mindspore.hal.contiguous_tensors_handle.ContiguousTensorsHandle

class mindspore.hal.contiguous_tensors_handle.ContiguousTensorsHandle(tensor_list, enable_mem_align=True)

ContiguousTensorsHandle is a handle manage continuous memory.

Parameters

• tensor_list (list[Tensor], Tuple[Tensor]) – The tensor list to be stored.

• enable_mem_align (bool, optional) – Whether to enable the memory alignment function. False is not supported. Default: True .

Returns

ContiguousTensorsHandle, a manager with contiguous memory.

Examples

>>> import numpy as np
>>> import mindspore as ms
>>> from mindspore import Tensor
>>> from mindspore.hal.contiguous_tensors_handle import ContiguousTensorsHandle
>>> x = Tensor(np.array([1, 2, 3]).astype(np.float32))
>>> y = Tensor(np.array([4, 5, 6]).astype(np.float32))
>>> handle = ContiguousTensorsHandle([x, y], True)
>>> print(handle[0].shape)
[1]
>>> print(handle[1: 3].asnumpy())
[2, 3]

slice_by_tensor_index(start=None, end=None)

Return the tensor which is sliced by tensor index.

Parameters

• start (int, None) – Starting position. Default：None.

• end (int, None) – Deadline position. Default：None.

Returns

Tensor，is sliced by tensor index.

Raises

• TypeError – If start or end, is neither an 'int' nor a 'none'.

• ValueError – If values of start or end are negative, or out of the list range, or start >= end.

Examples

>>> import numpy as np
>>> import mindspore as ms
>>> from mindspore import Tensor
>>> from mindspore.hal.contiguous_tensors_handle import ContiguousTensorsHandle
>>> x = Tensor(np.array([1, 2, 3]).astype(np.float32))
>>> y = Tensor(np.array([4, 5, 6]).astype(np.float32))
>>> handle = ContiguousTensorsHandle([x, y], True)
>>> print(output.slice_by_tensor_index(0, 1).asnumpy())
[1, 2, 3]