Function Differences with torch.prod

The following mapping relationships can be found in this file.

torch.prod

torch.prod(input, dim, keepdim=False, *, dtype=None) -> Tensor

For more information, see torch.prod.

mindspore.ops.prod

mindspore.ops.prod(input, axis=(), keep_dims=False) -> Tensor

For more information, see [mindspore.ops.prod](https://mindspore.cn/docs/en/r2.0/api_python/ops/mindspore.ops.prod.html.

Differences

PyTorch: Find the product on elements in input based on the specified dim. keepdim controls whether the output and input have the same dimension. dtype sets the data type of the output Tensor.

MindSpore: Find the product on the elements in input by the specified axis. The function of keep_dims is the same as PyTorch. MindSpore does not have a dtype parameter. MindSpore has a default value for axis, which is the product of all elements of input if axis is the default value.

Code Example

# PyTorch
import torch

input = torch.tensor([[1, 2.5, 3, 1], [2.5, 3, 2, 1]], dtype=torch.float32)
print(torch.prod(input, dim=1, keepdim=True))
# tensor([[ 7.5000],
#         [15.0000]])
print(torch.prod(input, dim=1, keepdim=True, dtype=torch.int32))
# tensor([[ 6],
#         [12]], dtype=torch.int32)

# MindSpore
import mindspore

x = mindspore.Tensor([[1, 2.5, 3, 1], [2.5, 3, 2, 1]], dtype=mindspore.float32)
print(mindspore.ops.prod(x, axis=1, keep_dims=True))
# [[ 7.5]
#  [15. ]]