diffusers/pr_11739/en/api/schedulers/multistep_dpm_solver.md

DPMSolverMultistepScheduler

DPMSolverMultistepScheduler is a multistep scheduler from DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps and DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models by Cheng Lu, Yuhao Zhou, Fan Bao, Jianfei Chen, Chongxuan Li, and Jun Zhu.

DPMSolver (and the improved version DPMSolver++) is a fast dedicated high-order solver for diffusion ODEs with convergence order guarantee. Empirically, DPMSolver sampling with only 20 steps can generate high-quality samples, and it can generate quite good samples even in 10 steps.

Tips

It is recommended to set solver_order to 2 for guide sampling, and solver_order=3 for unconditional sampling.

Dynamic thresholding from Imagen is supported, and for pixel-space diffusion models, you can set both algorithm_type="dpmsolver++" and thresholding=True to use the dynamic thresholding. This thresholding method is unsuitable for latent-space diffusion models such as Stable Diffusion.

The SDE variant of DPMSolver and DPM-Solver++ is also supported, but only for the first and second-order solvers. This is a fast SDE solver for the reverse diffusion SDE. It is recommended to use the second-order sde-dpmsolver++.

DPMSolverMultistepScheduler[[diffusers.DPMSolverMultistepScheduler]]

diffusers.DPMSolverMultistepScheduler[[diffusers.DPMSolverMultistepScheduler]]

Source

DPMSolverMultistepScheduler is a fast dedicated high-order solver for diffusion ODEs.

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.

add_noisediffusers.DPMSolverMultistepScheduler.add_noisehttps://github.com/huggingface/diffusers/blob/vr_11739/src/diffusers/schedulers/scheduling_dpmsolver_multistep.py#L1262[{"name": "original_samples", "val": ": Tensor"}, {"name": "noise", "val": ": Tensor"}, {"name": "timesteps", "val": ": IntTensor"}]- original_samples (torch.Tensor) -- The original samples without noise.

• noise (torch.Tensor) -- The noise to add to the samples.
• timesteps (torch.IntTensor) -- The timesteps at which to add noise to the samples.0torch.TensorThe noisy samples.

Add noise to the original samples according to the noise schedule at the specified timesteps.

Parameters:

num_train_timesteps (int, defaults to 1000) : The number of diffusion steps to train the model.

beta_start (float, defaults to 0.0001) : The starting beta value of inference.

beta_end (float, defaults to 0.02) : The final beta value.

beta_schedule ("linear", "scaled_linear", or "squaredcos_cap_v2", defaults to "linear") : The beta schedule, a mapping from a beta range to a sequence of betas for stepping the model.

trained_betas (np.ndarray, optional) : Pass an array of betas directly to the constructor to bypass beta_start and beta_end.

solver_order (int, defaults to 2) : The DPMSolver order which can be 1 or 2 or 3. It is recommended to use solver_order=2 for guided sampling, and solver_order=3 for unconditional sampling.

prediction_type ("epsilon", "sample", "v_prediction", or "flow_prediction", defaults to "epsilon") : Prediction type of the scheduler function. epsilon predicts the noise of the diffusion process, sample directly predicts the noisy sample, v_prediction predicts the velocity (see section 2.4 of Imagen Video paper), and flow_prediction predicts the flow.

thresholding (bool, defaults to False) : Whether to use the "dynamic thresholding" method. This is unsuitable for latent-space diffusion models such as Stable Diffusion.

dynamic_thresholding_ratio (float, defaults to 0.995) : The ratio for the dynamic thresholding method. Valid only when thresholding=True.

sample_max_value (float, defaults to 1.0) : The threshold value for dynamic thresholding. Valid only when thresholding=True and algorithm_type="dpmsolver++".

algorithm_type ("dpmsolver", "dpmsolver++", "sde-dpmsolver", or "sde-dpmsolver++", defaults to "dpmsolver++") : Algorithm type for the solver. The dpmsolver type implements the algorithms in the DPMSolver paper, and the dpmsolver++ type implements the algorithms in the DPMSolver++ paper. It is recommended to use dpmsolver++ or sde-dpmsolver++ with solver_order=2 for guided sampling like in Stable Diffusion.

solver_type ("midpoint" or "heun", defaults to "midpoint") : Solver type for the second-order solver. The solver type slightly affects the sample quality, especially for a small number of steps. It is recommended to use midpoint solvers.

lower_order_final (bool, defaults to True) : Whether to use lower-order solvers in the final steps. Only valid for [!TIP] > The algorithm and model type are decoupled. You can use either DPMSolver or DPMSolver++ for both noise > prediction and data prediction models.

