huggingface/pytorch-image-models · error · ValueError

Invalid epsilon value: {}

Error message

Invalid epsilon value: {}

What it means

Raised by timm's Adan optimizer constructor when eps is negative. eps is the numerical-stability term added to denominators, so it must be >= 0.

Source

Thrown at timm/optim/adan.py:76

        no_prox: How to perform the weight decay
        caution: Enable caution from 'Cautious Optimizers'
        foreach: If True would use torch._foreach implementation. Faster but uses slightly more memory.
    """

    def __init__(self,
            params,
            lr: float = 1e-3,
            betas: Tuple[float, float, float] = (0.98, 0.92, 0.99),
            eps: float = 1e-8,
            weight_decay: float = 0.0,
            no_prox: bool = False,
            caution: bool = False,
            foreach: Optional[bool] = None,
    ):
        if not 0.0 <= lr:
            raise ValueError('Invalid learning rate: {}'.format(lr))
        if not 0.0 <= eps:
            raise ValueError('Invalid epsilon value: {}'.format(eps))
        if not 0.0 <= betas[0] < 1.0:
            raise ValueError('Invalid beta parameter at index 0: {}'.format(betas[0]))
        if not 0.0 <= betas[1] < 1.0:
            raise ValueError('Invalid beta parameter at index 1: {}'.format(betas[1]))
        if not 0.0 <= betas[2] < 1.0:
            raise ValueError('Invalid beta parameter at index 2: {}'.format(betas[2]))

        defaults = dict(
            lr=lr,
            betas=betas,
            eps=eps,
            weight_decay=weight_decay,
            no_prox=no_prox,
            caution=caution,
            foreach=foreach,
        )
        super().__init__(params, defaults)

View on GitHub (pinned to 9a5261e31b)

Solutions

1. Use a small non-negative eps, typically 1e-8
2. Check the config/CLI mapping that produced the negative eps

Example fix

# before
opt = timm.optim.Adan(model.parameters(), lr=1e-3, eps=-1e-8)
# after
opt = timm.optim.Adan(model.parameters(), lr=1e-3, eps=1e-8)

assert eps >= 0.0, f'eps must be >= 0, got {eps}'

def valid_eps(eps) -> bool:
    return isinstance(eps, (int, float)) and eps >= 0.0

When it happens

Trigger: Calling timm.optim.Adan(params, eps=value) with value < 0.0.

Common situations: Typo in config (eps: -1e-8), or accidentally binding another hyperparameter's value into eps during a sweep.

Related errors

• Invalid learning rate: {}
• Invalid beta parameter at index 0: {}
• Invalid beta parameter at index 1: {}
• Invalid beta parameter at index 2: {}
• Invalid beta parameter at index 0: {}

AI-assisted analysis of huggingface/pytorch-image-models@9a5261e31b (2026-08-27). Data as JSON: /api/errors/50be02d0b224677d. Report an issue: GitHub.