+hMC bSSKrSSKJrJrJrJrJrJrJr SSK r SSK r SSK r SSKJs Jr SSKJrJrJrJrJr SSKJr SSKJrJr SSKJrJr SSK J!r!J"r"J#r#J$r$ SS K%J&r&J'r'J(r(J)r)J*r* SS K+J,r,J-r- SS K.J/r/ SS K0J1r1 SS K2J3r3J4r4J5r5J6r6J7r7J8r8 SSK9J:r:J;r;Jr>J?r? SSK@JArA \"5(aSSKBJCrC SSK2JDrD \D"5(a SSKEJFs JGrH SrIOSrI\5R"\K5rLSrMSS\\NS\\\O\ R4S\\\NS\\\Q4SjjrR"SS\>\?\$\#\"\!5rSg)N)AnyCallableDictListOptionalTupleUnion)CLIPImageProcessor CLIPTextModelCLIPTextModelWithProjection CLIPTokenizerCLIPVisionModelWithProjection) is_invisible_watermark_available)MultiPipelineCallbacksPipelineCallback)PipelineImageInputVaeImageProcessor)FromSingleFileMixinIPAdapterMixin StableDiffusionXLLoraLoaderMixinTextualInversionLoaderMixin) AutoencoderKLControlNetModelImageProjectionMultiControlNetModelUNet2DConditionModel)AttnProcessor2_0XFormersAttnProcessor)adjust_lora_scale_text_encoder)KarrasDiffusionSchedulers)USE_PEFT_BACKEND deprecateloggingreplace_example_docstringscale_lora_layersunscale_lora_layers)is_compiled_moduleis_torch_version randn_tensor)DiffusionPipelineStableDiffusionMixin)StableDiffusionXLPipelineOutput)StableDiffusionXLWatermarker)is_torch_xla_availableTFa/ Examples: ```py >>> # !pip install opencv-python transformers accelerate >>> from diffusers import StableDiffusionXLControlNetPipeline, ControlNetModel, AutoencoderKL >>> from diffusers.utils import load_image >>> import numpy as np >>> import torch >>> import cv2 >>> from PIL import Image >>> prompt = "aerial view, a futuristic research complex in a bright foggy jungle, hard lighting" >>> negative_prompt = "low quality, bad quality, sketches" >>> # download an image >>> image = load_image( ... "https://hf.co/datasets/hf-internal-testing/diffusers-images/resolve/main/sd_controlnet/hf-logo.png" ... ) >>> # initialize the models and pipeline >>> controlnet_conditioning_scale = 0.5 # recommended for good generalization >>> controlnet = ControlNetModel.from_pretrained( ... "diffusers/controlnet-canny-sdxl-1.0", torch_dtype=torch.float16 ... ) >>> vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch.float16) >>> pipe = StableDiffusionXLControlNetPipeline.from_pretrained( ... "stabilityai/stable-diffusion-xl-base-1.0", controlnet=controlnet, vae=vae, torch_dtype=torch.float16 ... ) >>> pipe.enable_model_cpu_offload() >>> # get canny image >>> image = np.array(image) >>> image = cv2.Canny(image, 100, 200) >>> image = image[:, :, None] >>> image = np.concatenate([image, image, image], axis=2) >>> canny_image = Image.fromarray(image) >>> # generate image >>> image = pipe( ... prompt, controlnet_conditioning_scale=controlnet_conditioning_scale, image=canny_image ... ).images[0] ``` num_inference_stepsdevice timestepssigmasc UbUb [S5eUbS[[R"UR5R R 55;nU(d[SURS35eUR"S X2S.UD6 URn[U5nX14$UbS[[R"UR5R R 55;nU(d[SURS35eUR"S XBS.UD6 URn[U5nX14$UR"U4S U0UD6 URnX14$) a Calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. Handles custom timesteps. Any kwargs will be supplied to `scheduler.set_timesteps`. Args: scheduler (`SchedulerMixin`): The scheduler to get timesteps from. num_inference_steps (`int`): The number of diffusion steps used when generating samples with a pre-trained model. If used, `timesteps` must be `None`. device (`str` or `torch.device`, *optional*): The device to which the timesteps should be moved to. If `None`, the timesteps are not moved. timesteps (`List[int]`, *optional*): Custom timesteps used to override the timestep spacing strategy of the scheduler. If `timesteps` is passed, `num_inference_steps` and `sigmas` must be `None`. sigmas (`List[float]`, *optional*): Custom sigmas used to override the timestep spacing strategy of the scheduler. If `sigmas` is passed, `num_inference_steps` and `timesteps` must be `None`. Returns: `Tuple[torch.Tensor, int]`: A tuple where the first element is the timestep schedule from the scheduler and the second element is the number of inference steps. zYOnly one of `timesteps` or `sigmas` can be passed. Please choose one to set custom valuesr3zThe current scheduler class zx's `set_timesteps` does not support custom timestep schedules. Please check whether you are using the correct scheduler.)r3r2r4zv's `set_timesteps` does not support custom sigmas schedules. Please check whether you are using the correct scheduler.)r4r2r2) ValueErrorsetinspect signature set_timesteps parameterskeys __class__r3len) schedulerr1r2r3r4kwargsaccepts_timesteps accept_sigmass r/home/james-whalen/.local/lib/python3.13/site-packages/diffusers/pipelines/controlnet/pipeline_controlnet_sd_xl.pyretrieve_timestepsrE}s}>!3tuu'3w/@/@AXAX/Y/d/d/i/i/k+ll .y/B/B.CDab  M)MfM'' !)n  ))   C(9(9):Q:Q(R(](](b(b(d$ee .y/B/B.CD_`  GvGG'' !)n  ))  3MFMfM''  ))cR.