Pytorch学习（1）一维线性回归实现

掌握要点：

1、tensor

2、Variable

3、Dataset

4、nn.Module

5、torch.optim

6、模型的保存和加载

import torch
from torch import nn
from torch.autograd import Variable
from torch import optim
import numpy as np
import matplotlib.pyplot as plt

#线性模型
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)

x_train = torch.from_numpy(x_train)
y_train = torch.from_numpy(y_train)

# 建立模型
class LinearRegression(nn.Module):
    def __init__(self):
        super(LinearRegression, self).__init__()
        self.linear = nn.Linear(1, 1) # input and output is 1 dimension
        
    def forward(self, x):
        out = self.linear(x)
        return out

# define model
model = LinearRegression()
    
# define Loss and optimizer
criterion = nn.MSELoss()
optimizer = optim.SGD(model.parameters(), lr = 1e-3)

# train model
num_epochs = 10000
for epoch in range(num_epochs):
    inputs = Variable(x_train)
    target = Variable(y_train)
        
    #forward
    out = model(inputs)
    loss = criterion(out, target)
    
    #bachward
    optimizer.zero_grad()
    loss.backward()
    optimizer.step()
    
    if (epoch+1) % 20 == 0:
        print('Epoch[{}/{}], loss: {:.6}'.format(epoch+1, 
              num_epochs, loss))
       
# model test
model.eval()
model.cpu()
predict = model(Variable(x_train))
predict = predict.data.numpy()

plt.figure()
plt.plot(x_train.numpy(), y_train.numpy(), 'ro', label = 'Original data')
plt.plot(x_train.numpy(), predict, label = 'Fitting Line')
plt.show()

# plot the loss curve    
plt.figure()
index = range(len(loss_all))
plt.plot(index, loss_all)
plt.show

问题探究：

（1）super(LinearRegression, self).__init__()

super作用

    1、如果子类(Puple)继承父类(Person)不做初始化，那么会自动继承父类(Person)属性name。
    2、如果子类(Puple_Init)继承父类(Person)做了初始化，且不调用super初始化父类构造函数，那么子类(Puple_Init)不会自动继承父类的属性(name)。
    3、如果子类(Puple_super)继承父类(Person)做了初始化，且调用了super初始化了父类的构造函数，那么子类(Puple_Super)也会继承父类的(name)属性。

 转自： https://blog.csdn.net/paopaohll/article/details/83063349