Skip to content

fine_tuning

Classes

fastvideo.train.methods.fine_tuning.DiffusionForcingSFTMethod 

DiffusionForcingSFTMethod(*, cfg: Any, role_models: dict[str, ModelBase])

Bases: TrainingMethod

Diffusion-forcing SFT (DFSFT): train only student with inhomogeneous timesteps.

Source code in fastvideo/train/methods/fine_tuning/dfsft.py 
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
def __init__(
    self,
    *,
    cfg: Any,
    role_models: dict[str, ModelBase],
) -> None:
    super().__init__(cfg=cfg, role_models=role_models)

    if "student" not in role_models:
        raise ValueError("DFSFT requires role 'student'")
    if not self.student._trainable:
        raise ValueError("DFSFT requires student to be trainable")
    self._attn_kind: Literal["dense", "vsa"] = (self._infer_attn_kind())

    self._chunk_size = self._parse_chunk_size(self.method_config.get("chunk_size", None))
    self._timestep_index_range = (self._parse_timestep_index_range())
    self._training_weights = self._build_training_weights()

    # Initialize preprocessors on student.
    self.student.init_preprocessors(self.training_config)

    self._init_optimizers_and_schedulers()

fastvideo.train.methods.fine_tuning.FineTuneMethod 

FineTuneMethod(*, cfg: Any, role_models: dict[str, ModelBase])

Bases: TrainingMethod

Supervised finetuning: only student participates.

Source code in fastvideo/train/methods/fine_tuning/finetune.py 
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
def __init__(
    self,
    *,
    cfg: Any,
    role_models: dict[str, ModelBase],
) -> None:
    super().__init__(cfg=cfg, role_models=role_models)

    if "student" not in role_models:
        raise ValueError("FineTuneMethod requires role 'student'")
    if not self.student._trainable:
        raise ValueError("FineTuneMethod requires student to be "
                         "trainable")
    self._attn_kind: Literal["dense", "vsa"] = (self._infer_attn_kind())

    # Initialize preprocessors on student.
    self.student.init_preprocessors(self.training_config)

    self._init_optimizers_and_schedulers()