+h)8SSKJrJrJrJrJr SSKrSSKJrJ r SSK J r SSK J r JrJrJr SSKJr SS KJrJr \"5(a SSKJs Jr S rOS r\R6"\5rS rSS jr"SS\5r g))CallableDictListOptionalUnionN)UNet2DConditionModelVQModel) DDPMScheduler) deprecateis_torch_xla_availableloggingreplace_example_docstring) randn_tensor)DiffusionPipelineImagePipelineOutputTFae Examples: ```py >>> from diffusers import KandinskyV22Pipeline, KandinskyV22PriorPipeline >>> import torch >>> pipe_prior = KandinskyV22PriorPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-prior") >>> pipe_prior.to("cuda") >>> prompt = "red cat, 4k photo" >>> out = pipe_prior(prompt) >>> image_emb = out.image_embeds >>> zero_image_emb = out.negative_image_embeds >>> pipe = KandinskyV22Pipeline.from_pretrained("kandinsky-community/kandinsky-2-2-decoder") >>> pipe.to("cuda") >>> image = pipe( ... image_embeds=image_emb, ... negative_image_embeds=zero_image_emb, ... height=768, ... width=768, ... num_inference_steps=50, ... ).images >>> image[0].save("cat.png") ``` cnXS--nXS--S:waUS- nXS--nXS--S:waUS- nX2-XB-4$)Nrr)heightwidth scale_factor new_height new_widths p/home/james-whalen/.local/lib/python3.13/site-packages/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2.pydownscale_height_and_widthr>s[?*J a1$a q(I Q!#Q  $i&> >>c^\rSrSrSrSr/SQrS\S\S\ 4U4Sjjr S r \ S 5r \ S 5r\ S 5r\R""5\"\5S S SSSSSSSSS/4 S\\R*\\R*4S\\R*\\R*4S\S\S\S\S\S\\\R4\\R44S\\R*S\\S\S\\\\\/S4S \\4S!jj55rS"r U=r!$)#KandinskyV22PipelineHak Pipeline for text-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: scheduler (Union[`DDIMScheduler`,`DDPMScheduler`]): 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 Decoder to generate the image from the latents. z unet->movq)latents image_embedsnegative_image_embedsunet schedulermovqc>[TU]5 URUUUS9 S[URR R 5S- -Ulg)N)r%r&r'rr)super__init__register_moduleslenr'configblock_out_channelsmovq_scale_factor)selfr%r&r' __class__s rr*KandinskyV22Pipeline.__init__[sV    "#s499+;+;+N+N'ORS'S!TrcUc [XX2S9nO)_num_timestepsrAs r num_timesteps"KandinskyV22Pipeline.num_timesteps~s"""ridg@rNpilTr"r#r$rrnum_inference_stepsrBnum_images_per_promptr4 output_type return_dictcallback_on_step_end"callback_on_step_end_tensor_inputsc V ^URSS5nURSS5nUb [SSS5 Ub [SSS5 U bX[U4SjU 55(d>[STRS U Vs/sHnUTR;dMUPM sn35eTR nUTl[U[5(a[R"US S 9nURS U-n[U[5(a[R"US S 9nTR(aXURUS S 9nURUS S 9n[R"X!/S S 9RTRR US 9nTR"R%UUS 9 TR"R&nTRR(R*n[-X4TR.5up4TR1UUX44UR UUU TR"5n [3U5Tl[7TR9U55GH unnTR(a[R"U /S-5OU nSU0nTRUUSUSS9S nTR(atUR;U RSSS 9unnUR=S5unnUR=S5unnUTR>UU- --n[R"UU/SS 9n[ATR"R(S5(a$TR"R(RBS;d UR;U RSSS 9unnTR"REUUU US9S n U b\0n U Hn[G5UU U'M U "TUUU 5n!U!RSU 5n U!RSU5nU!RSU5nUb-UU-S :Xa$U[ITR"SS5-n"U"U"UU 5 [J(dGM[LRN"5 GM U S;a[SU 35eU S:XdTRPRSU SS9Sn#U S;aXU#S -S -n#U#RUS S5n#U#RW5RYS SS!S5R[5R]5n#U S":XaTR_U#5n#OU n#TRa5 U (dU#4$[cU#S#9$s snf)$a Function invoked when calling the pipeline for generation. Args: 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. 