paddle完成一个简单线性回归最后输出一个散点图

Paddle（PyTorch的开源版本）是一个强大的深度学习框架，用于实现机器学习模型。对于简单的线性回归任务，你可以按照以下步骤操作：

1. 导入必要的库：

import paddle
import numpy as np
from sklearn import datasets
import matplotlib.pyplot as plt

2. 加载数据集并预处理：

boston_dataset = datasets.load_boston()
X_train, y_train = boston_dataset.data, boston_dataset.target

3. 创建模型，这里是一个基础的线性回归模型：

class LinearRegression(paddle.nn.Layer):
    def __init__(self):
        super(LinearRegression, self).__init__()
        self.linear = paddle.nn.Linear(in_features=13, out_features=1)

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

4. 初始化模型并设置损失函数和优化器：

model = LinearRegression()
criterion = paddle.nn.MSELoss()  # Mean Squared Error Loss for regression
optimizer = paddle.optimizer.SGD(parameters=model.parameters(), learning_rate=0.01)

5. 训练模型：

num_epochs = 100
for epoch in range(num_epochs):
    pred = model(X_train)  # Predictions
    loss = criterion(pred, y_train.reshape(-1, 1))  # Compute loss
    loss.backward()  # Backpropagation
    optimizer.step()
    optimizer.zero_grad()

    if (epoch+1) % 10 == 0:
        print(f"Epoch {epoch+1}, Loss: {loss.item():.4f}")

6. 完成训练后，可以预测新数据并绘制散点图展示结果：

# Generate predictions for the whole dataset and plot them
y_pred = model(X_train).numpy()
plt.scatter(X_train, y_train, color='blue')
plt.plot(X_train, y_pred, color='red', linewidth=2)
plt.xlabel('Feature')
plt.ylabel('Target')
plt.title('Boston Housing Linear Regression Result')
plt.show()