B站刘二大人-线性回归 Pytorch

系列文章：

• • 线性回归 Pytorch

import matplotlib.pyplot as plt
import  torch
x_data = torch.Tensor([[1.0, 2.0, 3.0]])
y_data = torch.Tensor([[2.0, 4.0, 6.0]])

class LinearModel(torch.nn.Module):
    # 成员方法__init__ 及 forward 必须实现
    def __init__(self):
        super(LinearModel, self).__init__()
        # 1,1，代表权重weight和偏置b的size
        self.linear = torch.nn.Linear(3, 1)

    def forward(self, x):
        y_pred = self.linear(x)
        return  y_pred

model = LinearModel()
# 损失函数计算一个batch中所有loss的均值
criterion =  torch.nn.MSELoss(size_average=False)
optimizer = torch.optim.SGD(model.parameters(), lr=0.0005)
loss_list = []
epoch_list = []
for epoch in range(200):
    y_pred = model(x_data)
    loss = criterion(y_pred, y_data)
    print(epoch, loss.item())
    loss_list.append(loss.item())
    epoch_list.append(epoch)
    # 梯度会在backward时累计，所以在backward之前必须重置
    optimizer.zero_grad()
    loss.backward()
    # update
    optimizer.step()

plt.plot(epoch_list, loss_list)
plt.xlabel("epoch")
plt.ylabel("loss")
plt.show()