Pytorch---使用Pytorch进行线性回归的分解写法

一、代码中的数据集可以点击以下链接进行下载

百度网盘提取码：lala

二、代码运行环境

Pytorch-gpu==1.7.1
Python=3.7

三、代码的构建如下所示

import torch
import pandas as pd
import numpy as np
import tqdm
import matplotlib.pyplot as plt

# 读取数据
data = pd.read_csv(r'dataset/dataset.csv')
X = torch.from_numpy(data.Education.values.reshape(-1, 1).astype(np.float32))
Y = torch.from_numpy(data.Income.values.reshape(-1, 1).astype(np.float32))

# 参数的初始化
w = torch.randn(1, requires_grad=True)
b = torch.zeros(1, requires_grad=True)
learning_rate = 0.0001

tqdm_train = tqdm.tqdm(range(5000))

# 开始进行迭代
for epoch in tqdm_train:
    for x, y in zip(X, Y):
        y_pred = torch.matmul(x, w) + b
        loss = (y - y_pred).pow(2).mean()
        # 清空梯度
        if w.grad is not None:
            w.grad.data.zero_()
        if b.grad is not None:
            b.grad.data.zero_()
        # 进行自动微分运算
        loss.backward()
        # 进行参数优化
        with torch.no_grad():
            w.data -= w.grad.data * learning_rate
            b.data -= b.grad.data * learning_rate

# 训练结果的简单展示
plt.scatter(data.Education, data.Income)
plt.plot(X.numpy(), (X * w + b).data.numpy(), c='r')
plt.show()

四、运行结果如下所示