Loss Functions#
Classes#
- class BaseScheduler[source]#
Bases:
ABCAbstract base class for schedulers.
- extra_repr()[source]#
Additional string representation of the scheduler.
- Returns:
strAdditional string representation of the scheduler
- class BaseFuncScheduler(max_epochs, *, warmup_epochs=0, final_value=1, initial_value=0)[source]#
Bases:
BaseScheduler,ABCAbstract base class for simple function schedulers.
- Attributes:
- Parameters:
- class LinearScheduler(max_epochs, *, warmup_epochs=0, final_value=1, initial_value=0)[source]#
Bases:
BaseFuncSchedulerLinear scheduler.
- Attributes:
- Parameters:
- class SigmoidScheduler(max_epochs, *, warmup_epochs=0, final_value=1, initial_value=0, compression=0)[source]#
Bases:
BaseFuncSchedulerSigmoid scheduler.
- Attributes:
- Parameters:
- max_epochs
int Maximum number of epochs
- warmup_epochs
int,Optional Number of warmup epochs, default = 0
- final_value
float,Optional Maximum value of the scheduler, default = 1
- initial_value
float,Optional Minimum value of the scheduler, default = 0
- compression
float,Optional Compression factor for the tanh function, higher values result in a steeper curve, default = 10
- max_epochs
- class StepScheduler(warmup, max_steps, step_size, gamma, initial_value=1)[source]#
Bases:
BaseSchedulerStep scheduler.
- Attributes:
- Parameters:
- class TanhScheduler(max_epochs, *, warmup_epochs=0, final_value=1, initial_value=0, compression=0)[source]#
Bases:
SigmoidSchedulerTanh scheduler.
- Attributes:
- Parameters:
- max_epochs
int Maximum number of epochs
- warmup_epochs
int,Optional Number of warmup epochs, default = 0
- final_value
float,Optional Maximum value of the scheduler, default = 1
- initial_value
float,Optional Minimum value of the scheduler, default = 0
- compression
float,Optional Compression factor for the sigmoid function, higher values result in a steeper curve, default = 5
- max_epochs