Pytorch深度学习之线性回归入门案例

#导入相应包
import torch
import torch.nn as nn
import numpy as np
import matplotlib.pyplot as plt
#设置输入维数、输出维数、训练次数、学习率
input_size = 1
output_size = 1
num_epochs = 60
learning_rate = 0.001
#定义训练集数据
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)
#定义线性模型
model = nn.Linear(input_size, output_size)
#定义损失函数和优化器
criterion = nn.MSELoss()
optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)
#进行训练
for epoch in range(num_epochs):
    inputs = torch.from_numpy(x_train)
    targets = torch.from_numpy(y_train)
    #前向传播
    outputs = model(inputs)
    #计算损失函数
    loss = criterion(outputs, targets)
	#优化器梯度置0，这点很重要
    optimizer.zero_grad()
    #通过损失函数反向传播
    loss.backward()
    #优化器梯度优化
    optimizer.step()
	#每五次输出训练损失
    if(epoch+1)%5 == 0:
        print('Epoch [{}/{}], Loss: {:.4f}'.format(epoch+1, num_epochs, loss.item()))
#可视化
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='Fitted line')
plt.legend()
plt.show()
#保存模型
torch.save(model.state_dict(), 'model.ckpt')

最后的结果如下：

Epoch [5/60], Loss: 24.6505
Epoch [10/60], Loss: 10.0899
Epoch [15/60], Loss: 4.1912
Epoch [20/60], Loss: 1.8015
Epoch [25/60], Loss: 0.8334
Epoch [30/60], Loss: 0.4412
Epoch [35/60], Loss: 0.2823
Epoch [40/60], Loss: 0.2179
Epoch [45/60], Loss: 0.1918
Epoch [50/60], Loss: 0.1812
Epoch [55/60], Loss: 0.1769
Epoch [60/60], Loss: 0.1752

绘制图像如下：