huggingface/pytorch-image-models · error · ValueError

Invalid weight_decay value: {weight_decay}

Error message

Invalid weight_decay value: {weight_decay}

What it means

NAdamW constructor validation: weight_decay must satisfy 0.0 <= weight_decay. Negative decoupled weight decay is rejected at construction.

Source

Thrown at timm/optim/nadamw.py:64

            betas: Tuple[float, float] = (0.9, 0.999),
            eps: float = 1e-8,
            weight_decay: float = 1e-2,
            caution: bool = False,
            corrected_weight_decay: bool = False,
            maximize: bool = False,
            foreach: Optional[bool] = None,
            capturable: bool = False,
    ):
        if not 0.0 <= lr:
            raise ValueError(f'Invalid learning rate: {lr}')
        if not 0.0 <= eps:
            raise ValueError(f'Invalid epsilon value: {eps}')
        if not 0.0 <= betas[0] < 1.0:
            raise ValueError(f'Invalid beta parameter at index 0: {betas[0]}')
        if not 0.0 <= betas[1] < 1.0:
            raise ValueError(f'Invalid beta parameter at index 1: {betas[1]}')
        if not 0.0 <= weight_decay:
            raise ValueError(f'Invalid weight_decay value: {weight_decay}')
        defaults = dict(
            lr=lr,
            betas=betas,
            eps=eps,
            weight_decay=weight_decay,
            caution=caution,
            corrected_weight_decay=corrected_weight_decay,
            foreach=foreach,
            maximize=maximize,
            capturable=capturable,
        )
        super().__init__(params, defaults)

    def __setstate__(self, state):
        super().__setstate__(state)
        for group in self.param_groups:
            group.setdefault('caution', False)
            group.setdefault('corrected_weight_decay', False)

View on GitHub (pinned to 9a5261e31b)

Solutions

1. Use a non-negative weight_decay such as 0.05
2. If you intended weight growth, that is not supported; use 0 or positive decay

Example fix

# before
opt = NAdamW(model.parameters(), weight_decay=-0.05)

# after
opt = NAdamW(model.parameters(), weight_decay=0.05)

assert weight_decay >= 0.0

When it happens

Trigger: Calling timm.optim.NAdamW(params, weight_decay=-0.05).

Common situations: Sign confusion when implementing weight growth experiments, or a config value typo.

Related errors

• Invalid learning rate: {lr}
• Invalid epsilon value: {eps}
• Invalid beta parameter at index 0: {betas[0]}
• Invalid beta parameter at index 1: {betas[1]}
• Preset '{value}' is empty or invalid

AI-assisted analysis of huggingface/pytorch-image-models@9a5261e31b (2026-08-27). Data as JSON: /api/errors/ecccc2171a9aea60. Report an issue: GitHub.