^\rSrSrSrSr/SQr/SQrS_S\S \ S \ S \ S \ S \ S\ \\\\\\4S\S\S\\S\S\4U4SjjjrS`S\S\\S\\R4S\S\S\\S\\S\\R8S\\R8S\\R8S \\R8S!\\S"\\4S#jjrSaS$jrS%r S&r!SbS)jr"S*r#ScS,jr$SaS-jr%SaS.jr&S/r'S0\RP4S1\R8S2\S3\RRS4\R84S5jjr*\+S65r,\+S75r-\+S85r.\+S95r/\+S:5r0\+S;5r1\+S<5r2\Rf"5\4"\55SSSSSS=SSSS>SSSS(SSSSSSSSS?SSS'S+S(S'SS@SSS@SSSSA/4&S\ \\\4S\\ \\\4SB\6SC\\SD\\SE\SF\\SG\\SH\\SI\S\\ \\\4S\\ \\\4S\\SJ\SK\\ \Rn\\Rn4SA\\R8S\\R8S\\R8S\\R8S \\R8SL\\6SM\\\R8SN\\SO\SP\\8\\94SQ\ \\\4SR\SS\ \\\4ST\ \\\4SU\\\4SV\\\4SW\\\4SX\\\\4SY\\\4SZ\\\\4S"\\S[\\ \:\\\8/S4\;\<4S\\\4LS]jj55r=S^r>U=r?$)d#StableDiffusionXLControlNetPipelineay Pipeline for text-to-image generation using Stable Diffusion XL with ControlNet guidance. 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.). The pipeline also inherits the following loading methods: - [`~loaders.TextualInversionLoaderMixin.load_textual_inversion`] for loading textual inversion embeddings - [`~loaders.StableDiffusionXLLoraLoaderMixin.load_lora_weights`] for loading LoRA weights - [`~loaders.StableDiffusionXLLoraLoaderMixin.save_lora_weights`] for saving LoRA weights - [`~loaders.FromSingleFileMixin.from_single_file`] for loading `.ckpt` files - [`~loaders.IPAdapterMixin.load_ip_adapter`] for loading IP Adapters Args: vae ([`AutoencoderKL`]): Variational Auto-Encoder (VAE) model to encode and decode images to and from latent representations. text_encoder ([`~transformers.CLIPTextModel`]): Frozen text-encoder ([clip-vit-large-patch14](https://huggingface.co/openai/clip-vit-large-patch14)). text_encoder_2 ([`~transformers.CLIPTextModelWithProjection`]): Second frozen text-encoder ([laion/CLIP-ViT-bigG-14-laion2B-39B-b160k](https://huggingface.co/laion/CLIP-ViT-bigG-14-laion2B-39B-b160k)). tokenizer ([`~transformers.CLIPTokenizer`]): A `CLIPTokenizer` to tokenize text. tokenizer_2 ([`~transformers.CLIPTokenizer`]): A `CLIPTokenizer` to tokenize text. unet ([`UNet2DConditionModel`]): A `UNet2DConditionModel` to denoise the encoded image latents. controlnet ([`ControlNetModel`] or `List[ControlNetModel]`): Provides additional conditioning to the `unet` during the denoising process. If you set multiple ControlNets as a list, the outputs from each ControlNet are added together to create one combined additional conditioning. scheduler ([`SchedulerMixin`]): A scheduler to be used in combination with `unet` to denoise the encoded image latents. Can be one of [`DDIMScheduler`], [`LMSDiscreteScheduler`], or [`PNDMScheduler`]. force_zeros_for_empty_prompt (`bool`, *optional*, defaults to `"True"`): Whether the negative prompt embeddings should always be set to 0. Also see the config of `stabilityai/stable-diffusion-xl-base-1-0`. add_watermarker (`bool`, *optional*): Whether to use the [invisible_watermark](https://github.com/ShieldMnt/invisible-watermark/) library to watermark output images. If not defined, it defaults to `True` if the package is installed; otherwise no watermarker is used. z6text_encoder->text_encoder_2->image_encoder->unet->vae) tokenizer tokenizer_2 text_encodertext_encoder_2feature_extractor image_encoder)latents prompt_embedsnegative_prompt_embedsadd_text_embeds add_time_idsnegative_pooled_prompt_embedsnegative_add_time_idsimageTNvaerLrMrJrKunet controlnetr@force_zeros_for_empty_promptadd_watermarkerrNrOc  >[T U]5 [U[[45(a [ U5nUR UUUUUUUUU U S9 [USS5(a/S[URRR5S- -OSUl [URSS9Ul[URSSS 9UlU bU O [!5n U (a[#5UlOSUlUR'U S 9 g) N) rXrLrMrJrKrYrZr@rNrOrXr+T)vae_scale_factordo_convert_rgbF)r`ra do_normalize)r[)super__init__ isinstancelisttuplerregister_modulesgetattrr?rXconfigblock_out_channelsr`rimage_processorcontrol_image_processorrr/ watermarkregister_to_config)selfrXrLrMrJrKrYrZr@r[r\rNrOr>s rDrd,StableDiffusionXLControlNetPipeline.__init__s  j4- 0 0-j9J %)#!/'  W^^bdikoVpVpc$((//*L*L&MPQ&Q Rvw0$BWBWhlm'8!224V[( $.=-H/NnNp 9;DN!DN =YZrFr^promptprompt_2r2num_images_per_promptdo_classifier_free_guidancenegative_promptnegative_prompt_2rQrRpooled_prompt_embedsrU lora_scale clip_skipc  U=(d URnU b[U[5(aXlURb8[ (d[ URU 5 O[URU 5 URb8[ (d[ URU 5 O[URU 5 [U[5(aU/OUnUb [U5nOURSnURbURUR/O UR/nURbURUR/O UR/nUGcU=(d Un[U[5(aU/OUn/nX/n[UUU5GHQunnn[U[5(aUR!UU5nU"USUR"SSS9nUR$nU"USSS9R$nURS URS :ah[&R("UU5(dLUR+USS2UR"S - S 245n[,R/S UR"S U35 U"UR1U5SS 9nU cUSR2S:XaUSn U cUR4SnOUR4U S-*nUR7U5 GMT [&R8"US S9nUSL=(a UR:R<nU(a8U c5U(a.