pytorch:一维线性回归(一)

╰半夏微凉° 2022-02-01 09:53 329阅读 0赞

# -- coding:utf-8 --
    import torch
    import numpy as np
    from torch import nn, optim
    import matplotlib.pyplot as plt
    from torch.autograd import Variable
    # 设置字体为中文
    plt.rcParams['font.sans-serif'] = ['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    # 训练数据
    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)
    # 训练数据可视化
    plt.scatter(x_train, y_train, marker='o', c='red')
    plt.title(label='一维线性回归-原始数据分布展示')
    plt.show()
    # 把训练数据转换为张量numpy->torch
    x_train = torch.from_numpy(x_train)
    y_train = torch.from_numpy(y_train)
    
    # 构建线性模型
    class LinearRegression(nn.Module):
        def __init__(self):
            super(LinearRegression, self).__init__()
            # 输入和输出的维度都是1
            self.linear = nn.Linear(1, 1)
    
        def forward(self, x):
            out = self.linear(x)
            return out
    
    
    if torch.cuda.is_available():
        model = LinearRegression().cuda()
    else:
        model = LinearRegression()
    # 损失函数：均方误差
    criterion = nn.MSELoss()
    # 采用随机梯度下降
    optimizer = optim.SGD(model.parameters(), lr=1e-3)
    
    num_epochs = 1000  # 训练总次数
    ctn = []  # 训练次数
    lo = []  # 损失值
    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)
    
        # 向前传播
        out = model(inputs)
        loss = criterion(out, target)
    
        # 向后传播
        optimizer.zero_grad()  # 梯度清零
        loss.backward() # 向后传播
        optimizer.step()  # 更新参数
        # 统计每次对应的损失值
        ctn.append(epoch+1)
        lo.append(loss.item())
    
        if (epoch + 1) % 20 == 0:
            print('Epoch[{}/{}], loss:{:.6f}'.format(epoch + 1, num_epochs, loss.item()))
    
    model.eval()
    if torch.cuda.is_available():
        predict = model(Variable(x_train).cuda())
        predict = predict.data.cpu().numpy()
    else:
        predict = model(Variable(x_train))
        predict = predict.data.numpy()
    
    plt.plot(x_train.numpy(), y_train.numpy(),'ro',label='Original Data')
    plt.plot(x_train.numpy(), predict, label='拟合的直线')
    plt.title(label='拟合的直线')
    plt.show()
    # 绘制训练次数与损失值之间的关系
    plt.plot(ctn,lo)
    plt.title(label='训练次数与损失')
    plt.xlabel('训练次数')
    plt.ylabel('损失值')
    plt.show()

实验结果：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3Nvbmd4aWFvbGluZ2Jhb2Jhbw_size_16_color_FFFFFF_t_70][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3Nvbmd4aWFvbGluZ2Jhb2Jhbw_size_16_color_FFFFFF_t_70 1][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3Nvbmd4aWFvbGluZ2Jhb2Jhbw_size_16_color_FFFFFF_t_70 2][]

[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3Nvbmd4aWFvbGluZ2Jhb2Jhbw_size_16_color_FFFFFF_t_70]: /images/20220201/722672cf19614eb2a4e78d7e20eaa764.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3Nvbmd4aWFvbGluZ2Jhb2Jhbw_size_16_color_FFFFFF_t_70 1]: /images/20220201/1889695076d140eab519e2b4f5a36f6a.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3Nvbmd4aWFvbGluZ2Jhb2Jhbw_size_16_color_FFFFFF_t_70 2]: /images/20220201/0cb4f0799c694e988b3a6f903686393f.png