+hW SSKrSSKJrJrJrJr SSKrSSKJr SSK J r J r SSK Jr SSKJrJrJr SSKJr SS KJrJr SS KJr S S KJrJrJr S SKJr \"5(a SSK J!s J"r# Sr$OSr$\RJ"\&5r'"SS\\5r(g)N)ListOptionalTupleUnion) functional)CLIPTextModelWithProjection CLIPTokenizer)CLIPTextModelOutput)PriorTransformerUNet2DConditionModel UNet2DModel)UnCLIPScheduler)is_torch_xla_availablelogging) randn_tensor)DeprecatedPipelineMixinDiffusionPipelineImagePipelineOutput)UnCLIPTextProjModelTFc x^\rSrSr%SrSrS/r\\S'\ \S'\ \S'\ \S'\ \S'\ \S '\ \S '\\S '\\S '\\S 'SrS\S\ S\ S\ S\ S \ S \ S \S \S \4U4SjjrSrS#S\\\\4S\\R.4Sjjr\R2"5S$S\\\\\4S\S\S\S\S\\\R:\\R:4S\\R.S\\R.S\\R.S\\\\4S\\R.S\S\S\\S \4S!jj5r S"r!U=r"$)%UnCLIPPipeline)a Pipeline for text-to-image generation using unCLIP. 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.). Args: text_encoder ([`~transformers.CLIPTextModelWithProjection`]): Frozen text-encoder. tokenizer ([`~transformers.CLIPTokenizer`]): A `CLIPTokenizer` to tokenize text. prior ([`PriorTransformer`]): The canonical unCLIP prior to approximate the image embedding from the text embedding. text_proj ([`UnCLIPTextProjModel`]): Utility class to prepare and combine the embeddings before they are passed to the decoder. decoder ([`UNet2DConditionModel`]): The decoder to invert the image embedding into an image. super_res_first ([`UNet2DModel`]): Super resolution UNet. Used in all but the last step of the super resolution diffusion process. super_res_last ([`UNet2DModel`]): Super resolution UNet. Used in the last step of the super resolution diffusion process. prior_scheduler ([`UnCLIPScheduler`]): Scheduler used in the prior denoising process (a modified [`DDPMScheduler`]). decoder_scheduler ([`UnCLIPScheduler`]): Scheduler used in the decoder denoising process (a modified [`DDPMScheduler`]). super_res_scheduler ([`UnCLIPScheduler`]): Scheduler used in the super resolution denoising process (a modified [`DDPMScheduler`]). z0.33.1priordecoder text_proj text_encoder tokenizersuper_res_firstsuper_res_lastprior_schedulerdecoder_schedulersuper_res_schedulerzAtext_encoder->text_proj->decoder->super_res_first->super_res_lastc R>[T U]5 URUUUUUUUUU U S9 g)N) rrrr rr!r"r#r$r%)super__init__register_modules) selfrrrr rr!r"r#r$r% __class__s d/home/james-whalen/.local/lib/python3.13/site-packages/diffusers/pipelines/unclip/pipeline_unclip.pyr(UnCLIPPipeline.__init__YsA  %+)+/ 3  cUc [XX2S9nO truncationreturn_tensorslongest)r@rBz\The following part of your input was truncated because CLIP can only handle sequences up to z tokens: rdim) isinstancelistlenr model_max_length input_idsattention_maskboolr5r3torchequal batch_decodeloggerwarningr text_embedslast_hidden_staterepeat_interleaverepeatviewcat)r*promptr1num_images_per_promptdo_classifier_free_guidancer;r< batch_size text_inputstext_input_ids text_maskuntruncated_ids removed_texttext_encoder_output prompt_embedstext_enc_hid_states uncond_tokens uncond_inputuncond_text_mask*negative_prompt_embeds_text_encoder_outputnegative_prompt_embedsuncond_text_enc_hid_statesseq_lens r,_encode_promptUnCLIPPipeline._encode_promptsr  $(264(@(@VaJ..