2 Pytorch实例——线性回归

import torch
import torch.nn as nn
import numpy as np
import matplotlib.pyplot as plt


# 设置超参数 Hyper-parameters
input_size = 1
output_size = 1
num_epochs = 60
learning_rate = 0.001

# 生成训练数据
x_train = np.array([[3.3], [4.4], [5.5], [6.71], [6.93], [4.168], 
                    [9.779], [6.182], [7.59], [2.167], [7.042], 
                    [10.791], [5.313], [7.997], [3.1]], dtype=np.float32)

y_train = np.array([[1.7], [2.76], [2.09], [3.19], [1.694], [1.573], 
                    [3.366], [2.596], [2.53], [1.221], [2.827], 
                    [3.465], [1.65], [2.904], [1.3]], dtype=np.float32)

# 线性回归模型
model = nn.Linear(input_size, output_size)

# 定义损失和优化器
criterion = nn.MSELoss()
optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate) 

# 训练模型
for epoch in range(num_epochs):
    # 转化numpy arrays 为 torch tensors
    inputs = torch.from_numpy(x_train)
    targets = torch.from_numpy(y_train)

    # 前向传递
    outputs = model(inputs)
    loss = criterion(outputs, targets)
    
    # 反向传播并且进行优化
    optimizer.zero_grad()  #将梯度初始化为零
    loss.backward()        #计算模型参数的导数
    optimizer.step()       #更新模型参数
    
    if (epoch+1) % 5 == 0:
        print ('Epoch [{}/{}], Loss: {:.4f}'.format(epoch+1, num_epochs, loss.item()))

# 画图
predicted = model(torch.from_numpy(x_train)).detach().numpy()
plt.plot(x_train, y_train, 'ro', label='original data')
plt.plot(x_train, predicted, label='predict')
plt.legend()
plt.show()

# 保存模型的checkpoint
torch.save(model.state_dict(), 'model.ckpt')

最终输出结果：

Epoch [5/60], Loss: 3.7651
Epoch [10/60], Loss: 1.7614
Epoch [15/60], Loss: 0.9493
Epoch [20/60], Loss: 0.6199
Epoch [25/60], Loss: 0.4862
Epoch [30/60], Loss: 0.4316
Epoch [35/60], Loss: 0.4092
Epoch [40/60], Loss: 0.3998
Epoch [45/60], Loss: 0.3956
Epoch [50/60], Loss: 0.3936
Epoch [55/60], Loss: 0.3924
Epoch [60/60], Loss: 0.3916