{"record":{"id":"8350fd6baf53e93c","repo":"huggingface/pytorch-image-models","slug":"invalid-learning-rate-lr-8350fd","errorCode":null,"errorMessage":"Invalid learning rate: {lr}","messagePattern":"Invalid learning rate: (.+?)","errorType":"validation","errorClass":"ValueError","httpStatus":null,"severity":"error","filePath":"timm/optim/nadamw.py","lineNumber":56,"sourceCode":"        caution: enable caution\n        corrected_weight_decay: apply corrected weight decay (lr**2 / max_lr)\n    \"\"\"\n\n    def __init__(\n            self,\n            params: ParamsT,\n            lr: float = 1e-3,\n            betas: Tuple[float, float] = (0.9, 0.999),\n            eps: float = 1e-8,\n            weight_decay: float = 1e-2,\n            caution: bool = False,\n            corrected_weight_decay: bool = False,\n            maximize: bool = False,\n            foreach: Optional[bool] = None,\n            capturable: bool = False,\n    ):\n        if not 0.0 <= lr:\n            raise ValueError(f'Invalid learning rate: {lr}')\n        if not 0.0 <= eps:\n            raise ValueError(f'Invalid epsilon value: {eps}')\n        if not 0.0 <= betas[0] < 1.0:\n            raise ValueError(f'Invalid beta parameter at index 0: {betas[0]}')\n        if not 0.0 <= betas[1] < 1.0:\n            raise ValueError(f'Invalid beta parameter at index 1: {betas[1]}')\n        if not 0.0 <= weight_decay:\n            raise ValueError(f'Invalid weight_decay value: {weight_decay}')\n        defaults = dict(\n            lr=lr,\n            betas=betas,\n            eps=eps,\n            weight_decay=weight_decay,\n            caution=caution,\n            corrected_weight_decay=corrected_weight_decay,\n            foreach=foreach,\n            maximize=maximize,\n            capturable=capturable,","sourceCodeStart":38,"sourceCodeEnd":74,"githubUrl":"https://github.com/huggingface/pytorch-image-models/blob/9a5261e31b3b5128526eb2658333b4c0a54464ae/timm/optim/nadamw.py#L38-L74","documentation":"NAdamW optimizer constructor validation: learning rate must satisfy 0.0 <= lr. Negative values are rejected when timm.optim.NAdamW is created.","triggerScenarios":"Calling timm.optim.NAdamW(params, lr=-0.001) or with lr computed negative/NaN from hyperparameter search.","commonSituations":"Sweep scripts producing negative lr, config typos, or lr read from an env var with a leading dash.","solutions":["Set lr to a valid non-negative value","Validate hyperparameters before optimizer creation when they come from search/sweeps"],"exampleFix":"# before\nopt = NAdamW(model.parameters(), lr=-1e-4)\n\n# after\nopt = NAdamW(model.parameters(), lr=1e-4)","handlingStrategy":"validation","validationCode":"assert lr >= 0.0, f'lr must be >= 0, got {lr}'","typeGuard":null,"tryCatchPattern":null,"preventionTips":["Validate sweep outputs before constructing optimizers","Assert on signs of parsed CLI floats"],"tags":["timm","nadamw","learning-rate","valueerror"],"backgroundTag":"invalid-optimizer-hyperparameter","analyzedSha":"9a5261e31b3b5128526eb2658333b4c0a54464ae","analyzedAt":"2026-08-27T02:34:25.417Z","schemaVersion":2},"datasetVersion":"2026-08-27T03:17:27.898Z"}