class Config:
optim: Optim
model: str = chika.choices("resnet20", "resnet56", "se_resnet56", "wrn28_2", "resnext29_32x4d")
batch_size: int = 128
use_amp: bool = False
seed: int = 1
gpu: int = chika.bounded(0, 0, torch.cuda.device_count())
class Optim:
epochs: int = 200
name: str = chika.choices("abel", "steps", "cosine")
lr: float = 0.1
weight_decay: float = 5e-4
gamma: float = 0.1
steps: List[int] = chika.sequence(100, 150)
class Config:
model: str = chika.choices(*MODEL_REGISTRY.choices())
batch_size: int = 128
epochs: int = 200
lr: float = 0.1
weight_decay: float = 1e-4
data: str = chika.choices("cifar10", "cifar100", "svhn")
bn_no_wd: bool = False
use_amp: bool = False
use_accimage: bool = False
use_multi_tensor: bool = False
use_channel_last: bool = False
prefetch_factor: int = 2
persistent_workers: bool = False
debug: bool = False
download: bool = False
class Config:
name: str = chika.choices(*MODEL_REGISTRY.choices())
batch_size: int = 128
epochs: int = 200
lr: float = 0.1
weight_decay: float = 1e-4
lr_decay: float = 0.1
bn_no_wd: bool = False
use_amp: bool = False
use_accimage: bool = False
use_prefetcher: bool = False
debug: bool = False
class ModelConfig:
name: str = chika.choices(*MLPMixers.choices())
droppath_rate: float = 0.1
grad_clip: float = 1
ema: bool = False
ema_rate: float = chika.bounded(0.999, 0, 1)