+h fSSKJrJrJrJrJr SSKrSSKJrJ r SSK J r SSK J r SSKJrJrJrJr SSKJr S S KJrJr S S KJr \"5(a SSKJs Jr S rOS r\R>"\ 5r!Sr""SS\5r#g))CallableDictListOptionalUnionN)CLIPTextModelWithProjection CLIPTokenizer)StableCascadeUNet)DDPMWuerstchenScheduler)is_torch_versionis_torch_xla_availableloggingreplace_example_docstring) randn_tensor)DiffusionPipelineImagePipelineOutput) PaellaVQModelTFa Examples: ```py >>> import torch >>> from diffusers import StableCascadePriorPipeline, StableCascadeDecoderPipeline >>> prior_pipe = StableCascadePriorPipeline.from_pretrained( ... "stabilityai/stable-cascade-prior", torch_dtype=torch.bfloat16 ... ).to("cuda") >>> gen_pipe = StableCascadeDecoderPipeline.from_pretrain( ... "stabilityai/stable-cascade", torch_dtype=torch.float16 ... ).to("cuda") >>> prompt = "an image of a shiba inu, donning a spacesuit and helmet" >>> prior_output = pipe(prompt) >>> images = gen_pipe(prior_output.image_embeddings, prompt=prompt) ``` c%2^\rSrSrSrSrSrSr/SQrS,S\ S\ S\ S\ S \ S \S S 4U4S jjjrSrS-S\\R&S\\R&S\\R&S\\R&4SjjrS.Sjr\S5r\S5r\S5rSr\R6"5\"\5S SSS S S S S SS S SSS S/4S\\R&\\R&4S \\ \\ 4S!\!S"\S#\\\ \\ 4S\\R&S\\R&S\\R&S\\R&S$\!S%\\\RD\\RD4S\\R&S&\\ S'\#S(\\$\!\!\%/S 4S)\\ 4 S*jj55r&S+r'U=r($)/StableCascadeDecoderPipeline:a Pipeline for generating images from the Stable Cascade model. 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: tokenizer (`CLIPTokenizer`): The CLIP tokenizer. text_encoder (`CLIPTextModelWithProjection`): The CLIP text encoder. decoder ([`StableCascadeUNet`]): The Stable Cascade decoder unet. vqgan ([`PaellaVQModel`]): The VQGAN model. scheduler ([`DDPMWuerstchenScheduler`]): A scheduler to be used in combination with `prior` to generate image embedding. latent_dim_scale (float, `optional`, defaults to 10.67): Multiplier to determine the VQ latent space size from the image embeddings. If the image embeddings are height=24 and width=24, the VQ latent shape needs to be height=int(24*10.67)=256 and width=int(24*10.67)=256 in order to match the training conditions. decoder text_encoderztext_encoder->decoder->vqgan)latentsprompt_embeds_poolednegative_prompt_embedsimage_embeddings tokenizer schedulervqganlatent_dim_scalereturnNcf>[TU]5 URUUUUUS9 URUS9 g)N)rrrr r!)r")super__init__register_modulesregister_to_config)selfrrrr r!r" __class__s t/home/james-whalen/.local/lib/python3.13/site-packages/diffusers/pipelines/stable_cascade/pipeline_stable_cascade.pyr&%StableCascadeDecoderPipeline.__init__\sF  %   1ABc `URuppX-S[XRR-5[XRR-54n Uc [ XXTS9nO>::# )K)22N(77N"nnVYW[n\ffO$$R(N,@,@,DDU[[NN $~~::#At~~'F'F'JR'O$OP  778 ,Q"03TT^^5T5T3T0T!U!/3TT^^5T5T3T0T!U"&"3"3!!&).:K:KF:Sjn#4# 0==bAM#+':'F'F'P'PQR'S$%((t/@/@/F/Fv(V 366T=N=N=T=T]c6d%778MST7U 3EEF[abEc ! ).I&!