1. 内容来源
2. 底层实现
2.1 数据集获取
本文依旧使用Fashion Mnist,数据下载源码详见本系列博客Softmax分类器一文
代码实现:
import torch
from torch import nn
from d2l import torch as d2l
batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size)
2.2 模型参数初始化
MLP模型与线性回归以及Softmax分类不同在于使用多余1层的网络架构,中间层被称作隐藏层(Hidden Layer),并加入了非线性激活函数将网络转化成非线性网络(否则N层线性层可等价成单层线性层)
设置中间隐藏层神经元个数为256,则数据维数变化为784(输入)->256(隐藏层)->10(输出)
则参数初始化代码实现:
# Prameters
num_inputs, num_outputs, num_hiddens = 784, 10, 256
W1 = nn.Parameter(
torch.randn(num_inputs, num_hiddens, requires_grad=True)
)
b1 = nn.Parameter(
torch.zeros(num_hiddens, requires_grad=True))
W2 = nn.Parameter(
torch.randn(num_hiddens, num_outputs, requires_grad=True)
)
b2 = nn.Parameter(
torch.zeros(num_outputs, requires_grad=True))
params = [W1, b1, W2, b2]
2.3 模型构建
激活函数使用ReLU函数
代码实现:
def relu(X):
a = torch.zeros_like(X)
return torch.max(X, a)
第一层输出公式表示:
其中,分别为输入、第一层(输入层)的权重、偏置
第二层输出公式表示:
其中,分别为第二层(隐藏层)的权重和偏置
则代码实现:
def net(X):
X = X.reshape((-1, num_inputs))
H = relu(X @ W1 + b1)
return (H @ W2 + b2)
Loss函数选用交叉熵(原始实现代码见本系列博客Softmax分类器一文)
代码实现:
loss = nn.CrossEntropyLoss()
2.4 模型训练
# Training
num_epochs, lr = 10, 0.01
updater = torch.optim.SGD(params, lr)
d2l.train_ch3(net, train_iter, test_iter, loss, num_epochs, updater)
结果:
其中Test Acc处于Train Acc下方,说明模型有一定的过拟合
3. 简洁实现
3.1 模型构建
输入层和隐藏层使用torch.nn.Linear()表示
代码实现:
import torch
from torch import nn
from d2l import torch as d2l
# Model
net = nn.Sequential(
nn.Flatten(),
nn.Linear(784, 256),
nn.ReLU(),
nn.Linear(256, 10)
)
def init_weights(m):
if type(m) == nn.Linear:
nn.init.normal_(m.weight, std=0.1)
net.apply(init_weights)
模型展示:
Sequential(
(0): Flatten(start_dim=1, end_dim=-1)
(1): Linear(in_features=784, out_features=256, bias=True)
(2): ReLU()
(3): Linear(in_features=256, out_features=10, bias=True)
)
3.2 模型训练
# Training
batch_size, lr, num_epochs = 256, 0.1, 10
loss = nn.CrossEntropyLoss()
trainer = torch.optim.SGD(net.parameters(), lr=lr)
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size)
d2l.train_ch3(net, train_iter, test_iter, loss, num_epochs, trainer)
结果: