lucid.tensor¶

lucid.tensor(data: Tensor | int | float | complex | list[int | float | complex] | ndarray | array, requires_grad: bool = False, keep_grad: bool = False, dtype: type[bool | int | float | complex] | Numeric | None = None, device: Literal['cpu', 'gpu'] = 'cpu') → Tensor¶

Creates a new Tensor object with specified data, requires_grad setting, and data type. This function is similar to PyTorch’s torch.tensor, allowing you to initialize a Tensor either from another Tensor or directly from array-like data.

Function Signature¶

def tensor(
    data: Tensor | _ArrayOrScalar,
    requires_grad: bool = False,
    dtype: Any = np.float32
) -> Tensor

Parameters¶

data (Tensor | _ArrayOrScalar): The input data to be wrapped in a Tensor. If a Tensor is provided, its underlying data is used.
requires_grad (bool, optional): If True, the resulting Tensor will track gradients for backpropagation. Defaults to False.
dtype (Any, optional): The desired data type for the new Tensor. Defaults to np.float32.

Returns¶

Tensor: A new Tensor instance wrapping the input data, with the specified requires_grad setting and data type.

Examples¶

Create a Tensor from a list

>>> import lucid
>>> t = lucid.tensor([1, 2, 3], requires_grad=True)
>>> print(t)
Tensor([1, 2, 3], requires_grad=True)

Use another Tensor as input

>>> import lucid
>>> t1 = lucid.tensor([4.0, 5.0, 6.0])
>>> t2 = lucid.tensor(t1, dtype=np.float64)
>>> print(t2)
Tensor([4.0, 5.0, 6.0], dtype=float64)