ride.unfreeze

Module Contents

Classes

Unfreezable

Abstract base-class for Ride mixins

Functions

freeze_layers_except_names(parent_module, ...)
get_modules_to_unfreeze(→ Sequence[Tuple[str, ...)
unfreeze_from_end(layers, num_layers_from_end[, ...])
linear_unfreeze_schedule(→ Dict[int, int])

Attributes

logger

ride.unfreeze.logger

class ride.unfreeze.Unfreezable(hparams: pytorch_lightning.utilities.parsing.AttributeDict, *args, **kwargs)

Bases: ride.core.RideMixin

Abstract base-class for Ride mixins

hparams: Ellipsis

validate_attributes()

static configs() → ride.core.Configs

on_init_end(hparams, layers_to_unfreeze: Sequence[Tuple[str, torch.nn.Module]] = None, names_to_unfreeze: Sequence[str] = None, *args, **kwargs)

on_traning_epoch_start(epoch: int)

ride.unfreeze.freeze_layers_except_names(parent_module: torch.nn.Module, names_to_unfreeze: Sequence[str])

ride.unfreeze.get_modules_to_unfreeze(parent_module: torch.nn.Module, name_must_include='') → Sequence[Tuple[str, torch.nn.Module]]

ride.unfreeze.unfreeze_from_end(layers: Sequence[Tuple[str, torch.nn.Module]], num_layers_from_end: int, freeze_others=False)

ride.unfreeze.linear_unfreeze_schedule(initial_epoch: int, total_layers: int, step_size: int = 1, init_layers: int = 0, max_layers: int = -1, epoch_step: int = 1) → Dict[int, int]