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. 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. latents (`torch.Tensor`, *optional*): Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image generation. Can be used to tweak the same generation with different prompts. If not provided, a latents tensor will ge generated by sampling using the supplied random `generator`. 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`). return_dict (`bool`, *optional*, defaults to `True`): Whether or not to return a [`~pipelines.ImagePipelineOutput`] instead of a plain tuple. callback_on_step_end (`Callable`, *optional*): A function that calls at the end of each denoising steps during the inference. The function is called with the following arguments: `callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict)`. `callback_kwargs` will include a list of all tensors as specified by `callback_on_step_end_tensor_inputs`. callback_on_step_end_tensor_inputs (`List`, *optional*): The list of tensor inputs for the `callback_on_step_end` function. The tensors specified in the list will be passed as `callback_kwargs` argument. You will only be able to include variables listed in the `._callback_tensor_inputs` attribute of your pipeline class. Examples: Returns: [`~pipelines.ImagePipelineOutput`] or `tuple` callbackNcallback_stepsz1.0.0zhPassing `callback` as an input argument to `__call__` is deprecated, consider use `callback_on_step_end`znPassing `callback_steps` as an input argument to `__call__` is deprecated, consider use `callback_on_step_end`c3@># UHoTR;v M g7fr>)_callback_tensor_inputs).0kr0s r 0KandinskyV22Pipeline.__call__..sF 7Y!-- -7Ysz2`callback_on_step_end_tensor_inputs` has to be in z , but found r)dim)r6r5)r5rr#F)sampletimestepencoder_hidden_statesadded_cond_kwargsrPr variance_type)learned learned_range)r4r"r$order)ptnprLlatentzIOnly the output types `pt`, `pil` and `np` are supported not output_type=rgT)force_not_quantizer])rfrLg?rrL)images)2popr allr8rW_execution_devicer@ isinstancelisttorchcatr7rErepeat_interleaver9r%r6r& set_timesteps timestepsr- in_channelsrr/r;r,rH enumerate progress_barsplitchunkrBhasattrrasteplocalsgetattr XLA_AVAILABLExm mark_stepr'decodeclampcpupermutefloatnumpy numpy_to_pilmaybe_free_model_hooksr)$r0r#r$rrrMrBrNr4r"rOrPrQrRkwargsrTrUrYr5 batch_sizersnum_channels_latentsitlatent_model_inputr` noise_pred variance_prednoise_pred_uncondnoise_pred_text_variance_pred_textcallback_kwargscallback_outputsstep_idximages$` r__call__KandinskyV22Pipeline.__call__sL::j$/$4d;   z   %  A  . 9#F 7YF C C DTEaEaDbbn|^pH|^vwbckokGkGbGpq|^pHoIJ ''- lD ) ) 99\q9L!''*-BB +T 2 2$)II.C$K !  + +'99:OUV9WL$9$K$KLagh$K$i ! 99&;%JPQRUUiioofVL $$%8$HNN,, #yy//;;26$BXBXY && -v =       NN  ")nd// :;DAq=A=]=]G9q=!9cj !/ > )&*"3! #  J//,6,<,%%.1D1DZkHk1ll "YY 4F'GQO --??NN))77;WW * 0 0q1Aq 0 I Ann))# *  G$/"$;A)/!OA&<#7aO#T *..y'B/33NLQ (8(<(<=TVk(l%#N(:a(? CC1g.} crsi98 3'ZZ-C))MM   H %4?C1,C1r