path: root/schedulers/scheduling_euler_a.py

from typing import Optional, Tuple, Union

import numpy as np
import torch

from diffusers.configuration_utils import ConfigMixin, register_to_config
from diffusers.schedulers.scheduling_utils import SchedulerMixin, SchedulerOutput


class EulerAScheduler(SchedulerMixin, ConfigMixin):
    """
    Stochastic sampling from Karras et al. [1] tailored to the Variance-Expanding (VE) models [2]. Use Algorithm 2 and
    the VE column of Table 1 from [1] for reference.

    [1] Karras, Tero, et al. "Elucidating the Design Space of Diffusion-Based Generative Models."
    https://arxiv.org/abs/2206.00364 [2] Song, Yang, et al. "Score-based generative modeling through stochastic
    differential equations." https://arxiv.org/abs/2011.13456

    [`~ConfigMixin`] takes care of storing all config attributes that are passed in the scheduler's `__init__`
    function, such as `num_train_timesteps`. They can be accessed via `scheduler.config.num_train_timesteps`.
    [`~ConfigMixin`] also provides general loading and saving functionality via the [`~ConfigMixin.save_config`] and
    [`~ConfigMixin.from_config`] functions.

    For more details on the parameters, see the original paper's Appendix E.: "Elucidating the Design Space of
    Diffusion-Based Generative Models." https://arxiv.org/abs/2206.00364. The grid search values used to find the
    optimal {s_noise, s_churn, s_min, s_max} for a specific model are described in Table 5 of the paper.

    Args:
        sigma_min (`float`): minimum noise magnitude
        sigma_max (`float`): maximum noise magnitude
        s_noise (`float`): the amount of additional noise to counteract loss of detail during sampling.
            A reasonable range is [1.000, 1.011].
        s_churn (`float`): the parameter controlling the overall amount of stochasticity.
            A reasonable range is [0, 100].
        s_min (`float`): the start value of the sigma range where we add noise (enable stochasticity).
            A reasonable range is [0, 10].
        s_max (`float`): the end value of the sigma range where we add noise.
            A reasonable range is [0.2, 80].

    """

    @register_to_config
    def __init__(
        self,
        num_train_timesteps: int = 1000,
        beta_start: float = 0.0001,
        beta_end: float = 0.02,
        beta_schedule: str = "linear",
        trained_betas: Optional[np.ndarray] = None,
        tensor_format: str = "pt",
        num_inference_steps=None,
        device='cuda'
    ):
        if trained_betas is not None:
            self.betas = torch.from_numpy(trained_betas).to(device)
        if beta_schedule == "linear":
            self.betas = torch.linspace(beta_start, beta_end, num_train_timesteps, dtype=torch.float32, device=device)
        elif beta_schedule == "scaled_linear":
            # this schedule is very specific to the latent diffusion model.
            self.betas = torch.linspace(beta_start**0.5, beta_end**0.5, num_train_timesteps,
                                        dtype=torch.float32, device=device) ** 2
        else:
            raise NotImplementedError(f"{beta_schedule} does is not implemented for {self.__class__}")

        self.device = device
        self.tensor_format = tensor_format

        self.alphas = 1.0 - self.betas
        self.alphas_cumprod = torch.cumprod(self.alphas, dim=0)

        # standard deviation of the initial noise distribution
        self.init_noise_sigma = 1.0

        # setable values
        self.num_inference_steps = num_inference_steps
        self.timesteps = np.arange(0, num_train_timesteps)[::-1].copy()
        # get sigmas
        self.DSsigmas = ((1 - self.alphas_cumprod) / self.alphas_cumprod) ** 0.5
        self.sigmas = self.get_sigmas(self.DSsigmas, self.num_inference_steps)
        self.set_format(tensor_format=tensor_format)

    # A# take number of steps as input
    # A# store 1) number of steps 2) timesteps 3) schedule

    def set_timesteps(self, num_inference_steps: int, device: Union[str, torch.device] = None, **kwargs):
        """
        Sets the discrete timesteps used for the diffusion chain. Supporting function to be run before inference.