[&R>"U5n [&R>"U 5n GOU(GaU GcU=(d SnU=(d Un[U[5(aX/-OUn[U[5(aX/-OUnUb;[AU5[AU5La$[CS[AU5S[AU5S35eU[U5:wa![ESUS[U5SUSUS3 5eXg/n/n[UUU5Hunnn[U[5(aUR!UU5nURS nU"USUSSS9nU"UR$R1U5SS 9n U cU SR2S:XaU Sn U R4Sn UR7U 5 M [&R8"US S9n URb%UR1URRFUS9nO$UR1URHRFUS9nURunnn URKS US 5nURMUU-US 5nU(aU RS nURb%U R1URRFUS9n O$U R1URHRFUS9n U RKS US 5n U RMX-US 5n U RKS U5RMUU-S 5n U(a%U RKS U5RMUU-S 5n URb6[U[5(a![ (a[OURU 5 URb6[U[5(a![ (a[OURU 5 XX4$)a Encodes the prompt into text encoder hidden states. Args: prompt (`str` or `List[str]`, *optional*): prompt to be encoded prompt_2 (`str` or `List[str]`, *optional*): The prompt or prompts to be sent to the `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is used in both text-encoders device: (`torch.device`): torch device num_images_per_prompt (`int`): number of images that should be generated per prompt do_classifier_free_guidance (`bool`): whether to use classifier free guidance or not negative_prompt (`str` or `List[str]`, *optional*): The prompt or prompts not to guide the image generation. If not defined, one has to pass `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is less than `1`). negative_prompt_2 (`str` or `List[str]`, *optional*): The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and `text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders 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. 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. pooled_prompt_embeds (`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_pooled_prompt_embeds (`torch.Tensor`, *optional*): Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` input argument. lora_scale (`float`, *optional*): A lora scale that will be applied to all LoRA layers of the text encoder if LoRA layers are loaded. clip_skip (`int`, *optional*): Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that the output of the pre-final layer will be used for computing the prompt embeddings. Nr max_lengthTpt)paddingr| truncationreturn_tensorslongest)r~rr^z\The following part of your input was truncated because CLIP can only handle sequences up to z tokens: output_hidden_statesr+dimz?`negative_prompt` should be the same type to `prompt`, but got z != .z`negative_prompt`: z has batch size z, but `prompt`: zT. Please make sure that passed `negative_prompt` matches the batch size of `prompt`.)dtyper2)(_execution_devicerer _lora_scalerLr"r r&rMstrr?shaperJrKziprmaybe_convert_promptmodel_max_length input_idstorchequal batch_decodeloggerwarningtondim hidden_statesappendconcatrjr[ zeros_liketype TypeErrorr7rrYrepeatviewr')!rprrrsr2rtrurvrwrQrRrxrUryrz batch_size tokenizers text_encodersprompt_embeds_listpromptsrJrL text_inputstext_input_idsuntruncated_ids removed_textzero_out_negative_prompt uncond_tokensnegative_prompt_embeds_listr| uncond_inputbs_embedseq_len_s! rD encode_prompt1StableDiffusionXLControlNetPipeline.encode_prompt/st1411  !j7W&X&X)   ,''243D3DjQ%d&7&7D"".''243F3F S%d&9&9:F'44&&  VJ&,,Q/J<@>>;Udnnd&6&67\`\l\l[m 8<8I8I8UT   3 3 4\`\o\o[p   )6H%/#%>%>zHH"$ (G36w M3Z/ <d$?@@!66vyIF'((99##'  "-!6!6"+FIVZ"["e"e"((,0D0DR0HHQVQ\Q\"ORR$-#9#9/!YMgMgjkMknpMpJp:q#rLNN%667yP !-^->->v-F]a b (/M!4D4I4IQ4N+8+;($$1$?$?$CM%2$?$?)a-@P$QM"))-8I4[L"LL);DM$3d#:#gt{{?g?g &+A+INf%*%5%5m%D ",1,<,<=Q,R ) (-C-K-3O 1 D_ AK?\_@`@`j+<?PRU4V4V 00\m  !d6l$:O&OUVZ[jVkUlmV ~Q(s?33 )/)::J3K_J`ax/ |<33 "1 D *, 'STYY , ,b  ',I,P,PQRTi,j,o,o00"- )    ($ @AAFVFV#D$5$5zB    *$ @AAFVFV#D$7$7D6JiirFcd[URR55Rn[ U[ R 5(dURUSS9RnURX%S9nU(aqURUSS9RSnURUSS9nUR[ R"U5SS9RSnURUSS9nXg4$URU5RnURUSS9n[ R"U5n X4$) Nr})rr2rTrrrr)nextrOr<rrerTensorrN pixel_valuesrrrepeat_interleaver image_embeds) rprWr2rtrrimage_enc_hidden_statesuncond_image_enc_hidden_statesruncond_image_embedss rD encode_image0StableDiffusionXLControlNetPipeline.encode_images?T''2245;;%..**5*FSSE4 &*&8&8UY&8&Z&h&hik&l #&=&O&OPekl&O&m #-1-?-?  'd.@.mB. *.L-]-]%1.^. *+J J--e4AAL'99:OUV9WL"'"2"2<"@ 4 4rFc /nU(a/nUGc&[U[5(dU/n[U5[URRR 5:waB[ S[U5S[URRR 5S35e[XRRR 5Hhup[U [5(+n URXSU 5upURU SSS245 U(dMPWRU SSS245 Mj OEUH?n U(a$U RS5upWRU 5 URU 5 MA /n[U5Hwup[R"U /U-SS9n U(a2[R"WU /U-SS9n [R"X/SS9n U RUS9n URU 5 My U$) NzK`ip_adapter_image` must have same length as the number of IP Adapters. Got images and z IP Adapters.r^r+rr)r2)rerfr?rYencoder_hid_projimage_projection_layersr7rrrrchunk enumeratercatr)rpip_adapter_imageip_adapter_image_embedsr2rtrurnegative_image_embedssingle_ip_adapter_imageimage_proj_layeroutput_hidden_statesingle_image_embedssingle_negative_image_embedsis rDprepare_ip_adapter_image_embedsCStableDiffusionXLControlNetPipeline.prepare_ip_adapter_image_embeds7s9 &$& ! " *.55$4#5 #$DII,F,F,^,^(__ abefvbwaxyEFIJNJSJSJdJdJ|J|F}E~~KL>A ))"<"<"T"T>9'+55E*W&W#DHDUDU+Q8KEA###$7a$@A..)001MdTUg1VW>(?#.H[HaHabcHdE0)001MN##$78 (? #%&/ &= "A"'))-@,ADY,Y_`"a */4yy:OPQ:R9SVk9kqr/s,&+ii1M0cij&k#"5"8"8"8"G  # * *+> ?'