o piMC@sddlZddlmZddlmZmZmZddlZddlm Z m Z ddl m Z m Z ddlmZdd lmZe eZeGd d d e ZGd d d ee ZdS)N) dataclass)OptionalTupleUnion) ConfigMixinregister_to_config) BaseOutputlogging) randn_tensor)SchedulerMixinc@s.eZdZUdZejed<dZeejed<dS)EDMEulerSchedulerOutputaq Output class for the scheduler's `step` function output. Args: prev_sample (`torch.Tensor` of shape `(batch_size, num_channels, height, width)` for images): Computed sample `(x_{t-1})` of previous timestep. `prev_sample` should be used as next model input in the denoising loop. pred_original_sample (`torch.Tensor` of shape `(batch_size, num_channels, height, width)` for images): The predicted denoised sample `(x_{0})` based on the model output from the current timestep. `pred_original_sample` can be used to preview progress or for guidance. prev_sampleNpred_original_sample) __name__ __module__ __qualname____doc__torchTensor__annotations__rrrrg/home/ubuntu/SoloSpeech/.venv/lib/python3.10/site-packages/diffusers/schedulers/scheduling_edm_euler.pyrs  rc@seZdZdZgZdZe       dId ed ed ed ede dedefddZ e ddZ e ddZ e ddZdJde fddZddZdd Zd!d"Zd#ejd$eeejfd%ejfd&d'ZdKd)e d*eeejffd+d,ZdLd%ejfd-d.ZdLd%ejfd/d0ZdKd1d2Zd3d4Zd5d5ed6d7d(d8fd9ejd$eeejfd#ejd:ed;edeejd?ed%ee e!ffd@dAZ"dBejdCejdDejd%ejfdEdFZ#dGdHZ$d(S)MEDMEulerSchedulera Implements the Euler scheduler in EDM formulation as presented in Karras et al. 2022 [1]. [1] Karras, Tero, et al. "Elucidating the Design Space of Diffusion-Based Generative Models." https://arxiv.org/abs/2206.00364 This model inherits from [`SchedulerMixin`] and [`ConfigMixin`]. Check the superclass documentation for the generic methods the library implements for all schedulers such as loading and saving. Args: sigma_min (`float`, *optional*, defaults to 0.002): Minimum noise magnitude in the sigma schedule. This was set to 0.002 in the EDM paper [1]; a reasonable range is [0, 10]. sigma_max (`float`, *optional*, defaults to 80.0): Maximum noise magnitude in the sigma schedule. This was set to 80.0 in the EDM paper [1]; a reasonable range is [0.2, 80.0]. sigma_data (`float`, *optional*, defaults to 0.5): The standard deviation of the data distribution. This is set to 0.5 in the EDM paper [1]. sigma_schedule (`str`, *optional*, defaults to `karras`): Sigma schedule to compute the `sigmas`. By default, we the schedule introduced in the EDM paper (https://arxiv.org/abs/2206.00364). Other acceptable value is "exponential". The exponential schedule was incorporated in this model: https://huggingface.co/stabilityai/cosxl. num_train_timesteps (`int`, defaults to 1000): The number of diffusion steps to train the model. prediction_type (`str`, defaults to `epsilon`, *optional*): Prediction type of the scheduler function; can be `epsilon` (predicts the noise of the diffusion process), `sample` (directly predicts the noisy sample`) or `v_prediction` (see section 2.4 of [Imagen Video](https://imagen.research.google/video/paper.pdf) paper). rho (`float`, *optional*, defaults to 7.0): The rho parameter used for calculating the Karras sigma schedule, which is set to 7.0 in the EDM paper [1]. r Mb`?T@?karrasepsilon@ sigma_min sigma_max sigma_datasigma_schedulenum_train_timestepsprediction_typerhoc Cs|dvr td|dd|_tdd|}|dkr ||} n |dkr)||} || |_t| tj d| j dg|_ d |_ d|_ d|_|j d |_ dS) N)r exponentialz-Wrong value for provided for `sigma_schedule=z`.`rr rr)deviceFcpu) ValueErrornum_inference_stepsrlinspace_compute_karras_sigmas_compute_exponential_sigmasprecondition_noise timestepscatzerosr+sigmasis_scale_input_called _step_index _begin_indexto) selfr"r#r$r%r&r'r(rampr6rrr__init__Us    zEDMEulerScheduler.__init__cCs|jjdddS)Nrr r)configr#r;rrrinit_noise_sigmavsz"EDMEulerScheduler.init_noise_sigmacC|jS)zg The index counter for current timestep. It will increase 1 after each scheduler step. )r8r?rrr step_index{zEDMEulerScheduler.step_indexcCrA)zq The index for the first timestep. It should be set from pipeline with `set_begin_index` method. r9r?rrr begin_indexrCzEDMEulerScheduler.begin_indexrrEcCs ||_dS)z Sets the begin index for the scheduler. This function should be run from pipeline before the inference. Args: begin_index (`int`): The begin index for the scheduler. NrD)r;rErrrset_begin_indexs z!EDMEulerScheduler.set_begin_indexcCs(d|d|jjdd}||}|S)Nr rr)r>r$)r;samplesigmac_in scaled_samplerrrprecondition_inputssz%EDMEulerScheduler.precondition_inputscCs*t|tjs t|g}dt|}|S)Ng?) isinstancerrtensorlog)r;rHc_noiserrrr2s  z$EDMEulerScheduler.precondition_noisecCs|jj}|d|d|d}|jjdkr%|||d|dd}n |jjdkr;| ||d|dd}n td|jjd||||}|S)Nrr r v_predictionzPrediction type z is not supported.)r>r$r'r-)r;rG model_outputrHr$c_skipc_outdenoisedrrrprecondition_outputss   z&EDMEulerScheduler.precondition_outputsrGtimestepreturncCs6|jdur |||j|j}|||}d|_|S)a Ensures interchangeability with schedulers that need to scale the denoising model input depending on the current timestep. Scales the denoising model input by `(sigma**2 + 1) ** 0.5` to match the Euler algorithm. Args: sample (`torch.Tensor`): The input sample. timestep (`int`, *optional*): The current timestep in the diffusion chain. Returns: `torch.Tensor`: A scaled input sample. NT)rB_init_step_indexr6rKr7)r;rGrVrHrrrscale_model_inputs    z#EDMEulerScheduler.scale_model_inputNr.r+cCs||_tdd|j}|jjdkr||}n |jjdkr"||}|jtj|d}| ||_ t |tj d|j dg|_d|_d|_|jd|_dS) a Sets the discrete timesteps used for the diffusion chain (to be run before inference). Args: num_inference_steps (`int`): The number of diffusion steps used when generating samples with a pre-trained model. device (`str` or `torch.device`, *optional*): The device to which the timesteps should be moved to. If `None`, the timesteps are not moved. rr rr))dtyper+r*Nr,)r.rr/r>r%r0r1r:float32r2r3r4r5r+r6r8r9)r;r.r+r<r6rrr set_timestepss      zEDMEulerScheduler.set_timestepscCsP|p|jj}|p |jj}|jj}|d|}|d|}|||||}|S)z6Constructs the noise schedule of Karras et al. (2022).r )r>r"r#r()r;r<r"r#r( min_inv_rho max_inv_rhor6rrrr0s    z(EDMEulerScheduler._compute_karras_sigmascCsD|p|jj}|p |jj}tt|t|t| d}|S)zImplementation closely follows k-diffusion. https://github.com/crowsonkb/k-diffusion/blob/6ab5146d4a5ef63901326489f31f1d8e7dd36b48/k_diffusion/sampling.py#L26 r) r>r"r#rr/mathrNlenexpflip)r;r<r"r#r6rrrr1s  (z-EDMEulerScheduler._compute_exponential_sigmascCs:|dur|j}||k}t|dkrdnd}||S)Nr r)r3nonzeror`item)r;rVschedule_timestepsindicesposrrrindex_for_timesteps   z$EDMEulerScheduler.index_for_timestepcCs@|jdurt|tjr||jj}|||_dS|j |_dSN) rErLrrr:r3r+rhr8r9)r;rVrrrrXs   z"EDMEulerScheduler._init_step_indexinfg?TrQs_churns_tmins_tmaxs_noise generator return_dictc CsDt|ttjtjfrtd|jstd|j dur | || tj }|j |j } || kr6|krDnn t|t|j ddnd} t|j|j|j|d} | |} | | d}| dkrm|| |d | d d }||||}|||}|j |j d|}|||}| |j}|jd7_| s|fSt||d S) a* Predict the sample from the previous timestep by reversing the SDE. This function propagates the diffusion process from the learned model outputs (most often the predicted noise). Args: model_output (`torch.Tensor`): The direct output from learned diffusion model. timestep (`float`): The current discrete timestep in the diffusion chain. sample (`torch.Tensor`): A current instance of a sample created by the diffusion process. s_churn (`float`): s_tmin (`float`): s_tmax (`float`): s_noise (`float`, defaults to 1.0): Scaling factor for noise added to the sample. generator (`torch.Generator`, *optional*): A random number generator. return_dict (`bool`): Whether or not to return a [`~schedulers.scheduling_euler_discrete.EDMEulerSchedulerOutput`] or tuple. Returns: [`~schedulers.scheduling_euler_discrete.EDMEulerSchedulerOutput`] or `tuple`: If return_dict is `True`, [`~schedulers.scheduling_euler_discrete.EDMEulerSchedulerOutput`] is returned, otherwise a tuple is returned where the first element is the sample tensor. zPassing integer indices (e.g. from `enumerate(timesteps)`) as timesteps to `EDMEulerScheduler.step()` is not supported. Make sure to pass one of the `scheduler.timesteps` as a timestep.zThe `scale_model_input` function should be called before `step` to ensure correct denoising. See `StableDiffusionPipeline` for a usage example.Nr g4y?rj)rZr+rprrr)rr)rLintr IntTensor LongTensorr-r7loggerwarningrBrXr:r[r6minr`r shaperZr+rUr8r)r;rQrVrGrlrmrnrorprqrHgammanoiseeps sigma_hatr derivativedtrrrrsteps:'    4     zEDMEulerScheduler.steporiginal_samplesrzr3cs jj|j|jd}|jjdkr)t|r)jj|jtjd|j|jtjd}n j|j||j}j durFfdd|D}nj durUj g|j d}n j g|j d}|| }t |j t |j kr}|d}t |j t |j ksn|||}|S)N)r+rZmps)rZcsg|]}|qSr)rh).0trer;rr sz/EDMEulerScheduler.add_noise..r)r6r:r+rZtyperis_floating_pointr3r[rErBrxflattenr` unsqueeze)r;rrzr3r6 step_indicesrH noisy_samplesrrr add_noiseps"      zEDMEulerScheduler.add_noisecCs|jjSri)r>r&r?rrr__len__szEDMEulerScheduler.__len__)rrrrrr r!)rri)NN)%rrrr _compatiblesorderrfloatstrrrr=propertyr@rBrErFrKr2rUrrrrYr+r\r0r1rhrXr Generatorboolrrrrrrrrrr1s      $      ` !r)r_ dataclassesrtypingrrrrconfiguration_utilsrrutilsr r utils.torch_utilsr scheduling_utilsr get_loggerrrurrrrrrs