PyTorch示例——LinearModel

PyTorch示例——LinearModel

版本信息

• PyTorch: 1.12.1
• Python: 3.7.13

导包

import torch
import matplotlib.pyplot as plt

原始数据

X_data = torch.Tensor([[1.0, 2.9], [2.0, 6.1], [3.0, 9.2], [4.0, 12.3], [5.0, 14.9], [6.0, 18.1]])
y_data = torch.Tensor([[0.4], [1.4], [2.8], [3.5], [4.8], [5.5]])

构建模型

class MyLinearModel(torch.nn.Module):
    
    def __init__(self):
        super(MyLinearModel, self).__init__()
        # 缩写形式torch.nn.Linear(2, 1)
        self.linear = torch.nn.Linear(in_features=2, out_features=1)
        
    def forward(self, x):
        output = self.linear(x)
        return output

开始训练

• 代码

# 参数
learning_rate = 0.005
epoch_num = 20

# 构建模型
my_model = MyLinearModel()
# 用MSE计算损失
loss = torch.nn.MSELoss()
# 优化器，即学习器，使用SGD，会更新模型参数my_model.parameters()
optimizer = torch.optim.SGD(my_model.parameters(), lr=learning_rate)

# 开始训练
loss_list = []
for epoch in range(epoch_num):
    y_pred = my_model(X_data)
    l = loss(y_pred, y_data)
    
    optimizer.zero_grad() # 梯度置零
    l.backward() # 反向传播梯度
    optimizer.step() # 更新梯度
    
    print(f"Train... ===> epoch = {epoch}, loss = {l.item()}")
    loss_list.append(l.item())

• 输出信息

Train... ===> epoch = 0, loss = 29.548913955688477
Train... ===> epoch = 1, loss = 8.797457695007324
Train... ===> epoch = 2, loss = 2.6332108974456787
Train... ===> epoch = 3, loss = 0.8021142482757568
Train... ===> epoch = 4, loss = 0.2581847608089447
Train... ===> epoch = 5, loss = 0.09661001712083817
Train... ===> epoch = 6, loss = 0.048613857477903366
Train... ===> epoch = 7, loss = 0.034356359392404556
Train... ===> epoch = 8, loss = 0.03012099117040634
Train... ===> epoch = 9, loss = 0.028862761333584785
Train... ===> epoch = 10, loss = 0.02848881296813488
Train... ===> epoch = 11, loss = 0.02837762050330639
Train... ===> epoch = 12, loss = 0.028344422578811646
Train... ===> epoch = 13, loss = 0.028334492817521095
Train... ===> epoch = 14, loss = 0.028331367298960686
Train... ===> epoch = 15, loss = 0.028330326080322266
Train... ===> epoch = 16, loss = 0.02832990325987339
Train... ===> epoch = 17, loss = 0.028329646214842796
Train... ===> epoch = 18, loss = 0.0283293928951025
Train... ===> epoch = 19, loss = 0.028329243883490562

绘制曲线：epoch与loss

plt.plot(epochs, loss_list)
plt.xlabel("epoch")
plt.ylabel("loss_val")
plt.show()

查看权重、偏置信息

• 代码

print(f"w = {my_model.linear.weight}")
print(f"b = {my_model.linear.bias}")

• 输出信息

w = Parameter containing:
tensor([[-0.0156,  0.3502]], requires_grad=True)
b = Parameter containing:
tensor([-0.5856], requires_grad=True)

利用模型做预测

• 代码

X_test = torch.Tensor([[4.0, 12.0]])
y_pred = my_model(X_test)
print(f"y_pred = {y_pred.data}")

• 输出信息

y_pred = tensor([[3.5542]])