\ "%;%@%@Acel%m "7==a@G0Q0X0XYZ\qst0u -0Q0V0V2GR1 - 0AABW]^A_ "II'=}&MNM).4UWq3r)s & #3Y"?@I)CCroriginal_samplesrTrBreturnc"[R"SSS[RS9nSU- n[R"USS9nUR UR UR S9nUR UR 5nXcS -nUR5n[UR5[UR5:a?URS 5n[UR5[UR5:aM?S Xc- S -nUR5n[UR5[UR5:a?URS 5n[UR5[UR5:aM?Xq-X--n U $) Ng-C6?g{Gz?i)rL?rrd)rErLg?rar) rjlinspacefloat32cumprodrOrErLflattenr4rM unsqueeze) r:rrTrBbetasalphasalphas_cumprodsqrt_alpha_prodsqrt_one_minus_alpha_prod noisy_sampless rrQ"KandinskyImg2ImgPipeline.add_noisesh vtTGuv15'**2B2I2IQaQgQg*hLL!1!8!89 (3s:)113/''(3/?/E/E+FF-77;O/''(3/?/E/E+FF&')B%Bs$J!$=$E$E$G!+112S9I9O9O5PP(A(K(KB(O %+112S9I9O9O5PP(:=V=^^ rrwimage image_embedsnegative_image_embedsrzheightwidthrCrDguidance_scalerxrK output_typecallbackcallback_steps return_dictc B [U[5(aSnO8[U[5(a [U5nO[ S[ U535eUR nUU -nU S:nURUUU UU5unnn[U[5(a[R"USS9n[U[5(a[R"USS9nU(aNURU SS9nURU SS9n[R"XC/SS9RURUS9n[U[5(dU/n[SU55(d)[ SUVs/sHn[ U5PM snS 35e[R"UVs/sHnURRUXv5PM! snSS9nURURUS9nUR R#U5S nURU SS9nUR$R'UUS 9 UR)XU5unn[+UR$R,R.U -5S - n[R0"U/U-URUS9nUR2R,R4ngUR85upgUR;UUUUXg4URUU UR$5n[=UR?U55GHunnU(a[R"U/S -5OUnUUS .nUR3UUUUSS9SnU(ajURAURBSSS9unn URES 5un!n"U RES 5unn#U!U U"U!- --n[R"UU#/SS9n[GUR$R,S5(a$UR$R,RHS;d URAURBSSS9unnUR$RKUUUU S9RLnUb-UU-S:Xa$U[OUR$SS5-n$U"U$UU5 [P(dGM[RRT"5 GM UR RWUSS9SnURY5 U S;a[ SU 35eURR[X-5nU(dU4$[]US9$s snfs snf)a Function invoked when calling the pipeline for generation. Args: prompt (`str` or `List[str]`): The prompt or prompts to guide the image generation. image (`torch.Tensor`, `PIL.Image.Image`): `Image`, or tensor representing an image batch, that will be used as the starting point for the process. image_embeds (`torch.Tensor` or `List[torch.Tensor]`): The clip image embeddings for text prompt, that will be used to condition the image generation. negative_image_embeds (`torch.Tensor` or `List[torch.Tensor]`): The clip image embeddings for negative text prompt, will be used to condition the image generation. negative_prompt (`str` or `List[str]`, *optional*): The prompt or prompts not to guide the image generation. Ignored when not using guidance (i.e., ignored if `guidance_scale` is less than `1`). height (`int`, *optional*, defaults to 512): The height in pixels of the generated image. width (`int`, *optional*, defaults to 512): The width in pixels of the generated image. num_inference_steps (`int`, *optional*, defaults to 100): The number of denoising steps. More denoising steps usually lead to a higher quality image at the expense of slower inference. strength (`float`, *optional*, defaults to 0.3): Conceptually, indicates how much to transform the reference `image`. Must be between 0 and 1. `image` will be used as a starting point, adding more noise to it the larger the `strength`. The number of denoising steps depends on the amount of noise initially added. When `strength` is 1, added noise will be maximum and the denoising process will run for the full number of iterations specified in `num_inference_steps`. A value of 1, therefore, essentially ignores `image`. guidance_scale (`float`, *optional*, defaults to 4.0): Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2. of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to the text `prompt`, usually at the expense of lower image quality. num_images_per_prompt (`int`, *optional*, defaults to 1): The number of images to generate per prompt. generator (`torch.Generator` or `List[torch.Generator]`, *optional*): One or a list of [torch generator(s)](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 generate image. Choose between: `"pil"` (`PIL.Image.Image`), `"np"` (`np.array`) or `"pt"` (`torch.Tensor`). callback (`Callable`, *optional*): A function that calls every `callback_steps` steps during inference. The function is called with the following arguments: `callback(step: int, timestep: int, latents: torch.Tensor)`. callback_steps (`int`, *optional*, defaults to 1): The frequency at which the `callback` function is called. If not specified, the callback is called at every step. return_dict (`bool`, *optional*, defaults to `True`): Whether or not to return a [`~pipelines.ImagePipelineOutput`] instead of a plain tuple. Examples: Returns: [`~pipelines.ImagePipelineOutput`] or `tuple` rz2`prompt` has to be of type `str` or `list` but is rrrd)rLrEc3# UH9n[U[RR[R45v M; g7fN)rhPILImagerjTensor).0is r 4KandinskyImg2ImgPipeline.__call__..s+Q5a:a#))//5<%%.?UfCf1gg "YY 4F'GQO --??NN))77;WW * 0 0q1Aq 0 I Ann))# * k #N(:a(? CC1g.} OCT   T B8L ##% 1 1hithuvw w$$00D8O"%00s2J]s "T &T)r9r7r) Nrdg333333?g@rNrNrT)$__name__ __module__ __qualname____firstlineno____doc__model_cpu_offload_seqrr rr r r1rHrUrrjr IntTensorrQno_gradrEXAMPLE_DOC_STRINGrrsrrrrr@float Generatorrboolr__static_attributes__ __classcell__)r<s@rr!r!as'&7 &  '  #  !  88, dDN,,||??   4 ]]_12<@#& #%&MQ%*GK #J1c49n%J1U\\399??D4FSYY__H]]^J1ll J1 %|| J1 "%T#Y"78 J1J1J1!J1J1J1 #J1E%//43H"HIJJ1c]J18S#u||$rs32 13' .C*))MM   H %&R6X10X1r