FlowMatchEulerDiscreteScheduler (original) (raw)

FlowMatchEulerDiscreteScheduler is based on the flow-matching sampling introduced in Stable Diffusion 3.

class diffusers.FlowMatchEulerDiscreteScheduler

< source >

( num_train_timesteps: int = 1000 shift: float = 1.0 use_dynamic_shifting: bool = False base_shift: typing.Optional[float] = 0.5 max_shift: typing.Optional[float] = 1.15 base_image_seq_len: typing.Optional[int] = 256 max_image_seq_len: typing.Optional[int] = 4096 invert_sigmas: bool = False shift_terminal: typing.Optional[float] = None use_karras_sigmas: typing.Optional[bool] = False use_exponential_sigmas: typing.Optional[bool] = False use_beta_sigmas: typing.Optional[bool] = False time_shift_type: str = 'exponential' stochastic_sampling: bool = False )

Parameters

• num_train_timesteps (int, defaults to 1000) — The number of diffusion steps to train the model.
• shift (float, defaults to 1.0) — The shift value for the timestep schedule.
• use_dynamic_shifting (bool, defaults to False) — Whether to apply timestep shifting on-the-fly based on the image resolution.
• base_shift (float, defaults to 0.5) — Value to stabilize image generation. Increasing base_shift reduces variation and image is more consistent with desired output.
• max_shift (float, defaults to 1.15) — Value change allowed to latent vectors. Increasing max_shift encourages more variation and image may be more exaggerated or stylized.
• base_image_seq_len (int, defaults to 256) — The base image sequence length.
• max_image_seq_len (int, defaults to 4096) — The maximum image sequence length.
• invert_sigmas (bool, defaults to False) — Whether to invert the sigmas.
• shift_terminal (float, defaults to None) — The end value of the shifted timestep schedule.
• use_karras_sigmas (bool, defaults to False) — Whether to use Karras sigmas for step sizes in the noise schedule during sampling.
• use_exponential_sigmas (bool, defaults to False) — Whether to use exponential sigmas for step sizes in the noise schedule during sampling.
• use_beta_sigmas (bool, defaults to False) — Whether to use beta sigmas for step sizes in the noise schedule during sampling.
• time_shift_type (str, defaults to “exponential”) — The type of dynamic resolution-dependent timestep shifting to apply. Either “exponential” or “linear”.
• stochastic_sampling (bool, defaults to False) — Whether to use stochastic sampling.

Euler scheduler.

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.

scale_noise

< source >

( sample: FloatTensor timestep: typing.Union[float, torch.FloatTensor] noise: typing.Optional[torch.FloatTensor] = None ) → torch.FloatTensor

Parameters

• sample (torch.FloatTensor) — The input sample.
• timestep (int, optional) — The current timestep in the diffusion chain.

Returns

torch.FloatTensor

A scaled input sample.

Forward process in flow-matching

set_begin_index

< source >

( begin_index: int = 0 )

Parameters

• begin_index (int) — The begin index for the scheduler.

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

set_timesteps

< source >

( num_inference_steps: typing.Optional[int] = None device: typing.Union[str, torch.device] = None sigmas: typing.Optional[typing.List[float]] = None mu: typing.Optional[float] = None timesteps: typing.Optional[typing.List[float]] = None )

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.
• sigmas (List[float], optional) — Custom values for sigmas to be used for each diffusion step. If None, the sigmas are computed automatically.
• mu (float, optional) — Determines the amount of shifting applied to sigmas when performing resolution-dependent timestep shifting.
• timesteps (List[float], optional) — Custom values for timesteps to be used for each diffusion step. If None, the timesteps are computed automatically.

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

step

< source >

( model_output: FloatTensor timestep: typing.Union[float, torch.FloatTensor] sample: FloatTensor s_churn: float = 0.0 s_tmin: float = 0.0 s_tmax: float = inf s_noise: float = 1.0 generator: typing.Optional[torch._C.Generator] = None per_token_timesteps: typing.Optional[torch.Tensor] = None return_dict: bool = True ) → FlowMatchEulerDiscreteSchedulerOutput or tuple

Parameters

• model_output (torch.FloatTensor) — The direct output from learned diffusion model.
• timestep (float) — The current discrete timestep in the diffusion chain.
• sample (torch.FloatTensor) — 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.
• per_token_timesteps (torch.Tensor, optional) — The timesteps for each token in the sample.
• return_dict (bool) — Whether or not to return aFlowMatchEulerDiscreteSchedulerOutput or tuple.

Returns

FlowMatchEulerDiscreteSchedulerOutput or tuple

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

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).