#z 1 fT/%::UVZ[jVkUlmV ~Q(OS11!0 1 s?33 )/)::J3K_J`ax/ |<33 !0 >>$>>::# *L:>9J9J&&))&1+::==fE%):K: 6 &P%]%]^`%a ",V,b,b,l,lmn,o ) &,2215G%;%>%>TEVEVE\E\ek%>%l "%;%B%B1F[]^%_ "%;%@%@Acelnp%q "399!# UHoTR;v M g7fN)_callback_tensor_inputs).0kr)s r+ .sF 7Y!-- -7Ysz2`callback_on_step_end_tensor_inputs` has to be in z , but found zCannot forward both `prompt`: z and `prompt_embeds`: z2. Please make sure to only forward one of the two.zeProvide either `prompt` or `prompt_embeds`. Cannot leave both `prompt` and `prompt_embeds` undefined.z2`prompt` has to be of type `str` or `list` but is z'Cannot forward both `negative_prompt`: z and `negative_prompt_embeds`: zu`prompt_embeds` and `negative_prompt_embeds` must have the same shape when passed directly, but got: `prompt_embeds` z != `negative_prompt_embeds` rR)allr6rvr`ralistr^r3)r)rfrgrBr"callback_on_step_end_tensor_inputsrxs` r+ check_inputs)StableCascadeDecoderPipeline.check_inputss . 9#F 7YF C C DTEaEaDbbn|^pH|^vwkokGkGbGpq|^pHoIJ   -";08N}o^00 ^ 5w  FC)@)@TZ\`IaIaQRVW]R^Q_`a a  &+A+M9/9JK*++]_   $)?)K""&<&B&BB --:-@-@,AB.445Q8C*L $+pHs D* D*cUR$ru_guidance_scaler)s r+guidance_scale+StableCascadeDecoderPipeline.guidance_scales###r-c URS:$)NrLrrs r+re8StableCascadeDecoderPipeline.do_classifier_free_guidances##a''r-cUR$ru)_num_timestepsrs r+ num_timesteps*StableCascadeDecoderPipeline.num_timesteps s"""r-c[R"S/5nSS/n[R"USU-- [R-S-5S-nX!nUR"U6nUR UR 5UR UR 5pSXe-S-R5[RS-- SU--U- nU$)NgMb?rrLg?r)rUtensorcospiclampr7r1acos)r)talphas_cumprods clamp_rangemin_varvarratios r+get_timestep_ratio_conditioningii%TT#**%wzz#**'=7=S(..0EHHsNCANQRR r- grLpilTrrrfnum_inference_stepsrrgr:r0 output_type return_dictcallback_on_step_endr}c  URnURRnX@l[ SS5(aU[ R :Xa [S5eURUUUUUS9 [U[5(a[ R"USS9nUb[U[5(aSnO3Ub![U[5(a [U5nOURSnXRSU--n Uc*Uc'URUUUU UR UUUUU S 9 unnnn UR (a[ R"Xy/5OUnUR (a,[ R"U[ R""U5/5OUnUR$R'UUS 9 UR$R(nUR+UXUUXUR$5n [UR$[,5(aUSS nOz[/UR$R0S 5(aUUR$R0R2(a0S UR$R0l[4R7S5 [/UR$S5(a/SUR$R8- n[ R:"USS9nO/n[U5Ul[?URAU55GHunn[UR$[,5(d[U5S:amURCURE5RG5U5nURIU RKS55RMU5RMU5nOURO5RQUR$R(S 5RIU RKS55RMU5nO/URIU RKS55RMU5nURUR (a[ R"U /S-5OU UR (a[ R"U/S-5OUUUS S9SnUR (a6URSS5unn[ RT"UUURV5n[UR$[,5(dUnUR$RYUUU U S9RZn Ub\0nUHn []5U UU 'M U"UUUU5n!U!R_SU 5n U!R_SU5nU!R_SU5n[`(dGM[bRd"5 GM U S;a[SU 35eU S:XdURfR0RhU -n URfRkU 5RlRoSS5n"U S:Xa?U"RqSSSS5RG5RO5Rs5n"OXU S:XaOU"RqSSSS5RG5RO5Rs5n"URuU"5n"OU n"URw5 U(dU"$[yU"5$)aW Function invoked when calling the pipeline for generation. Args: image_embedding (`torch.Tensor` or `List[torch.Tensor]`): Image Embeddings either extracted from an image or generated by a Prior Model. prompt (`str` or `List[str]`): The prompt or prompts to guide the image generation. num_inference_steps (`int`, *optional*, defaults to 12): 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 0.0): Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://huggingface.co/papers/2207.12598). `decoder_guidance_scale` is defined as `w` of equation 2. of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting `decoder_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. 