+hLSSKJrJrJrJr SSKrSSKJr SSKJ r SSK J r J r SSK Jr SS KJrJrJr \ "5(a SSKJs Jr S rOS r\ R0"\5r"S S \\5rg))ListOptionalTupleUnionN) UNet1DModel)SchedulerMixin)is_torch_xla_availablelogging) randn_tensor)AudioPipelineOutputDeprecatedPipelineMixinDiffusionPipelineTFc^\rSrSrSrSrSrS\S\4U4Sjjr \ R"5SS\ S\ S \ \\ R\\ R4S \ \S \S \\\44 S jj5rSrU=r$)DanceDiffusionPipeline%a Pipeline for audio generation. This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods implemented for all pipelines (downloading, saving, running on a particular device, etc.). Parameters: unet ([`UNet1DModel`]): A `UNet1DModel` to denoise the encoded audio. scheduler ([`SchedulerMixin`]): A scheduler to be used in combination with `unet` to denoise the encoded audio latents. Can be one of [`IPNDMScheduler`]. z0.33.1unet schedulerc@>[TU]5 URXS9 g)N)rr)super__init__register_modules)selfrr __class__s v/home/james-whalen/.local/lib/python3.13/site-packages/diffusers/pipelines/dance_diffusion/pipeline_dance_diffusion.pyrDanceDiffusionPipeline.__init__7s  4= batch_sizenum_inference_steps generatoraudio_length_in_s return_dictreturncUcAURRRURRR- nX@RRR-nS[ URR 5-nUSU-:a5[ USSU-URRR- S35e[U5nXg-S:waX@RRR-U-S-U-n[RUSX`RRR- S XRRR- S 35 [U5n[URR55Rn XRRRU4n [U[5(a*[ U5U:wa[ S [ U5S US 35e[!XUR"U S9n UR$R'X+R(S9 UR$R*R-U 5UR$lUR/UR$R*5Hfn URX5R0n UR$R3XU 5R4n [6(dMQ[8R:"5 Mh U R=SS5R?5RA5RC5n U SS2SS2SU24n U(dU 4$[EU S9$)at The call function to the pipeline for generation. Args: batch_size (`int`, *optional*, defaults to 1): The number of audio samples to generate. num_inference_steps (`int`, *optional*, defaults to 50): The number of denoising steps. More denoising steps usually lead to a higher-quality audio sample at the expense of slower inference. generator (`torch.Generator`, *optional*): A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make generation deterministic. audio_length_in_s (`float`, *optional*, defaults to `self.unet.config.sample_size/self.unet.config.sample_rate`): The length of the generated audio sample in seconds. return_dict (`bool`, *optional*, defaults to `True`): Whether or not to return a [`~pipelines.AudioPipelineOutput`] instead of a plain tuple. Example: ```py from diffusers import DiffusionPipeline from scipy.io.wavfile import write model_id = "harmonai/maestro-150k" pipe = DiffusionPipeline.from_pretrained(model_id) pipe = pipe.to("cuda") audios = pipe(audio_length_in_s=4.0).audios # To save locally for i, audio in enumerate(audios): write(f"maestro_test_{i}.wav", pipe.unet.sample_rate, audio.transpose()) # To display in google colab import IPython.display as ipd for audio in audios: display(ipd.Audio(audio, rate=pipe.unet.sample_rate)) ``` Returns: [`~pipelines.AudioPipelineOutput`] or `tuple`: If `return_dict` is `True`, [`~pipelines.AudioPipelineOutput`] is returned, otherwise a `tuple` is returned where the first element is a list with the generated audio. Nr rz1 is too small. Make sure it's bigger or equal to .rz is increased to z; so that it can be handled by the model. It will be cut to z after the denoising process.z/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.)r!devicedtype)r()audios)#rconfig sample_size sample_ratelen up_blocks ValueErrorintloggerinfonext parametersr) in_channels isinstancelistr _execution_devicer set_timestepsr( timestepsto progress_barsamplestep prev_sample XLA_AVAILABLExm mark_stepclampfloatcpunumpyr)rrr r!r"r#r-down_scale_factororiginal_sample_sizer)shapeaudiot model_outputs r__call__DanceDiffusionPipeline.__call__;sn  $ $ 0 0 < Z+GA#i.AQR&<'gi  U@V@V^cd $$%8$N#'>>#;#;#>#>u#E ""4>>#;#;[>^> ]]_#&MQ-1 m1m1!m1E%//43H"HIJ m1 $E? m1  m1 "E) *m1m1rr)typingrrrrrZmodelsr schedulersr utilsr r utils.torch_utilsr pipeline_utilsrrrtorch_xla.core.xla_modelcore xla_modelrCrB get_loggerrSr3rrQrrrjs_ 0/ !(4-\\))MM   H %D146GD1r