        Args:
            num_inference_steps (`int`):
                the number of diffusion steps used when generating samples with a pre-trained model.
        """

        self.num_inference_steps = num_inference_steps
        self.DSsigmas = ((1 - self.alphas_cumprod) / self.alphas_cumprod) ** 0.5
        self.sigmas = self.get_sigmas(self.DSsigmas, self.num_inference_steps)
        self.timesteps = self.sigmas[:-1]
        self.is_scale_input_called = False

    def scale_model_input(self, sample: torch.FloatTensor, timestep: int) -> torch.FloatTensor:
        """
        Ensures interchangeability with schedulers that need to scale the denoising model input depending on the
        current timestep.
        Args:
            sample (`torch.FloatTensor`): input sample
            timestep (`int`, optional): current timestep
        Returns:
            `torch.FloatTensor`: scaled input sample
        """
        if isinstance(timestep, torch.Tensor):
            timestep = timestep.to(self.timesteps.device)
        if self.is_scale_input_called:
            return sample
        step_index = (self.timesteps == timestep).nonzero().item()
        sigma = self.sigmas[step_index]
        sample = sample * sigma
        self.is_scale_input_called = True
        return sample

    def step(
        self,
        model_output: torch.FloatTensor,
        timestep: Union[float, torch.FloatTensor],
        sample: torch.FloatTensor,
        generator: torch.Generator = None,
        return_dict: bool = True,
    ) -> Union[SchedulerOutput, Tuple]:
        """
        Predict the sample at the previous timestep by reversing the SDE. Core function to propagate the diffusion
        process from the learned model outputs (most often the predicted noise).

        Args:
            model_output (`torch.FloatTensor`): direct output from learned diffusion model.
            sigma_hat (`float`): TODO
            sigma_prev (`float`): TODO
            sample_hat (`torch.FloatTensor`): TODO
            return_dict (`bool`): option for returning tuple rather than SchedulerOutput class

            EulerAOutput: updated sample in the diffusion chain and derivative (TODO double check).
        Returns:
            [`~schedulers.scheduling_karras_ve.EulerAOutput`] or `tuple`:
            [`~schedulers.scheduling_karras_ve.EulerAOutput`] if `return_dict` is True, otherwise a `tuple`. When
            returning a tuple, the first element is the sample tensor.

        """
        if isinstance(timestep, torch.Tensor):
            timestep = timestep.to(self.timesteps.device)
        step_index = (self.timesteps == timestep).nonzero().item()
        step_prev_index = step_index + 1

        s = self.sigmas[step_index]
        s_prev = self.sigmas[step_prev_index]
        latents = sample

        sigma_down, sigma_up = self.get_ancestral_step(s, s_prev)
        d = self.to_d(latents, s, model_output)
        dt = sigma_down - s
        latents = latents + d * dt
        latents = latents + torch.randn(latents.shape, layout=latents.layout, device=latents.device, dtype=latents.dtype,
                                        generator=generator) * sigma_up

        return SchedulerOutput(prev_sample=latents)

    def step_correct(
        self,
        model_output: torch.FloatTensor,
        sigma_hat: float,
        sigma_prev: float,
        sample_hat: torch.FloatTensor,
        sample_prev: torch.FloatTensor,
        derivative: torch.FloatTensor,
        return_dict: bool = True,
    ) -> Union[SchedulerOutput, Tuple]:
        """
        Correct the predicted sample based on the output model_output of the network. TODO complete description

        Args:
            model_output (`torch.FloatTensor`): direct output from learned diffusion model.
            sigma_hat (`float`): TODO
            sigma_prev (`float`): TODO
            sample_hat (`torch.FloatTensor`): TODO
            sample_prev (`torch.FloatTensor`): TODO
            derivative (`torch.FloatTensor`): TODO
            return_dict (`bool`): option for returning tuple rather than SchedulerOutput class

        Returns:
            prev_sample (TODO): updated sample in the diffusion chain. derivative (TODO): TODO

