pyProject/test_pytorch.py

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


print(torch.__version__)  # pytorch版本
print(torch.version.cuda)  # cuda版本
print(torch.cuda.is_available())  # 查看cuda是否可用

#
# 使用GPU or CPU
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")



# 生成一些随机的训练数据
np.random.seed(42)
x = np.random.rand(1000, 1)
y = 2 * x + 1 + 0.5 * np.random.randn(1000, 1)

# 将数据转换为张量
x_tensor = torch.from_numpy(x).float()
y_tensor = torch.from_numpy(y).float()

# 定义线性回归模型
class LinearRegressionModel(nn.Module):
    def __init__(self):
        super(LinearRegressionModel, self).__init__()
        self.linear = nn.Linear(1, 1)

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

# 创建模型实例
model = LinearRegressionModel()

# 定义损失函数和优化器
criterion = nn.MSELoss()
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)

# 训练模型
num_epochs = 1000
for epoch in range(num_epochs):
    # 前向传播
    outputs = model(x_tensor)
    loss = criterion(outputs, y_tensor)

    # 反向传播和优化
    optimizer.zero_grad()
    loss.backward()
    optimizer.step()

    if (epoch + 1) % 10 == 0:
        print(f'Epoch [{epoch + 1}/{num_epochs}], Loss: {loss.item()}')

# 进行预测
with torch.no_grad():
    predicted = model(x_tensor)

# 绘制原始数据和预测结果
plt.scatter(x, y, label='Original Data')
plt.plot(x, predicted.numpy(), color='red', label='Predicted Line')
plt.xlabel('x')
plt.ylabel('y')
plt.legend()
plt.show()