Function Differences with torch.nn.Sigmoid

torch.nn.Sigmoid

class torch.nn.Sigmoid()(input) -> Tensor

For more information, see torch.nn.Sigmoid.

mindspore.nn.Sigmoid

class mindspore.nn.Sigmoid()(input_x) -> Tensor

For more information, see mindspore.nn.Sigmoid.

Differences

PyTorch: Compute Sigmoid activation function element-wise, which maps the input to between 0 and 1.

MindSpore: MindSpore API implements the same functionality as PyTorch, and only the input parameter names after instantiation are different.

Categories	Subcategories	PyTorch	MindSpore	Difference
Input	Single input	input	input_x	Same function, different parameter names

Code Example

The two APIs achieve the same function and have the same usage.

# PyTorch
import torch
from torch import tensor

input_x = tensor([-1, -2, 0, 2, 1], dtype=torch.float32)
sigmoid = torch.nn.Sigmoid()
output = sigmoid(input_x).numpy()
print(output)
# [0.26894143 0.11920292 0.5        0.880797   0.7310586 ]

# MindSpore
import mindspore
from mindspore import Tensor

input_x = Tensor([-1, -2, 0, 2, 1], mindspore.float32)
sigmoid = mindspore.nn.Sigmoid()
output = sigmoid(input_x)
print(output)
# [0.26894143 0.11920292 0.5        0.8807971  0.7310586 ]