>'&rFcnS[[R"URR5R R 55;n0nU(aX$S'S[[R"URR5R R 55;nU(aXS'U$)Neta generator)r8r9r:r@stepr<r=)rprr accepts_etaextra_step_kwargsaccepts_generators rDprepare_extra_step_kwargs=StableDiffusionXLControlNetPipeline.prepare_extra_step_kwargses s7#4#4T^^5H5H#I#T#T#Y#Y#[\\  '*e $(3w/@/@ATAT/U/`/`/e/e/g+hh -6k *  rF?c ^Ub6[U[5(aUS::a[SUS[U5S35eUbX[ U4SjU55(d>[STR SUVs/sHnUTR ;dMUPM sn35eUbUb[SUS US 35eUbUb[S US US 35eUcUc [S 5eUbA[U[ 5(d,[U[5(d[S [U535eUbA[U[ 5(d,[U[5(d[S[U535eUbUb[SUSUS 35eUbUb[SUSUS 35eUbCUb@URUR:wa&[SURSURS35eUbU c [S5eUbU c [S5e[TR[5(aW[U[5(aB[RS[TRR5S[U5S35 [[ S5=(a8 [TR["R$R&R(5n[TR[*5(d0U(a=[TRR,[*5(aTR/X1U5 GO[TR[5(d0U(a[TRR,[5(a[U[5(d [1S5e[3SU55(a [S5e[U5[TRR5:wa8[S[U5S[TRR5S35eUHnTR/UX5 M Oe[TR[*5(d0U(aJ[TRR,[*5(a![U [45(d [1S 5eO[TR[5(d0U(a[TRR,[5(a[U [5(a#[3S!U 55(a [S5eOO[U [5(a7[U 5[TRR5:wa [S"5eOe[U[6[45(dU/n[U[6[45(dU/n[U5[U5:wa$[S#[U5S$[U5S%35e[TR[5(a[U5[TRR5:wa[[S&US'[U5S([TRR5S)[TRR5S3 5e[9X5HHunnUU:a[S*US+US35eUS,:a[S*US-35eUS.:dM;[S/US035e U bU b [S15eU b\[U [5(d[S2[U 535eU SR:S3;a[S4U SR:S535eggs snf)6Nrz5`callback_steps` has to be a positive integer but is z of type rc3@># UHoTR;v M g7fN)_callback_tensor_inputs).0krps rD CStableDiffusionXLControlNetPipeline.check_inputs..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.z Cannot forward both `prompt_2`: 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 z4`prompt_2` has to be of type `str` or `list` but is z'Cannot forward both `negative_prompt`: z and `negative_prompt_embeds`: z)Cannot forward both `negative_prompt_2`: zu`prompt_embeds` and `negative_prompt_embeds` must have the same shape when passed directly, but got: `prompt_embeds` z != `negative_prompt_embeds` zIf `prompt_embeds` are provided, `pooled_prompt_embeds` also have to be passed. Make sure to generate `pooled_prompt_embeds` from the same text encoder that was used to generate `prompt_embeds`.zIf `negative_prompt_embeds` are provided, `negative_pooled_prompt_embeds` also have to be passed. Make sure to generate `negative_pooled_prompt_embeds` from the same text encoder that was used to generate `negative_prompt_embeds`.z You have z! ControlNets and you have passed z= prompts. The conditionings will be fixed across the prompts.scaled_dot_product_attentionz5For multiple controlnets: `image` must be type `list`c3B# UHn[U[5v M g7frrerfrrs rDrrs8%QZ4((%zEA single batch of multiple conditionings are supported at the moment.zbFor multiple controlnets: `image` must have the same length as the number of controlnets, but got rz ControlNets.zLFor single controlnet: `controlnet_conditioning_scale` must be type `float`.c3B# UHn[U[5v M g7frrrs rDrrsR4Qqz!T**4QrzFor multiple controlnets: When `controlnet_conditioning_scale` is specified as `list`, it must have the same length as the number of controlnetsz`control_guidance_start` has z* elements, but `control_guidance_end` has zI elements. Make sure to provide the same number of elements to each list.z`control_guidance_start`: z has z elements but there are z- controlnets available. Make sure to provide zcontrol guidance start: z4 cannot be larger or equal to control guidance end: rz can't be smaller than 0.rzcontrol guidance end: z can't be larger than 1.0.zProvide either `ip_adapter_image` or `ip_adapter_image_embeds`. Cannot leave both `ip_adapter_image` and `ip_adapter_image_embeds` defined.z:`ip_adapter_image_embeds` has to be of type `list` but is )rzF`ip_adapter_image_embeds` has to be a list of 3D or 4D tensors but is D)reintr7rallrrrfrrZrrrr?netshasattrFr_dynamo eval_frameOptimizedModuler _orig_mod check_imageranyfloatrgrr)rprrrsrWcallback_stepsrvrwrQrRrxrrrUcontrolnet_conditioning_scalecontrol_guidance_startcontrol_guidance_end"callback_on_step_end_tensor_inputsr is_compiledimage_startends` rD check_inputs0StableDiffusionXLControlNetPipeline.check_inputsvsc&  %z.#/N/NR`deReGGWX(),  . 9#F 7YF C C DTEaEaDbbn|^pH|^vwbckokGkGbGpq|^pHoIJ   -";08N}o^00  !m&?28*>R4QRRR$%lmmS94@@SIfEgkn$$lF!D 505$-@@&<%= ". >>$8#9 % &#.B*C C/4J0K/LLvwz|PxQwRR[\  doo'; < <)*c$//2F2F.GG 01G0HcRhNiMjkCDGHLHWHWH\H\D]C^^KLOPTP_P_PdPdLeKffgh4KJE3| .ug5ijminnops{ #;E7B[!\]]Sy #9#>X!YZZL  ',C,O^  # .5t<< PQUVmQnPop)+00> \]tuv]w]|]|\}}~? /mpHs /]]c[U[RR5n[U[R5n[U[ R 5n[U[5=(a' [US[RR5n[U[5=(a [US[R5n[U[5=(a [US[ R 5n U(d:U(d3U(d,U(d%U(dU (d[S[U535eU(aSn O [U5n Ub[U[5(aSn O6Ub![U[5(a [U5n OUbURSn U S:waU W :wa[SU SU 35egg)Nrzimage must be passed and be one of PIL image, numpy array, torch tensor, list of PIL images, list of numpy arrays or list of torch tensors, but is r^zdIf image batch size is not 1, image batch size must be same as prompt batch size. image batch size: z, prompt batch size: )rePILImagerrnpndarrayrfrrr?rrr7) rprWrrrQ image_is_pilimage_is_tensor image_is_npimage_is_pil_listimage_is_tensor_listimage_is_np_listimage_batch_sizeprompt_batch_sizes rDr/StableDiffusionXLControlNetPipeline.check_image4s!%9 $UELL9  3 &ud3] 58SYY__8])%6]:eAhPUP\P\;]%eT2Wz%(BJJ7W#%($fgklqgrfst   "5z   *VS"9"9 !   