Parameters:

model_output (torch.Tensor) : The direct output from the learned diffusion model.

sample (torch.Tensor) : A current instance of a sample created by the diffusion process.

Returns:

torch.Tensor

The converted model output.

dpm_solver_first_order_update[[diffusers.DPMSolverMultistepScheduler.dpm_solver_first_order_update]]

Source

One step for the first-order DPMSolver (equivalent to DDIM).

Parameters:

model_output (torch.Tensor) : The direct output from the learned diffusion model.

sample (torch.Tensor) : A current instance of a sample created by the diffusion process.

Returns:

torch.Tensor

The sample tensor at the previous timestep.

index_for_timestep[[diffusers.DPMSolverMultistepScheduler.index_for_timestep]]

Source

Find the index for a given timestep in the schedule.

Parameters:

timestep (int or torch.Tensor) : The timestep for which to find the index.

schedule_timesteps (torch.Tensor, optional) : The timestep schedule to search in. If None, uses self.timesteps.

Returns:

int

The index of the timestep in the schedule.

multistep_dpm_solver_second_order_update[[diffusers.DPMSolverMultistepScheduler.multistep_dpm_solver_second_order_update]]

Source

One step for the second-order multistep DPMSolver.

Parameters:

model_output_list (List[torch.Tensor]) : The direct outputs from learned diffusion model at current and latter timesteps.

sample (torch.Tensor) : A current instance of a sample created by the diffusion process.

Returns:

torch.Tensor

The sample tensor at the previous timestep.

multistep_dpm_solver_third_order_update[[diffusers.DPMSolverMultistepScheduler.multistep_dpm_solver_third_order_update]]

Source

One step for the third-order multistep DPMSolver.

Parameters:

model_output_list (List[torch.Tensor]) : The direct outputs from learned diffusion model at current and latter timesteps.

sample (torch.Tensor) : A current instance of a sample created by diffusion process.

Returns:

torch.Tensor

The sample tensor at the previous timestep.

scale_model_input[[diffusers.DPMSolverMultistepScheduler.scale_model_input]]

Source

Ensures interchangeability with schedulers that need to scale the denoising model input depending on the current timestep.

Parameters:

sample (torch.Tensor) : The input sample.

Returns:

torch.Tensor

A scaled input sample.

set_begin_index[[diffusers.DPMSolverMultistepScheduler.set_begin_index]]

Source

Sets the begin index for the scheduler. This function should be run from pipeline before the inference.

Parameters:

begin_index (int, defaults to 0) : The begin index for the scheduler.

set_timesteps[[diffusers.DPMSolverMultistepScheduler.set_timesteps]]

Source

Sets the discrete timesteps used for the diffusion chain (to be run before inference).

Parameters:

num_inference_steps (int, optional) : 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.

mu (float, optional) : The mu parameter for dynamic shifting. If provided, requires use_dynamic_shifting=True and time_shift_type="exponential".

timesteps (List[int], optional) : Custom timesteps used to support arbitrary timesteps schedule. If None, timesteps will be generated based on the timestep_spacing attribute. If timesteps is passed, num_inference_steps and sigmas must be None, and timestep_spacing attribute will be ignored.

step[[diffusers.DPMSolverMultistepScheduler.step]]

Source

Predict the sample from the previous timestep by reversing the SDE. This function propagates the sample with the multistep DPMSolver.

Parameters:

model_output (torch.Tensor) : The direct output from the learned diffusion model.

timestep (int or torch.Tensor) : The current discrete 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.

variance_noise (torch.Tensor, optional) : Alternative to generating noise with generator by directly providing the noise for the variance itself. Useful for methods such as LEdits++.

return_dict (bool, defaults to True) : Whether or not to return a SchedulerOutput or tuple.

Returns:

[SchedulerOutput](/docs/diffusers/pr_11739/en/api/schedulers/overview#diffusers.schedulers.scheduling_utils.SchedulerOutput) or tuple``

If return_dict is True, SchedulerOutput is returned, otherwise a tuple is returned where the first element is the sample tensor.

SchedulerOutput[[diffusers.schedulers.scheduling_utils.SchedulerOutput]]

diffusers.schedulers.scheduling_utils.SchedulerOutput[[diffusers.schedulers.scheduling_utils.SchedulerOutput]]

Source

Base class for the output of a scheduler's step function.

Parameters:

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.