pytorch之一元线性回归

直接上代码：

# torch realize one-dimensional linear regression
import numpy as np
import torch
from torch import nn, optim
from torch.autograd import Variable
import matplotlib.pyplot as plt

x_train = np.array([[3.3], [4.4], [5.5], [7.5], [4.16]], dtype=np.float32)
y_train = np.array([[3.8], [4.7], [5.6], [7.4], [4.15]], dtype=np.float32)

# array transform into tensor
x_train = torch.from_numpy(x_train)
y_train = torch.from_numpy(y_train)


# 定义torch简单模型
class Linearregession(nn.Module):
    def __init__(self):
        super(Linearregession, self).__init__()
        self.linear = nn.Linear(1,1)  # input and output is 1 dimension

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


# use gpu or not
if torch.cuda.is_available():
    model = Linearregession().cuda()
else:
    model = Linearregession()

criterion = nn.MSELoss()  # 均方误差作为loss
optimizer = optim.SGD(model.parameters(),lr=1e-3)  # 梯度下降


# 开始训练
num_epochs=1000
for epoch in range(num_epochs):
    if torch.cuda.is_available():
        inputs = Variable(x_train).cuda()
        target = Variable(y_train).cuda()
    else:
        inputs = Variable(x_train)
        target = Variable(y_train)
    # forward
    out = model(inputs)  # 前向传播
    loss = criterion(out, target)  # 计算loss
    # backward
    optimizer.zero_grad()  # 梯度归零
    loss.backward()  # 反向传播
    optimizer.step()  # 更新参数
    if (epoch+1) % 20 == 0:
        print('Epoch[{}/{}], loss: {:.6f}'.format(epoch+1, num_epochs, loss.data[0]))

model.eval()# 将模型变成测试模式
predict = model(Variable(x_train.cuda()))
predict = predict.data.cpu().numpy()
# cpu版本torch代码：predict = model(Variable(x_train))
# cpu版本torch代码：predict = predict.data.numpy()
plt.plot(x_train.numpy(), y_train.numpy(), 'ro', label='Original data')
plt.plot(x_train.numpy(), predict, label='Fitting Line')
# 显示图例
plt.legend()
plt.show()

# 保存模型
# torch.save(model.state_dict(), './linear.pth')

注：
1.此代码为gpu环境下，若为cpu环境测试，请将其中已备注的两句cpu代码替换掉原代码
2.如果出现torch.from_numpy函数找不到的情况，（即代码显示常亮），笔者通过pycharm重启解决问题，也有其他作者提供了答案笔者不知是否有效，可参考解决方案