Jvt$<$< #F   & - 3 3A 6  q %59J%JvxHwII^_p^qr &K rFFc 2URRXUS9R[RS9nUR Sn U S:XaUn OUn UR U SS9nURXgS9nU(a!U (d[R"U/S-5nU$)N)heightwidthrrr^rrr+)rm preprocessrrfloat32rrr) rprWrrrrtr2rru guess_moder repeat_bys rD prepare_image1StableDiffusionXLControlNetPipeline.prepare_imageZs,,77TY7Z]]didqdq]r ;;q> q "I.I'' q'94 &zIIugk*E rFc VUU[U5UR-[U5UR-4n [U[5(a*[ U5U:wa[ S[ U5SUS35eUc [ XXeS9nOURU5nXRR-nU$)Nz/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.)rr2r) rr`rerfr?r7r*rr@init_noise_sigma) rprnum_channels_latentsrrrr2rrPrs rDprepare_latents3StableDiffusionXLControlNetPipeline.prepare_latentsys  K400 0 J$// /   i & &3y>Z+GA#i.AQR&<'gi  ?"5fZGjj(GNN;;;rFc2[X-U-5nURRR[ U5-U-nURR R RnX:wa[SUSUS35e[R"U/US9nU$)Nz7Model expects an added time embedding vector of length z, but a vector of z was created. The model has an incorrect config. Please check `unet.config.time_embedding_type` and `text_encoder_2.config.projection_dim`.r) rfrYrjaddition_time_embed_dimr? add_embeddinglinear_1 in_featuresr7rtensor) rp original_sizecrops_coords_top_left target_sizertext_encoder_projection_dimrTpassed_add_embed_dimexpected_add_embed_dims rD_get_add_time_ids5StableDiffusionXLControlNetPipeline._get_add_time_idssMAKOP  II   4 4s<7H HKf f "&!8!8!A!A!M!M ! 9IJ`IaasuItJJUV ||\N%@ rFc*URRnURR[RS9 [ URR RRSR[[45nU(aURRRU5 URR RRU5 URR RRU5 gg)Nrr)rXrrrrredecoder mid_block attentions processorrrpost_quant_convconv_in)rpruse_torch_2_0_or_xformerss rD upcast_vae.StableDiffusionXLControlNetPipeline.upcast_vaes  %-- ($. HH   & & 1 1! 4 > > % % ! % HH $ $ ' ' . HH   $ $ ' ' . HH   & & ) )% 0 %rFiw embedding_dimrreturncr[UR5S:XdeUS-nUS-n[R"[R"S55US- - n[R "[R "XCS9U*-5nURU5SS2S4USSS24-n[R"[R"U5[R"U5/SS9nUS-S:Xa*[RRRUS5nURURS U4:XdeU$) a, See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298 Args: w (`torch.Tensor`): Generate embedding vectors with a specified guidance scale to subsequently enrich timestep embeddings. embedding_dim (`int`, *optional*, defaults to 512): Dimension of the embeddings to generate. dtype (`torch.dtype`, *optional*, defaults to `torch.float32`): Data type of the generated embeddings. Returns: `torch.Tensor`: Embedding vectors with shape `(len(w), embedding_dim)`. r^g@@r+g@rNr)rr^r)r?rrlogr&exparangerrsincosnn functionalpad)rpr9r:rhalf_dimembs rDget_guidance_scale_embedding@StableDiffusionXLControlNetPipeline.get_guidance_scale_embeddings"177|q   J A%ii W-.(Q,?ii X;sdBCdd5k!T'"Sq\1ii338a@ 1  !((%%))#v6CyyQWWQZ7777 rFcUR$r)_guidance_scalerps rDguidance_scale2StableDiffusionXLControlNetPipeline.guidance_scales###rFcUR$r) _clip_skiprKs rDrz-StableDiffusionXLControlNetPipeline.clip_skip rFcrURS:=(a" URRRSL$)Nr^)rJrYrjtime_cond_proj_dimrKs rDru?StableDiffusionXLControlNetPipeline.do_classifier_free_guidances.##a'WDII,<,<,O,OSW,WWrFcUR$r)_cross_attention_kwargsrKs rDcross_attention_kwargs:StableDiffusionXLControlNetPipeline.cross_attention_kwargss+++rFcUR$r)_denoising_endrKs rD denoising_end1StableDiffusionXLControlNetPipeline.denoising_end"""rFcUR$r)_num_timestepsrKs rD num_timesteps1StableDiffusionXLControlNetPipeline.num_timestepsr]rFcUR$r) _interruptrKs rD interrupt-StableDiffusionXLControlNetPipeline.interruptrQrF2g@pil)rrrPrWrrr1r3r4r[rLrrrr output_type return_dictrWrrrrr'r(r)negative_original_sizenegative_crops_coords_top_leftnegative_target_sizecallback_on_step_endrc'  ^]U'RSS5n(U'RSS5n)U(b [SSS5 U)b [SSS5 [U%[[45(a U%R n&[ UR5(aURRO URn*[U[5(d%[U[5(a[U5U/-nO[U[5(d%[U[5(a[U5U/-nOb[U[5(dM[U[5(d8[U*[5(a[U*R5OSn+U+U/-U+U/-nnURUUUU)U U UUUUUUUUUU&5 XlU$UlUUlXlSUlUb[U[&5(aSn,O3Ub![U[5(a [U5n,OUR(S n,UR*n-[U*[5(a.[U[,5(aU/[U*R5-n[U*[.5(aU*R0R2O"U*RS R0R2n.U=(d U.nUR4bUR4R7S S5OSn/UR9UUU-U UR:U U UUUUU/UR<S 9 unnnnUcUb"UR?UUU-U,U -UR:5n0[U*[.5(a@URAUUUU,U -U U-U*RBUR:US 9 nUR(S SupEOx[U*[5(aa/n1UHBn2URAU2UUU,U -U U-U*RBUR:US 9 n2U1REU25 MD U1nUS R(S SupEOe[GURHUU-Xx5upv[U5Ul%URLR0RNn3URQU,U -U3UUURBU-UU5nSn4URLR0RRb{[TRV"URXS- 5R[U,U -5n5UR]U5URLR0RRS9R_U-URBS9n4URaX5n6/n7[c[U55Hn8[eUU5V9V:s/sH>un9n:S[-U8[U5- U9:=(d U8S-[U5- U::5- PM@ n;n9n:U7RE[U*[.5(aU;S OU;5 M [U[5(aU=(d US R(S SnOU=(d UR(S SnU =(d XE4n UnU!b!U#bURmU!U"U#URBU=S9n?OU>n?UR:(aE[TRn"UU/S S9n[TRn"UU/S S9n>UR_U-5nUR_U-5R[U,U -S5n>[U5X`RHRp-- n@URrb[URr[,5(aURrS :aURrS:a[i[uURHR0RvURrURHR0Rv-- 55m][[[yU]4SjU555nUSUn[ URL5nA[ UR5nB[{SS5nCUR}US9nD[U5GHun8nEUR(aM[TRR5(a3WA(a,WB(a%WC(a[TRR5 UR:(a[TRn"U/S-5OUnFURHRUFWE5nFUS.nGU(amUR:(a\UnHURHRUHWE5nHURS5SnIURS5SS.nJOWFnHUnIWGnJ[U7U8[5(a([eUU7U85VKV9s/sH unKn9UKU9-PM nLnKn9O$UnM[UM[5(aWMS nMWMU7U8-nLURWHWEWIUWLUWJSS9unNnOU(ayUR:(ahWNVPs/sH/nP[TRn"[TR"UP5UP/5PM1 nNnP[TRn"[TR"WO5UO/5nOUcUbW0WGS'URMWFWEUU4UR4WNWOWGSS9 S nQUR:(aWQRS5unRnSURU USUR- --nQURHR"WQWEU40U6DSS0D6S nU%b0nTU&HnU[5UUWTUU'M U%"UU8WEWT5nVUVRSU5nUVRSU5nUVRS