Function Differences with torch.logical_and

View Source On Gitee

torch.logical_and

class torch.logical_and(input, other, out=None)

For more information, see torch.logical_and.

mindspore.ops.LogicalAnd

class class mindspore.ops.LogicalAnd()(x, y)

For more information, see mindspore.ops.LogicalAnd.

Differences

PyTorch: Computes the element-wise logical AND of the given input tensors. Zeros are treated as False and nonzeros are treated as True.

MindSpore: Computes the “logical AND” of two tensors element-wise. The input should be a bool or a tensor whose data type is bool.

Code Example

import numpy as np
import torch
import mindspore.ops as ops
from mindspore import Tensor, Parameter
from mindspore import dtype as mstype

# MindSpore
x = Tensor(np.array([True, False, True]), mstype.bool_)
y = Tensor(np.array([True, True, False]), mstype.bool_)
logical_and = ops.LogicalAnd()
print(logical_and(x, y))
# [ True False False]
x = Tensor(np.array([True, False, True]), mstype.bool_)
y = Tensor(np.array([True, True, False]), mstype.int32)
logical_and = ops.LogicalAnd()
print(logical_and(x, y))
# TypeError: For primitive[LogicalAnd], the input argument[x, y, ] must be a type of {Tensor[Bool],}, but got Int32.

# Pytorch
print(torch.logical_and(torch.tensor([True, False, True]), torch.tensor([True, False, False])))
# tensor([ True, False, False])
a = torch.tensor([0, 1, 10, 0], dtype=torch.int8)
b = torch.tensor([4, 0, 1, 0], dtype=torch.int8)
print(torch.logical_and(a, b))
# tensor([False, False,  True, False])
print(torch.logical_and(a.double(), b.double()))
# tensor([False, False,  True, False])
print(torch.logical_and(a.double(), b))
# tensor([False, False,  True, False])
print(torch.logical_and(a, b, out=torch.empty(4, dtype=torch.bool)))
# tensor([False, False,  True, False])