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 `decoder_guidance_scale` is less than `1`). prompt_embeds (`torch.Tensor`, *optional*): Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not provided, text embeddings will be generated from `prompt` input argument. prompt_embeds_pooled (`torch.Tensor`, *optional*): Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not provided, pooled text embeddings will be generated from `prompt` input argument. negative_prompt_embeds (`torch.Tensor`, *optional*): Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input argument. negative_prompt_embeds_pooled (`torch.Tensor`, *optional*): Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not provided, negative_prompt_embeds_pooled will be generated from `negative_prompt` input argument. 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` [`~pipelines.ImagePipelineOutput`] if `return_dict` is True, otherwise a `tuple`. When returning a tuple, the first element is a list with the generated image embeddings. =2.2.0 when using `torch.bfloat16` dtype.)rgrBrr}rrONrL) rfr1r9r:rergrBrrrC)r1rK clip_sampleFz set `clip_sample` to be Falsebetasg?r)sampletimestep_ratioclip_text_pooledeffnetr) model_outputtimesteprr0rrBr)rFnprlatentzSOnly the output types `pt`, `np`, `pil` and `latent` are supported not output_type=rrr r)=_execution_devicerr2rr rUbfloat16r6r~r`r|catrarbr3rqre zeros_liker set_timesteps timestepsr@r hasattrr5rrXrYrcumprodr enumerate progress_barrlongcpuexpandsizer7floatdivchunklerprstep prev_samplelocalspop XLA_AVAILABLExm mark_stepr! scale_factordecoderrpermutenumpy numpy_to_pilmaybe_free_model_hooksr)#r)rrfrrrgrBrrrCr:r0rrrrr}r1r2r9r;rralphasrirrpredicted_latentspredicted_latents_textpredicted_latents_uncondcallback_kwargsrxcallback_outputsimagess# r+__call__%StableCascadeDecoderPipeline.__call__.sp'' ""- C ) )eu~~.Evw w  +'#9/Q   & - -$yy)9qA   *VS"9"9J  Jvt$<$<VJ&,,Q/J !69O9OPQ9RV`9` a  %;%CHLHZHZ%&;,0,L,L /+%9'=.KI[ I EA#Q(E // II+K L% // II')9)9:J)KL M!  $$%8$HNN,, && (PYdhdrdr  dnn&= > >!#2It~~,,m<<AVAVAbAb49%%1?@ 4>>7 + +4>>///F"]]6q9NN!)nd// :;DAqdnn.EFF~&*%)%I%I!&&(,,.Zh%iN%3%:%:7<<?%K%N%Nu%U%X%XY_%`N%&WWY]]4>>3K3KB3O%P%W%WX_XdXdefXg%h%k%klq%rN!"',,q/!:!=!=e!D!% 373S3Suyy'Q/Y`BFBbBbuyy.)9A)=>hv!5! !-!  ! //CTCZCZ[\C]@&(@$)JJ/GI_aeatat$u!dnn.EFF!"nn)).'# * k $/"$;A)/!OA&<#7aO#T *..y'B 0 4 4_m T )9)=)=>VXn)o&} [<^ ; ;efqers h&jj''44w>GZZ&&w/66<rsj98 C'1aa-C?))MM   H %(a+#4a+r-