Pytorch入门：Tensor加减乘除矩阵运算

Tensor加法：

a = torch.tensor([1, 2, 3])
b = torch.tensor([4, 5, 6])

c = a + b
print(c)    # tensor([5, 7, 9])
c = torch.add(a, b)
print(c)    # tensor([5, 7, 9])
c = a.add(b)
print(c)    # tensor([5, 7, 9])

Tensor减法：

a = torch.tensor([1, 2, 3])
b = torch.tensor([4, 5, 6])

c = a - b
print(c)    # tensor([-3, -3, -3])
c = torch.sub(a, b)
print(c)    # tensor([-3, -3, -3])
c = a.sub(b)
print(c)    # tensor([-3, -3, -3])

Tensor乘法：

a = torch.tensor([1, 2, 3])
b = torch.tensor([4, 5, 6])

c = a * b
print(c)    # tensor([ 4, 10, 18])
c = torch.mul(a, b)
print(c)    # tensor([ 4, 10, 18])
c = a.mul(b)
print(c)    # tensor([ 4, 10, 18])

Tensor除法：

a = torch.tensor([1, 2, 3])
b = torch.tensor([4, 5, 6])

c = a / b
print(c)    # tensor([0.2500, 0.4000, 0.5000])
c = torch.div(a, b)
print(c)    # tensor([0.2500, 0.4000, 0.5000])
c = a.div(b)
print(c)    # tensor([0.2500, 0.4000, 0.5000])

Tensor矩阵乘法：

a = torch.tensor([[1, 2, 3], [4, 5, 6]])    # size: 2, 3
b = torch.full([3, 4], 2)                   # size: 3, 4

c = torch.matmul(a, b)
print(c)    # tensor([[12, 12, 12, 12], [30, 30, 30, 30]]) size:2, 4

c = a.matmul(b)
print(c)    # tensor([[12, 12, 12, 12], [30, 30, 30, 30]]) size:2, 4

若张量维数大于2，则对最后两维进行matmul。进行此运算的要求是张量a与b除最后两维外的其他维必须一致：

a = torch.full([3, 2, 4, 3], 5)
b = torch.full([3, 2, 3, 6], 2)

c = torch.matmul(a, b)
print(c.size())        # size: 3, 2, 4, 6

c = a.matmul(b)
print(c.size())        # size: 3, 2, 4, 6