iron.tensor Namespace Reference

Classes
class	Tensor

Functions
	tensor (data, dtype=np.float32, device="npu")

Variables
	ones = Tensor.ones
	zeros = Tensor.zeros
	randint = Tensor.randint
	rand = Tensor.rand
	arange = Tensor.arange
	zeros_like = Tensor.zeros_like

Function Documentation

tensor()

iron.tensor.tensor	(		data,
			dtype = np.float32,
			device = "npu"
	)

Creates a Tensor from array-like input with the specified dtype and device.

Parameters:
    data (array-like): Input data (list, tuple, or  NumPy array.).
    dtype (np.dtype, optional): Desired data type. Defaults to np.float32.
    device (str, optional): Target device (e.g., "npu", "cpu"). Defaults to "npu".

Returns:
    Tensor: A new Tensor instance.

Variable Documentation

arange

iron.tensor.arange = Tensor.arange

ones

iron.tensor.ones = Tensor.ones

rand

iron.tensor.rand = Tensor.rand

randint

iron.tensor.randint = Tensor.randint

zeros

iron.tensor.zeros = Tensor.zeros

zeros_like

iron.tensor.zeros_like = Tensor.zeros_like