函数:class torch.nn.Linear(in_features,out_features,bias = True)
源码:
从init函数中可以看出Linear中包含四个属性:
1)in_features: 上层神经元个数【每个输入样本的大小】
2)out_features: 本层神经元个数【每个输出样本的大小】
3)weight:权重,形状[out_features , in_features]
4)bias: 偏置,形状[out_features]。网络层是否有偏置,默认存在,且维度为[out_features ];若bias=False,则该网络层无偏置,图层不会学习附加偏差。
示例说明:
import torch
x = torch.randn(128, 20) # 输入的维度是(128,20)
m = torch.nn.Linear(20, 30) # 20,30是指维度
output = m(x)
print('m.weight.shape:\n ', m.weight.shape)
print('m.bias.shape:\n', m.bias.shape)
print('output.shape:\n', output.shape)
# ans = torch.mm(input,torch.t(m.weight))+m.bias 等价于下面
ans = torch.mm(x, m.weight.t()) + m.bias
print('ans.shape:\n', ans.shape)
print(torch.equal(ans, output))
输出:
m.weight.shape:
torch.Size([30, 20])
m.bias.shape:
torch.Size([30])
output.shape:
torch.Size([128, 30])
ans.shape:
torch.Size([128, 30])
True
output.size()=矩阵size(128,20)*矩阵size(20,30)=(128,30)。