huggingface/pytorch-image-models · error · ValueError

Invalid beta parameter at index 1: {betas[1]}

Error message

Invalid beta parameter at index 1: {betas[1]}

What it means

NAdamW constructor validation: betas[1] (beta2, variance decay) must satisfy 0.0 <= beta2 < 1.0. Values at or above 1 break the bias-corrected second-moment estimate.

Source

Thrown at timm/optim/nadamw.py:62

            params: ParamsT,
            lr: float = 1e-3,
            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:

View on GitHub (pinned to 9a5261e31b)

Solutions

1. Use standard betas like (0.9, 0.999)
2. Keep beta2 strictly below 1 in search space definitions

Example fix

# before
opt = NAdamW(model.parameters(), betas=(0.9, 1.0))

# after
opt = NAdamW(model.parameters(), betas=(0.9, 0.999))

assert 0.0 <= betas[1] < 1.0

When it happens

Trigger: Calling timm.optim.NAdamW(params, betas=(0.9, 1.0)) or betas=(0.9, 1.5).

Common situations: Hyperparameter sweeps hitting the boundary 1.0, or configs written for optimizers that allow beta2=1.

Related errors

• Invalid beta parameter at index 0: {betas[0]}
• Invalid learning rate: {lr}
• Invalid epsilon value: {eps}
• Invalid weight_decay value: {weight_decay}
• Invalid beta parameter at index 0: {}

AI-assisted analysis of huggingface/pytorch-image-models@9a5261e31b (2026-08-27). Data as JSON: /api/errors/908e8b2cafbee7d0. Report an issue: GitHub.