U5nUVRS!U<5n5n>UVRS$U?5n?UVRS%U5nU8[U5S- :Xd)U8S-W@:a`U8S-URHRp-S :Xa@WDR5 U(b-U8U)-S :Xa$U8[URHS&S5-nWU("UWWEU5 [(dGM[R"5 GM SSS5 US':XGdURRB[TR:H=(a URR0RnXUX(a_UR5 UR_[[URRR555RB5n[URR0S(5=(a" URR0RSLnY[URR0S)5=(a" URR0RSLnZWY(GaWZ(a[TRV"URR0R5RSS*SS5R_URURB5n[[TRV"URR0R5RSS*SS5R_URURB5n\UU\-URR0R- U[-nO#UURR0R- nURRUSS+9S nWX(a'URR_[TRS,9 OUnUS':XdBURbURRU5nURRUUS-9nUR5 U(dU4$[US.9$s sn:n9fs sn9nKfs snPf!,(df  GNV=f)/u0 The call function to the pipeline for generation. Args: prompt (`str` or `List[str]`, *optional*): The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`. prompt_2 (`str` or `List[str]`, *optional*): The prompt or prompts to be sent to `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is used in both text-encoders. image (`torch.Tensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.Tensor]`, `List[PIL.Image.Image]`, `List[np.ndarray]`,: `List[List[torch.Tensor]]`, `List[List[np.ndarray]]` or `List[List[PIL.Image.Image]]`): The ControlNet input condition to provide guidance to the `unet` for generation. If the type is specified as `torch.Tensor`, it is passed to ControlNet as is. `PIL.Image.Image` can also be accepted as an image. The dimensions of the output image defaults to `image`'s dimensions. If height and/or width are passed, `image` is resized accordingly. If multiple ControlNets are specified in `init`, images must be passed as a list such that each element of the list can be correctly batched for input to a single ControlNet. height (`int`, *optional*, defaults to `self.unet.config.sample_size * self.vae_scale_factor`): The height in pixels of the generated image. Anything below 512 pixels won't work well for [stabilityai/stable-diffusion-xl-base-1.0](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0) and checkpoints that are not specifically fine-tuned on low resolutions. width (`int`, *optional*, defaults to `self.unet.config.sample_size * self.vae_scale_factor`): The width in pixels of the generated image. Anything below 512 pixels won't work well for [stabilityai/stable-diffusion-xl-base-1.0](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0) and checkpoints that are not specifically fine-tuned on low resolutions. num_inference_steps (`int`, *optional*, defaults to 50): The number of denoising steps. More denoising steps usually lead to a higher quality image at the expense of slower inference. timesteps (`List[int]`, *optional*): Custom timesteps to use for the denoising process with schedulers which support a `timesteps` argument in their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is passed will be used. Must be in descending order. sigmas (`List[float]`, *optional*): Custom sigmas to use for the denoising process with schedulers which support a `sigmas` argument in their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is passed will be used. denoising_end (`float`, *optional*): When specified, determines the fraction (between 0.0 and 1.0) of the total denoising process to be completed before it is intentionally prematurely terminated. As a result, the returned sample will still retain a substantial amount of noise as determined by the discrete timesteps selected by the scheduler. The denoising_end parameter should ideally be utilized when this pipeline forms a part of a "Mixture of Denoisers" multi-pipeline setup, as elaborated in [**Refining the Image Output**](https://huggingface.co/docs/diffusers/api/pipelines/stable_diffusion/stable_diffusion_xl#refining-the-image-output) guidance_scale (`float`, *optional*, defaults to 5.0): A higher guidance scale value encourages the model to generate images closely linked to the text `prompt` at the expense of lower image quality. Guidance scale is enabled when `guidance_scale > 1`. negative_prompt (`str` or `List[str]`, *optional*): The prompt or prompts to guide what to not include in image generation. If not defined, you need to pass `negative_prompt_embeds` instead. Ignored when not using guidance (`guidance_scale < 1`). negative_prompt_2 (`str` or `List[str]`, *optional*): The prompt or prompts to guide what to not include in image generation. This is sent to `tokenizer_2` and `text_encoder_2`. If not defined, `negative_prompt` is used in both text-encoders. num_images_per_prompt (`int`, *optional*, defaults to 1): The number of images to generate per prompt. eta (`float`, *optional*, defaults to 0.0): Corresponds to parameter eta (η) from the [DDIM](https://huggingface.co/papers/2010.02502) paper. Only applies to the [`~schedulers.DDIMScheduler`], and is ignored in other schedulers. generator (`torch.Generator` or `List[torch.Generator]`, *optional*): A [`torch.Generator`](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 