$>>::# )K)22N#22779<>$>>::# *L ,::??ADDVL 9=9J9J5n!O,[8R:"UUR4R<S4S S 9n!UR@RUUS9 UR@Rn"URBRRDn#URBRRFn$URBRRFn%URUU#U$U%4UR UUUUR@5n[#UR%U"55GH unnU(a[&R("U/S-5OUnURCUUUU U!S9RHn&U(avU&R-S5un'n(U'RKUR SSS9un'n)U(RKUR SSS9un(n*U'U U(U'- --n&[&R("U&U*/SS9n&US-U"R S:XaSnOU"US-nUR@R/U&UUUUS9R0nGM URMSS5nUn+URNRUUS9 URNRn,URPRRDS-n-URPRRFn$URPRRFn%URUU-U$U%4U+R UUU URN5n UR S :Xa[8RR"U+U$U%/S9n.OV0n/S[TRV"[8RR5RX;aS U/S'[8RR"U+4U$U%/SSS.U/D6n.[#UR%U,55HunnUU,R SS- :Xa URZn0O URPn0[&R("U U./SS9nU0"UUS9RHn&US-U,R S:XaSnOU,US-nURNR/U&UU UUS9R0n [\(dM[^R`"5 M U n1URc5 U1S-S-n1U1RMSS5n1U1Re5RgSSSS5Ri5Rk5n1US:XaURmU15n1U(dU14$[oU1S9$)a The call function to the pipeline for generation. Args: prompt (`str` or `List[str]`): The prompt or prompts to guide image generation. This can only be left undefined if `text_model_output` and `text_attention_mask` is passed. num_images_per_prompt (`int`, *optional*, defaults to 1): The number of images to generate per prompt. prior_num_inference_steps (`int`, *optional*, defaults to 25): The number of denoising steps for the prior. More denoising steps usually lead to a higher quality image at the expense of slower inference. decoder_num_inference_steps (`int`, *optional*, defaults to 25): The number of denoising steps for the decoder. More denoising steps usually lead to a higher quality image at the expense of slower inference. super_res_num_inference_steps (`int`, *optional*, defaults to 7): The number of denoising steps for super resolution. More denoising steps usually lead to a higher quality image at the expense of slower inference. generator (`torch.Generator` or `List[torch.Generator]`, *optional*): A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make generation deterministic. prior_latents (`torch.Tensor` of shape (batch size, embeddings dimension), *optional*): Pre-generated noisy latents to be used as inputs for the prior. decoder_latents (`torch.Tensor` of shape (batch size, channels, height, width), *optional*): Pre-generated noisy latents to be used as inputs for the decoder. super_res_latents (`torch.Tensor` of shape (batch size, channels, super res height, super res width), *optional*): Pre-generated noisy latents to be used as inputs for the decoder. prior_guidance_scale (`float`, *optional*, defaults to 4.0): 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`. decoder_guidance_scale (`float`, *optional*, defaults to 4.0): 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`. text_model_output (`CLIPTextModelOutput`, *optional*): Pre-defined [`CLIPTextModel`] outputs that can be derived from the text encoder. Pre-defined text outputs can be passed for tasks like text embedding interpolations. Make sure to also pass `text_attention_mask` in this case. `prompt` can the be left `None`. text_attention_mask (`torch.Tensor`, *optional*): Pre-defined CLIP text attention mask that can be derived from the tokenizer. Pre-defined text attention masks are necessary when passing `text_model_output`. 