{"record":{"id":"9fb137252ced195c","repo":"huggingface/pytorch-image-models","slug":"invalid-learning-rate-9fb137","errorCode":null,"errorMessage":"Invalid learning rate: {}","messagePattern":"Invalid learning rate: (.+?)","errorType":"validation","errorClass":"ValueError","httpStatus":null,"severity":"error","filePath":"timm/optim/mars.py","lineNumber":113,"sourceCode":"        https://arxiv.org/abs/2411.10438\n\n    \"\"\"\n    def __init__(\n            self,\n            params: ParamsT,\n            lr: float = 3e-3,\n            betas: Tuple[float, float] = (0.9, 0.99),\n            eps: float = 1e-8,\n            weight_decay: float = 0.,\n            gamma: float = 0.025,\n            mars_type: str = \"adamw\",\n            optimize_1d: bool = False,\n            lr_1d_factor: float = 1.0,\n            betas_1d: Optional[Tuple[float, float]] = None,\n            caution: bool = False\n    ):\n        if not 0.0 <= lr:\n            raise ValueError(\"Invalid learning rate: {}\".format(lr))\n        if not 0.0 <= eps:\n            raise ValueError(\"Invalid epsilon value: {}\".format(eps))\n        if not 0.0 <= betas[0] < 1.0:\n            raise ValueError(\"Invalid beta parameter at index 0: {}\".format(betas[0]))\n        if not 0.0 <= betas[1] < 1.0:\n            raise ValueError(\"Invalid beta parameter at index 1: {}\".format(betas[1]))\n        assert mars_type in [\"adamw\", \"lion\"], \"MARS type not supported\"\n\n        defaults = dict(\n            lr=lr,\n            betas=betas,\n            eps=eps,\n            weight_decay=weight_decay,\n            mars_type=mars_type,\n            gamma=gamma,\n            optimize_1d=optimize_1d,\n            lr_1d_factor=lr_1d_factor,\n            betas_1d=betas_1d or betas,","sourceCodeStart":95,"sourceCodeEnd":131,"githubUrl":"https://github.com/huggingface/pytorch-image-models/blob/9a5261e31b3b5128526eb2658333b4c0a54464ae/timm/optim/mars.py#L95-L131","documentation":"Mars constructor validates that lr is >= 0 is not enough phrasing-wise — it requires lr to satisfy 0.0 <= lr; any lr that fails that (i.e. negative) is rejected before the optimizer is built.","triggerScenarios":"Passing a negative lr (lr=-0.1) to mars.Mars(). Zero lr passes this check.","commonSituations":"Sign typos in configs; sweeps with negative lr values; misconfigured LR schedulers feeding the constructor.","solutions":["Pass a non-negative lr (e.g. 1e-3); note 0 is accepted here, unlike some other timm optimizers","Fix the config/sweep that produced the negative value"],"exampleFix":"# before\nopt = Mars(model.parameters(), lr=-1e-3)\n# after\nopt = Mars(model.parameters(), lr=1e-3)","handlingStrategy":"validation","validationCode":"assert cfg.lr >= 0, 'lr must be non-negative for Mars'","typeGuard":"def is_valid_mars_lr(lr: float) -> bool:\n    return isinstance(lr, (int, float)) and lr >= 0","tryCatchPattern":null,"preventionTips":["Validate lr sign at config load","Note Mars accepts lr=0 but rejects negatives"],"tags":["optimizer","mars","learning-rate","hyperparameter-validation"],"backgroundTag":"hyperparameter-out-of-range","analyzedSha":"9a5261e31b3b5128526eb2658333b4c0a54464ae","analyzedAt":"2026-08-27T02:34:25.417Z","schemaVersion":2},"datasetVersion":"2026-08-27T03:17:27.898Z"}