        """
        pred_original_sample = sample_prev + sigma_prev * model_output
        derivative_corr = (sample_prev - pred_original_sample) / sigma_prev
        sample_prev = sample_hat + (sigma_prev - sigma_hat) * (0.5 * derivative + 0.5 * derivative_corr)

        if not return_dict:
            return (sample_prev, derivative)

        return SchedulerOutput(prev_sample=sample_prev)

    def add_noise(
        self,
        original_samples: torch.FloatTensor,
        noise: torch.FloatTensor,
        timesteps: torch.FloatTensor,
    ) -> torch.FloatTensor:
        sigmas = self.sigmas.to(original_samples.device)
        schedule_timesteps = self.timesteps.to(original_samples.device)
        timesteps = timesteps.to(original_samples.device)
        step_indices = [(schedule_timesteps == t).nonzero().item() for t in timesteps]

        sigma = sigmas[step_indices].flatten()
        while len(sigma.shape) < len(original_samples.shape):
            sigma = sigma.unsqueeze(-1)

        noisy_samples = original_samples + noise * sigma
        self.is_scale_input_called = True
        return noisy_samples

    # from k_samplers sampling.py

    def get_ancestral_step(self, sigma_from, sigma_to):
        """Calculates the noise level (sigma_down) to step down to and the amount
        of noise to add (sigma_up) when doing an ancestral sampling step."""
        sigma_up = (sigma_to ** 2 * (sigma_from ** 2 - sigma_to ** 2) / sigma_from ** 2) ** 0.5
        sigma_down = (sigma_to ** 2 - sigma_up ** 2) ** 0.5
        return sigma_down, sigma_up

    def t_to_sigma(self, t, sigmas):
        t = t.float()
        low_idx, high_idx, w = t.floor().long(), t.ceil().long(), t.frac()
        return (1 - w) * sigmas[low_idx] + w * sigmas[high_idx]

    def append_zero(self, x):
        return torch.cat([x, x.new_zeros([1])])

    def get_sigmas(self, sigmas, n=None):
        if n is None:
            return self.append_zero(sigmas.flip(0))
        t_max = len(sigmas) - 1  # = 999
        device = self.device
        t = torch.linspace(t_max, 0, n, device=device)
        # t = torch.linspace(t_max, 0, n, device=sigmas.device)
        return self.append_zero(self.t_to_sigma(t, sigmas))

    # from k_samplers utils.py
    def append_dims(self, x, target_dims):
        """Appends dimensions to the end of a tensor until it has target_dims dimensions."""
        dims_to_append = target_dims - x.ndim
        if dims_to_append < 0:
            raise ValueError(f'input has {x.ndim} dims but target_dims is {target_dims}, which is less')
        return x[(...,) + (None,) * dims_to_append]

    # from k_samplers sampling.py
    def to_d(self, x, sigma, denoised):
        """Converts a denoiser output to a Karras ODE derivative."""
        return (x - denoised) / self.append_dims(sigma, x.ndim)

    def get_scalings(self, sigma):
        sigma_data = 1.
        c_out = -sigma
        c_in = 1 / (sigma ** 2 + sigma_data ** 2) ** 0.5
        return c_out, c_in

    # DiscreteSchedule DS
    def DSsigma_to_t(self, sigma, quantize=None):
        # quantize = self.quantize if quantize is None else quantize
        quantize = False
        dists = torch.abs(sigma - self.DSsigmas[:, None])
        if quantize:
            return torch.argmin(dists, dim=0).view(sigma.shape)
        low_idx, high_idx = torch.sort(torch.topk(dists, dim=0, k=2, largest=False).indices, dim=0)[0]
        low, high = self.DSsigmas[low_idx], self.DSsigmas[high_idx]
        w = (low - sigma) / (low - high)
        w = w.clamp(0, 1)
        t = (1 - w) * low_idx + w * high_idx
        return t.view(sigma.shape)

    def prepare_input(self, latent_in, t, batch_size):
        sigma = t.reshape(1)  # A# potential bug: doesn't work on samples > 1

        sigma_in = torch.cat([sigma] * 2 * batch_size)
        # noise_pred = CFGDenoiserForward(self.unet, latent_model_input, sigma_in, text_embeddings , guidance_scale,DSsigmas=self.scheduler.DSsigmas)
        # noise_pred = DiscreteEpsDDPMDenoiserForward(self.unet,latent_model_input, sigma_in,DSsigmas=self.scheduler.DSsigmas, cond=cond_in)
        c_out, c_in = [self.append_dims(x, latent_in.ndim) for x in self.get_scalings(sigma_in)]

        sigma_in = self.DSsigma_to_t(sigma_in)
        # s_in = latent_in.new_ones([latent_in.shape[0]])
        # sigma_in = sigma_in * s_in

        return c_out, c_in, sigma_in