{"record":{"id":"76d34cc0345a2ff5","repo":"huggingface/pytorch-image-models","slug":"name-must-be-a-scalar-or-scalar-tensor","errorCode":null,"errorMessage":"{name} must be a scalar or scalar tensor.","messagePattern":"(.+?) must be a scalar or scalar tensor\\.","errorType":"validation","errorClass":"ValueError","httpStatus":null,"severity":"error","filePath":"timm/optim/_helpers.py","lineNumber":71,"sourceCode":"    return capturable_supported_devices\n\n\ndef _check_capturable_devices(\n        params: Sequence[Tensor],\n        state_steps: Sequence[Tensor],\n        supports_xla: bool = True,\n) -> None:\n    capturable_supported_devices = _get_capturable_supported_devices(supports_xla=supports_xla)\n    assert all(\n        p.device.type == step.device.type and p.device.type in capturable_supported_devices\n        for p, step in zip(params, state_steps)\n    ), f\"If capturable=True, params and state_steps must be on supported devices: {capturable_supported_devices}.\"\n\n\ndef _validate_scalar(name: str, value, min_value: float = 0.0, max_value: Optional[float] = None) -> None:\n    if torch.is_tensor(value):\n        if value.numel() != 1:\n            raise ValueError(f\"{name} must be a scalar or scalar tensor.\")\n        value_float = float(value.detach().cpu())\n    else:\n        value_float = float(value)\n    if value_float < min_value or (max_value is not None and value_float >= max_value):\n        raise ValueError(f\"Invalid {name}: {value}\")\n\n\ndef _add_scaled_(param: Tensor, update: Tensor, scale) -> None:\n    if torch.is_tensor(scale):\n        param.add_(update * scale)\n    else:\n        param.add_(update, alpha=scale)\n\n\ndef _addcdiv_scaled_(param: Tensor, tensor1: Tensor, tensor2: Tensor, scale) -> None:\n    if torch.is_tensor(scale):\n        param.add_(tensor1 / tensor2 * scale)\n    else:","sourceCodeStart":53,"sourceCodeEnd":89,"githubUrl":"https://github.com/huggingface/pytorch-image-models/blob/9a5261e31b3b5128526eb2658333b4c0a54464ae/timm/optim/_helpers.py#L53-L89","documentation":"timm optimizer helper _validate_scalar accepts a python number or a single-element tensor for hyper-parameters like lr/beta/eps. A tensor with more than one element cannot be interpreted as a scalar hyper-parameter.","triggerScenarios":"Passing a multi-element tensor as an optimizer hyper-parameter, e.g. AdamW(model.parameters(), lr=torch.tensor([1e-3, 1e-4])) or a per-group param tensor passed where a scalar is expected.","commonSituations":"Programmatically building hyper-parameters as tensors (e.g. slices of arrays) instead of floats; migrating code that accidentally passes a shape-(1,1) or vector tensor.","solutions":["Pass plain floats (lr=1e-3, betas=(0.9, 0.999))","If a tensor arrives from elsewhere, extract a scalar: float(t) or t.item() after asserting t.numel()==1"],"exampleFix":"# before\nopt = timm.optim.create_optimizer_v2(model, opt='adamw', lr=torch.tensor([1e-3]))\n# after\nopt = timm.optim.create_optimizer_v2(model, opt='adamw', lr=1e-3)","handlingStrategy":"type-guard","validationCode":"hp = float(hp_tensor.item()) if torch.is_tensor(hp_tensor) else float(hp)\nassert not torch.is_tensor(hp_tensor) or hp_tensor.numel() == 1","typeGuard":"def as_scalar_hp(v):\n    if torch.is_tensor(v):\n        assert v.numel() == 1, 'hyper-parameter must be scalar'\n        return v.item()\n    return float(v)","tryCatchPattern":"try:\n    opt = timm.optim.AdamW(params, lr=lr, betas=betas)\nexcept ValueError as e:\n    if 'must be a scalar' in str(e):\n        opt = timm.optim.AdamW(params, lr=float(lr.item()), betas=(float(betas[0].item()), float(betas[1].item())))\n    else:\n        raise","preventionTips":["Always pass floats for optimizer hyper-parameters","Sanitize tensor-derived config values with .item()"],"tags":["timm","optimizer","hyperparameter","type-error"],"backgroundTag":"invalid-argument-type","analyzedSha":"9a5261e31b3b5128526eb2658333b4c0a54464ae","analyzedAt":"2026-08-27T02:34:25.417Z","schemaVersion":2},"datasetVersion":"2026-08-27T03:17:27.898Z"}