is generated by sampling using the supplied random `generator`. prompt_embeds (`torch.Tensor`, *optional*): Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not provided, text embeddings are generated from the `prompt` input argument. negative_prompt_embeds (`torch.Tensor`, *optional*): Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument. pooled_prompt_embeds (`torch.Tensor`, *optional*): Pre-generated pooled text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not provided, pooled text embeddings are generated from `prompt` input argument. negative_pooled_prompt_embeds (`torch.Tensor`, *optional*): Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not provided, pooled `negative_prompt_embeds` are generated from `negative_prompt` input argument. ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters. ip_adapter_image_embeds (`List[torch.Tensor]`, *optional*): Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters. Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. It should contain the negative image embedding if `do_classifier_free_guidance` is set to `True`. If not provided, embeddings are computed from the `ip_adapter_image` input argument. 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 [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] instead of a plain tuple. cross_attention_kwargs (`dict`, *optional*): A kwargs dictionary that if specified is passed along to the [`AttentionProcessor`] as defined in [`self.processor`](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py). controlnet_conditioning_scale (`float` or `List[float]`, *optional*, defaults to 1.0): The outputs of the ControlNet are multiplied by `controlnet_conditioning_scale` before they are added to the residual in the original `unet`. If multiple ControlNets are specified in `init`, you can set the corresponding scale as a list. guess_mode (`bool`, *optional*, defaults to `False`): The ControlNet encoder tries to recognize the content of the input image even if you remove all prompts. A `guidance_scale` value between 3.0 and 5.0 is recommended. control_guidance_start (`float` or `List[float]`, *optional*, defaults to 0.0): The percentage of total steps at which the ControlNet starts applying. control_guidance_end (`float` or `List[float]`, *optional*, defaults to 1.0): The percentage of total steps at which the ControlNet stops applying. original_size (`Tuple[int]`, *optional*, defaults to (1024, 1024)): If `original_size` is not the same as `target_size` the image will appear to be down- or upsampled. `original_size` defaults to `(height, width)` if not specified. Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952](https://huggingface.co/papers/2307.01952). crops_coords_top_left (`Tuple[int]`, *optional*, defaults to (0, 0)): `crops_coords_top_left` can be used to generate an image that appears to be "cropped" from the position `crops_coords_top_left` downwards. Favorable, well-centered images are usually achieved by setting `crops_coords_top_left` to (0, 0). Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952](https://huggingface.co/papers/2307.01952). target_size (`Tuple[int]`, *optional*, defaults to (1024, 1024)): For most cases, `target_size` should be set to the desired height and width of the generated image. If not specified it will default to `(height, width)`. Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952](https://huggingface.co/papers/2307.01952). negative_original_size (`Tuple[int]`, *optional*, defaults to (1024, 1024)): To negatively condition the generation process based on a specific image resolution. Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952](https://huggingface.co/papers/2307.01952). For more information, refer to this issue thread: https://github.com/huggingface/diffusers/issues/4208. negative_crops_coords_top_left (`Tuple[int]`, *optional*, defaults to (0, 0)): To negatively condition the generation process based on a specific crop coordinates. Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952](https://huggingface.co/papers/2307.01952). For more information, refer to this issue thread: https://github.com/huggingface/diffusers/issues/4208. negative_target_size (`Tuple[int]`, *optional*, defaults to (1024, 1024)): To negatively condition the generation process based on a target image resolution. It should be as same as the `target_size` for most cases. Part of SDXL's micro-conditioning as explained in section 2.2 of [https://huggingface.co/papers/2307.01952](https://huggingface.co/papers/2307.01952). For more information, refer to this issue thread: https://github.com/huggingface/diffusers/issues/4208. clip_skip (`int`, *optional*): Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that the output of the pre-final layer will be used for computing the prompt embeddings. callback_on_step_end (`Callable`, `PipelineCallback`, `MultiPipelineCallbacks`, *optional*): A function or a subclass of `PipelineCallback` or `MultiPipelineCallbacks` that is called at the end of each denoising step during the inference. 