比较与torch.nn.Linear的功能差异

torch.nn.Linear

torch.nn.Linear(
    in_features,
    out_features,
    bias=True
)

更多内容详见torch.nn.Linear。

mindspore.nn.Dense

class mindspore.nn.Dense(
    in_channels,
    out_channels,
    weight_init='normal',
    bias_init='zeros',
    has_bias=True,
    activation=None
)(input)

更多内容详见mindspore.nn.Dense。

使用方式

Pytorch：对传入数据应用线性变换，默认权重矩阵和偏移矩阵都由均匀分布初始化。

MindSpore：对传入数据应用线性变换，在输出数据之前可以选择应用激活函数activation，默认权重矩阵由标准正态分布初始化，偏移矩阵初始化为0。

代码示例

import mindspore as ms
from mindspore import nn
import torch
import numpy as np

# In MindSpore, default weight will be initialized through standard normal distribution.
# Default bias will be initialized by zero.
# Default none activation used.
input_net = ms.Tensor(np.array([[180, 234, 154], [244, 48, 247]]), ms.float32)
net = nn.Dense(3, 4)
output = net(input_net)
print(output.shape)
# Out：
# (2, 4)

# In torch, default weight and bias will be initialized through uniform distribution.
# No parameter to set the activation.
input_net = torch.Tensor(np.array([[180, 234, 154], [244, 48, 247]]))
net = torch.nn.Linear(3, 4)
output = net(input_net)
print(output.shape)
# Out：
# torch.Size([2, 4])