o Gi) @sddlZddlmZddlmZddlZddlmZmZddl m Z ddl m Z dd l mZ dd ed ededdejfddZeGddde ZGdddeeZdS)N) dataclass)Literal) ConfigMixinregister_to_config) BaseOutput) randn_tensor)SchedulerMixin+?cosinenum_diffusion_timestepsmax_betaalpha_transform_type)r explaplacereturncCs|dkr dd}n|dkrdd}n|dkrdd}ntd|g}t|D]}||}|d |}|td |||||q(tj|tjd S) a> Create a beta schedule that discretizes the given alpha_t_bar function, which defines the cumulative product of (1-beta) over time from t = [0,1]. Contains a function alpha_bar that takes an argument t and transforms it to the cumulative product of (1-beta) up to that part of the diffusion process. Args: num_diffusion_timesteps (`int`): The number of betas to produce. max_beta (`float`, defaults to `0.999`): The maximum beta to use; use values lower than 1 to avoid numerical instability. alpha_transform_type (`str`, defaults to `"cosine"`): The type of noise schedule for `alpha_bar`. Choose from `cosine`, `exp`, or `laplace`. Returns: `torch.Tensor`: The betas used by the scheduler to step the model outputs. r cSs t|ddtjddS)NgMb?gT㥛 ?r)mathcospitrg/home/ubuntu/.local/lib/python3.10/site-packages/diffusers/schedulers/scheduling_consistency_decoder.py alpha_bar_fn(s z)betas_for_alpha_bar..alpha_bar_fnrc SsPdtdd|tddtd|d}t|}t|d|S)Ngr ?rgư>)rcopysignlogfabsrsqrt)rlmbsnrrrrr-s4 rcSst|dS)Ng()rrrrrrr4sz"Unsupported alpha_transform_type: r )dtype) ValueErrorrangeappendmintorchtensorfloat32)r rrrbetasit1t2rrrbetas_for_alpha_bars     "r.c@seZdZUdZejed<dS)!ConsistencyDecoderSchedulerOutputa> Output class for the scheduler's `step` function. 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. prev_sampleN)__name__ __module__ __qualname____doc__r'Tensor__annotations__rrrrr/Bs  r/c@seZdZdZdZe  ddededdfd d Z  dd edBd e e j Bfd dZ e de jfddZdde jdedBde jfddZ  dde jdee jBde jde jdBdedeeBf ddZdS)ConsistencyDecoderSchedulera A scheduler for the consistency decoder used in Stable Diffusion pipelines. This scheduler implements a two-step denoising process using consistency models for decoding latent representations into images. 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: num_train_timesteps (`int`, *optional*, defaults to `1024`): The number of diffusion steps to train the model. sigma_data (`float`, *optional*, defaults to `0.5`): The standard deviation of the data distribution. Used for computing the skip and output scaling factors. r rnum_train_timesteps sigma_datarNcCst|}d|}tj|dd}t||_td||_td|d}td|}||d|d|d|_|||d|dd|_||d|dd|_dS)Ng?r)dimr rr) r.r'cumprodrsqrt_alphas_cumprodsqrt_one_minus_alphas_cumprodc_skipc_outc_in)selfr9r:r*alphasalphas_cumprodsigmassqrt_recip_alphas_cumprodrrr__init__cs z$ConsistencyDecoderScheduler.__init__num_inference_stepsdevicecCsr|dkrtdtjddgtj|d|_|j||_|j||_|j||_|j ||_ |j ||_ dS)Nrz8Currently more than 2 inference steps are not supported.ii)r"rI) r#r'r(long timestepsr=tor>r?r@rA)rBrHrIrrr set_timestepsysz)ConsistencyDecoderScheduler.set_timestepscCs|j|jdS)a  Return the standard deviation of the initial noise distribution. Returns: `torch.Tensor`: The initial noise sigma value from the precomputed `sqrt_one_minus_alphas_cumprod` at the first timestep. r)r>rK)rBrrrinit_noise_sigmas z,ConsistencyDecoderScheduler.init_noise_sigmasampletimestepcCs||j|S)a Ensures interchangeability with schedulers that need to scale the denoising model input depending on the current timestep. Args: sample (`torch.Tensor`): The input sample. timestep (`int`, *optional*): The current timestep in the diffusion chain. Returns: `torch.Tensor`: A scaled input sample. )rA)rBrOrPrrrscale_model_inputsz-ConsistencyDecoderScheduler.scale_model_inputT model_output generator return_dictc Cs|j|||j||}t|j|kd}|t|jdkr$|}n+t|j||j|j d} |j |j|d |j||j |j|d |j| }|sT|fSt |dS)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 the learned diffusion model. timestep (`float` or `torch.Tensor`): The current timestep in the diffusion chain. sample (`torch.Tensor`): A current instance of a sample created by the diffusion process. generator (`torch.Generator`, *optional*): A random number generator for reproducibility. return_dict (`bool`, *optional*, defaults to `True`): Whether or not to return a [`~schedulers.scheduling_consistency_decoder.ConsistencyDecoderSchedulerOutput`] or `tuple`. Returns: [`~schedulers.scheduling_consistency_decoder.ConsistencyDecoderSchedulerOutput`] or `tuple`: If `return_dict` is `True`, [`~schedulers.scheduling_consistency_decoder.ConsistencyDecoderSchedulerOutput`] is returned, otherwise a tuple is returned where the first element is the sample tensor. rr )rSr"rI)r0)r@r?r'whererKlenrshaper"rIr=rLr>r/) rBrRrPrOrSrTx_0 timestep_idxr0noiserrrsteps z ConsistencyDecoderScheduler.step)r8r)NN)N)NT)r1r2r3r4orderrintfloatrGstrr'rIrMpropertyr5rNrQ Generatorboolr/tupler[rrrrr7PsL   r7)r r )r dataclassesrtypingrr'configuration_utilsrrutilsrutils.torch_utilsrscheduling_utilsr r]r^r5r.r/r7rrrrs,      4