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.stable_diffusion.StableDiffusionPipelineOutput`] or `tuple`: If `return_dict` is `True`, [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] is returned, otherwise a `tuple` is returned containing the output images. callbackNrz1.0.0zjPassing `callback` as an input argument to `__call__` is deprecated, consider using `callback_on_step_end`zpPassing `callback_steps` as an input argument to `__call__` is deprecated, consider using `callback_on_step_end`r^Frscale)rQrRrxrUryrz) rWrrrrtr2rrurr)r:rrr)rr*rc>UT:$rr6)tsdiscrete_timestep_cutoffs rD>StableDiffusionXLControlNetPipeline.__call__..s RC[=[rFz>=z2.1)totalr+) text_embedstime_ids)encoder_hidden_statescontrolnet_condconditioning_scaleradded_cond_kwargsrir)ry timestep_condrWdown_block_additional_residualsmid_block_additional_residualr|ririrPrQrRrSrUrTrVrWorderlatent latents_mean latents_stdr)rir)rh)images)dpopr#rerr tensor_inputsr(rZrrfr?rrrrJrOrVrZrcrrrrrrjglobal_pool_conditionsrWgetrrurzrrrrrEr@r_rY in_channelsrrSrr&rLrrGrrrangerrMrprojection_dimr-rrr[roundnum_train_timestepsfilterr) progress_barrrdcuda is_available _inductorcudagraph_mark_step_beginscale_model_inputrrrlocalsupdateri XLA_AVAILABLExm mark_steprXfloat16 force_upcastr7riterr4r<rrrrr2scaling_factordecodernapply_watermarkrl postprocessmaybe_free_model_hooksr.)^rprrrsrWrrr1r3r4r[rLrvrwrtrrrPrQrRrxrUrrrhrirWrrrrr'r(r)rjrkrlrzrmrrArorrZmultrr2rtext_encoder_lora_scalerrrrr}guidance_scale_tensorrcontrolnet_keeprsekeepsrSr*rTrVnum_warmup_stepsis_unet_compiledis_controlnet_compiledis_torch_higher_equal_2_1rtlatent_model_inputr|control_model_inputcontrolnet_prompt_embedscontrolnet_added_cond_kwargsc cond_scalecontrolnet_cond_scaledown_block_res_samplesmid_block_res_sampled noise_prednoise_pred_uncondnoise_pred_textcallback_kwargsrcallback_outputsstep_idxneeds_upcastinghas_latents_meanhas_latents_stdrrrss^ @rD__call__,StableDiffusionXLControlNetPipeline.__call__s*L::j$/$4d;   |   %  C  *-=?U,V W W1E1S1S .2DT__2U2UT__..[_[j[j 0$77JG[]aBXAY%Y "0$77JG]_cBVAW#W 2D99*MacgBhBh+5jBV+W+W3z']^D.//,--%9 "         "  # ) ) " .! & .#'=$+  *VS"9"9J  Jvt$<$<VJ&,,Q/J'' j"6 7 7JGdfkY>Y>eD ' ' + +GT :ko        !  , ,  '#9!5*G.nn    " )$  '+B+N?? '2200 L j/ 2 2&&%(==&; &&,0,L,L%' E"KK,MFE  $8 9 9F++ !),AA*?!$**040P0P),  f% E!!HNN23/MFE 5*< NN/* & ")n $yy//;;&& . .           99   . . :$)LL1D1Dq1H$I$P$PQ[^sQs$t ! ==%TYY5E5E5X5X>bgmmb4  !::9Js9~&A 68LMMDAqeAI.2Rq1uI6NQR6RSSM   " "z*o/V/V58\a b ' eT " ")@U1X^^BC-@M)=U[[-=M!4f_ .    &*-.B.H.H.L*M '*.*=*=*D*D*S*S '--  ! %%(C .  " -2F2R$($:$:&.$#)),G %;% !%1 !  + +!II'=}&MSTUM#ii)F(X^_`O 99&;\%JPQRL%((0 ),,V4#v.55jCX6XZ[\ y>,?..BVBV,VV    *4--u55""Q&""Q&'*NN))==))DNN,A,A,U,UUW( $ #&d62[]f+g&h"i !"6#67I-dii8!3DOO!D$4T5$A!   %8  9\!),1>> JJ++--).D1OO==?AEAaAaUYYy1}%=gn"%)^^%E%EFXZ[%\"4CQ]$^!$"B"B*1'*...*J*JK^`a*b'/>0Ea0H-!69K!KJ?C'*B$)'1)&B %@O @<&(<$"B"B\r-r[qVWeii9I9I!9La8P.Q[q*-r+099e6F6FG[6\^r5s+t(#/3J3V8D%n5"YY&*7"/+/+F+F4J2F&7 %'   339C9I9I!9L6%!2^YjGj5k!kJ..--j!WmHYmglmnop'3&(O?-3Xa[*@';D!Q'X$.229gFG$4$8$8-$XM-=-A-ABZ\r-s*&6&:&:;Lo&^O4D4H4H79V51$4#7#7 #UL,<,@,@AXZo,p),00%@EI**A9I/IqSTuX\XfXfXlXlNlpqNq '')+N0Ba0G#$(K#K 1g6 =LLNY-:^h&"hhnn =^$((//B^B^O!!**T$txx/G/G/R/R/T*U%V%\%\] 'txxGtDHHOOLhLhptLt %dhhoo}Eq$((//JeJemqJqOOLL!=!=>CCAq!QORRSZSaSacjcpcpqLL!s@rDrHrHs)XU *.2*.047;,[ ,[$,[4 ,[ ! ,[ # ,[#,[/4+@%BXZnno,[-,['+,["$,[.,[5,[,[d#')-%&,0)-+/049=7;@D&*#'ljlj3-lj& lj # lj &* lj"#lj$C=lj  -lj!) 6lj'u||4lj(0 '=ljUOljC=lj^52+'\!.#! $&*&)" +/#{|#^%*>0ei&1&58emm.1@E  <$$ XX,,#### ]]_12)-48$( $##%#")- #;?=A/0MQ*.049=7;@D9=@D%* ;?CF #' 9B S\ =c49n%\ =5d3i01\ =" \ =  \ = } \ =!\ =9\ =U \ = \ =\ ="%T#Y"78\ =$E#tCy.$9:\ = (}\ =\ = E%//43H"HIJ!\ ="%,,'#\ =$  -%\ =&!) 6'\ =('u||4)\ =*(0 '=+\ =,##56-\ =."*$u||*S#X?\ =@ %S#XA\ =B38_C\ =D!)sCx 9E\ =F).c3hG\ =H'uS#X7I\ =JC=K\ =L' (Cd+T124DF\\ ] M\ =R-1IS\ =3\ =rFrH)NNNN)Tr9typingrrrrrrr numpyr PIL.Imagerrtorch.nn.functionalrBrCr transformersr r r r rdiffusers.utils.import_utilsr callbacksrrrlrrloadersrrrrmodelsrrrrrmodels.attention_processorrr models.lorar schedulersr!utilsr"r#r$r%r&r'utils.torch_utilsr(r)r*pipeline_utilsr,r-#stable_diffusion_xl.pipeline_outputr.stable_diffusion_xl.watermarkr/r0torch_xla.core.xla_modelcore xla_modelrr get_loggerrrrrrr2rrErHr6rFrDrs9 DDD JAD rq:3TSDQ$%%L,))MM   H %+b*.15%)$( 8*!#8* U3 ,- .8*S " 8* T%[ ! 8*vX=$ X=rF