Linear_regression

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


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

# Toy dataset         加载数据集
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)


# Linear regression model     定义模型
model = nn.Linear(input_size,output_size)


# Loss and optimizer       定义损失函数和优化器
critertion = nn.MSELoss()
optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)


# Train the model
for epoch in range(num_epochs):
    # Convert numpy arrays to torch tensors
    inputs = torch.from_numpy(x_train)
    targets = torch.from_numpy(y_train)


    # Farward pass
    outputs = model(inputs)
    loss = critertion(outputs, targets)

    # Backward and optimize
    optimizer.zero_grad()
    loss.backward()
    optimizer.step()


    if(epoch+1) % 5 == 0: #  每五轮输出损失函数值
        print('Epoch [{} / {}], Loss :{:.4f}'.format(epoch+1, num_epochs, loss.item()))


# Plot the graph
predicted = model(torch.from_numpy(x_train)).detach().numpy()  # detach就是截断反向传播的梯度流
plt.plot(x_train, y_train,'ro', label='Original data')  # r o表示红色圈
plt.plot(x_train,predicted,label='Fitted line')
plt.legend()  # plt.legend（）函数主要的作用就是给图加上图例，plt.legend([x,y,z])里面的参数使用的是list的的形式将图表的的名称喂给这和函数
plt.show()

# Save the model checkpoint
torch.save(model.state_dict(), 'model.ckpt') # model.ckpt是四个文件