You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
17 lines
761 B
17 lines
761 B
import torch
|
|
from typing import Any, Dict, Tuple
|
|
from pydantic import BaseModel, ConfigDict
|
|
|
|
class AdamWParams(BaseModel):
|
|
"""Configuration for `torch.optim.AdamW` optimizer."""
|
|
model_config = ConfigDict(frozen=True)
|
|
|
|
lr: float = 1e-3 # Learning rate
|
|
betas: Tuple[float, ...] = (0.9, 0.999) # Adam coefficients
|
|
eps: float = 1e-8 # Numerical stability
|
|
weight_decay: float = 1e-2 # L2 penalty (AdamW uses decoupled weight decay)
|
|
amsgrad: bool = False # Whether to use the AMSGrad variant
|
|
|
|
def asdict(self) -> Dict[str, Any]:
|
|
"""Returns a dictionary of valid parameters for `torch.optim.AdamW`."""
|
|
return self.model_dump() |