lucid.full

lucid.full(shape: int | list[int] | tuple[int], fill_value: int | float | complex, dtype: type[bool | int | float | complex] | Numeric | None = None, requires_grad: bool = False, keep_grad: bool = False, device: Literal['cpu', 'gpu'] = 'cpu') → Tensor

The full function creates a Tensor of a given shape and fills it with a specified scalar value.

Function Signature

def full(
    shape: int | _ShapeLike,
    fill_value: _Scalar,
    dtype: _BuiltinNumeric | Numeric | None = None,
    requires_grad: bool = False,
    keep_grad: bool = False,
    device: _DeviceType = "cpu",
) -> Tensor

Parameters

  • shape (int | tuple[int] | list[int]): Desired shape of the output tensor.

  • fill_value (_Scalar): The constant value to fill the tensor with.

  • dtype (type | Numeric, optional): Desired data type for the tensor. Can be one of the Python built-in numeric types (e.g. float, int) or a Numeric type. Defaults to None.

  • requires_grad (bool, optional): If set to True, gradients will be tracked for this tensor. Defaults to False.

  • keep_grad (bool, optional): Whether to retain the gradient after a backward pass. Defaults to False.

  • device (Literal[“cpu”, “gpu”], optional): Device to create the tensor on. “cpu” uses NumPy backend, “gpu” uses MLX backend. Defaults to “cpu”.

Returns

  • Tensor: A tensor filled with fill_value, of specified shape and dtype.

Examples

>>> import lucid
>>> t = lucid.full((2, 3), fill_value=7)
>>> print(t)
Tensor([[7, 7, 7],
        [7, 7, 7]], grad=None)

>>> # With gradients enabled
>>> t = lucid.full((2, 2), 3.14, dtype=lucid.Float32, requires_grad=True)
>>> print(t.requires_grad)
True

Notes

Note

This function is functionally similar to lucid.ones and lucid.zeros, but allows specifying any fill value.

Warning

If dtype is unspecified and fill_value is a float or int, implicit casting rules apply. Use dtype explicitly for precise control.