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.ImagePipelineOutput`] instead of a plain tuple. 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. Nrz2`prompt` has to be of type `str` or `list` but is rg?)r1r)timestepproj_embeddingencoder_hidden_statesrM)rzsampler0 prev_timestep)image_embeddingsrdtext_encoder_hidden_statesr\mps)valueT)r}rzr| class_labelsrMrE)r~r0rD)size antialiasbicubicF)rmode align_corners)r}rzg?r pil)images)8rHstrrIrJr4typer3_execution_devicermr# set_timesteps timestepsrconfig embedding_dimr9r2 enumerate progress_barrOrYpredicted_image_embeddingchunkstep prev_samplepost_process_latentsrintFpadclip_extra_context_tokensrNr$r in_channels sample_sizer}splitclampr%r! interpolateinspect signature parametersr" XLA_AVAILABLExm mark_stepmaybe_free_model_hookscpupermutefloatnumpy numpy_to_pilr)2r*rZr[rorprqr0rrrsrtr;r<rurvrwrxr]r1r\rdrer`prior_timesteps_tensorritlatent_model_inputr predicted_image_embedding_uncondpredicted_image_embedding_textr~radditive_clip_time_embeddingsdecoder_text_maskdecoder_timesteps_tensornum_channels_latentsheightwidth noise_prednoise_pred_uncondnoise_pred_text_predicted_variance image_smallsuper_res_timesteps_tensorchannelsimage_upscaledinterpolate_antialiasunetimages2 r,__call__UnCLIPPipeline.__call__sJ  &#&& FD)) [  #UVZ[aVbUc!dee*1-33A6J''"77 &:S&@&`DZ]`D`#8<8K8K F13NPa9 5 *I **+DV*T!%!5!5!?!? ))77 ,,  '         d//0FGHDAqC^M?Q+>!?dq (, ",&9( )3) (' &+SlSrSrstSuP02P,LOc25UUP-)1u.44Q77 $ 6q1u =  0055)$#+ 6 k /I> 77 F ( >B^^-'':(C >L> :: ;;% "uyy1I !i$..2Z2Z\]1^fg h  1 6 6uzz B  !i$..2Z2Z\]1^fj k  ,,-HQW,X#'#9#9#C#C #||22>>$$00 ##//.. -vu =  % %     " "  d//0HIJDAqE`O+'>?Q?W?WXY?Z`a'>'b$!16E6K6KL^LdLdefLgmn6K6o3!3.1G?]nKn1oo "YY 4F'GQO 1u066q99 $ 8Q ? #4499AmW`:k 3K:*//A6%   ../LU[.\%)%=%=%G%G"''..::a?%%,,88$$++77 00 65 1        $ $   ;;% ]];fe_MN$& !g// >III59%k2]]#)5/ QVZoNd//0JKLDAq.44Q7!;;**++!&,=~+NTU!V )f  1u288;; $ :1q5 A !% 8 8 = =A0 Yb!>!k } 5M8" ##% c! Aq! ##Aq!Q/557==? % %%e,E8O"%00r.)NN)NrrNNNNNNg@g @rT)#__name__ __module__ __qualname____firstlineno____doc___last_supported_version_exclude_from_cpu_offloadr __annotations__r rrr rrmodel_cpu_offload_seqr(r9rrr rrOTensorrmno_gradrrr GeneratorrrNr__static_attributes__ __classcell__)r+s@r,rr)sZ<'!(   !!""--  $$&&((_  & 2  !  '  % $ ) + - 8 "JN6:Y= $E*=u*D$EF Y=&ell3Y=v ]]_37%&)++--.MQ042648IM6:&)(+%* !\1sDI~./\1 #\1$' \1 &) \1 (+ \1E%//43H"HIJ\1  -\1"%,,/\1$ELL1\1$E*=u*D$EF\1&ell3\1$\1!&\1c]\1 !\1\1r.r))rtypingrrrrrOtorch.nnrr transformersrr &transformers.models.clip.modeling_clipr modelsr r r schedulersrutilsrrutils.torch_utilsrpipeline_utilsrrrrrtorch_xla.core.xla_modelcore xla_modelrr get_loggerrrRrrr.r,rsq// $CFII)4-\\*))MM   H %O1,.?O1r.