fastvideo.v1.training.training_utils

Module Contents

Functions

clip_grad_norm_	Clip the gradient norm of parameters.
clip_grad_norm_while_handling_failing_dtensor_cases
compute_density_for_timestep_sampling	Compute the density for sampling the timesteps when doing SD3 training.
custom_to_hf_state_dict	Convert fastvideo’s custom model format to diffusers format using reverse_param_names_mapping.
gather_state_dict_on_cpu_rank0
get_sigmas
load_checkpoint	Load checkpoint following finetrainer’s distributed checkpoint approach. Returns the step number from which training should resume.
normalize_dit_input
save_checkpoint	Save checkpoint following finetrainer’s distributed checkpoint approach. Saves both distributed checkpoint and consolidated model weights.
shard_latents_across_sp

Data

logger

API

fastvideo.v1.training.training_utils.clip_grad_norm_(parameters: torch.Tensor | list[torch.Tensor], max_norm: float, norm_type: float = 2.0, error_if_nonfinite: bool = False, foreach: bool | None = None, pp_mesh: torch.distributed.device_mesh.DeviceMesh | None = None) → torch.Tensor

Clip the gradient norm of parameters.

Gradient norm clipping requires computing the gradient norm over the entire model. torch.nn.utils.clip_grad_norm_ only computes gradient norm along DP/FSDP/TP dimensions. We need to manually reduce the gradient norm across PP stages. See https://github.com/pytorch/torchtitan/issues/596 for details.

Parameters:

• parameters (torch.Tensor or List[torch.Tensor]) – Tensors that will have gradients normalized.

• max_norm (float) – Maximum norm of the gradients after clipping.

• norm_type (float, defaults to 2.0) – Type of p-norm to use. Can be inf for infinity norm.

• error_if_nonfinite (bool, defaults to False) – If True, an error is thrown if the total norm of the gradients from parameters is nan, inf, or -inf.

• foreach (bool, defaults to None) – Use the faster foreach-based implementation. If None, use the foreach implementation for CUDA and CPU native tensors and silently fall back to the slow implementation for other device types.

• pp_mesh (torch.distributed.device_mesh.DeviceMesh, defaults to None) – Pipeline parallel device mesh. If not None, will reduce gradient norm across PP stages.

Returns:

Total norm of the gradients

Return type:

torch.Tensor

fastvideo.v1.training.training_utils.clip_grad_norm_while_handling_failing_dtensor_cases(parameters: torch.Tensor | list[torch.Tensor], max_norm: float, norm_type: float = 2.0, error_if_nonfinite: bool = False, foreach: bool | None = None, pp_mesh: torch.distributed.device_mesh.DeviceMesh | None = None) → torch.Tensor | None

fastvideo.v1.training.training_utils.compute_density_for_timestep_sampling(weighting_scheme: str, batch_size: int, generator, logit_mean: float | None = None, logit_std: float | None = None, mode_scale: float | None = None)

Compute the density for sampling the timesteps when doing SD3 training.

Courtesy: This was contributed by Rafie Walker in https://github.com/huggingface/diffusers/pull/8528.

SD3 paper reference: https://arxiv.org/abs/2403.03206v1.

fastvideo.v1.training.training_utils.custom_to_hf_state_dict(state_dict: dict[str, Any] | collections.abc.Iterator[tuple[str, torch.Tensor]], reverse_param_names_mapping: dict[str, tuple[str, int, int]]) → dict[str, Any]

Convert fastvideo’s custom model format to diffusers format using reverse_param_names_mapping.

Parameters:

• state_dict – State dict in fastvideo’s custom format

• reverse_param_names_mapping – Reverse mapping from fastvideo’s custom format to diffusers format

Returns:

State dict in diffusers format

fastvideo.v1.training.training_utils.gather_state_dict_on_cpu_rank0(model, device: torch.device | None = None) → dict[str, Any]

fastvideo.v1.training.training_utils.get_sigmas(noise_scheduler, device, timesteps, n_dim=4, dtype=torch.float32) → torch.Tensor

fastvideo.v1.training.training_utils.load_checkpoint(transformer, rank, checkpoint_path, optimizer=None, dataloader=None, scheduler=None, noise_generator=None) → int

Load checkpoint following finetrainer’s distributed checkpoint approach. Returns the step number from which training should resume.

fastvideo.v1.training.training_utils.logger

‘init_logger(…)’

fastvideo.v1.training.training_utils.normalize_dit_input(model_type, latents, args=None) → torch.Tensor

fastvideo.v1.training.training_utils.save_checkpoint(transformer, rank, output_dir, step, optimizer=None, dataloader=None, scheduler=None, noise_generator=None) → None

Save checkpoint following finetrainer’s distributed checkpoint approach. Saves both distributed checkpoint and consolidated model weights.

fastvideo.v1.training.training_utils.shard_latents_across_sp(latents: torch.Tensor, num_latent_t: int) → torch.Tensor