AI入门之神经网络（10）基于python/pytorch的线性回归学习

# 线性模型
# 使用pytorch的一般3步骤
# 1）设计模型的输入与输出和前向计算和各层的规模
# 2）选择损失函数和优化器
# 3）循环训练
# ——3.1前向计算
# ——3.2反向计算
# ——3.3更新权重
import torch
import torch.nn as nn
import numpy as np
import matplotlib.pyplot as plt # 画图包
from sklearn import datasets # 机器学习库这里用来导入数据

# 数据
x_numpy, y_numpy = datasets.make_regression(n_samples=100, n_features=1, noise=20, random_state=1)

x = torch.from_numpy(x_numpy.astype(np.float32))
y = torch.from_numpy(y_numpy.astype(np.float32))
y = y.view(y.shape[0], 1)

n_samples, n_features = x.shape
# 模型
input_size = n_features
output_size = 1
model = nn.Linear(input_size, output_size)
# 损失和优化器
learning_rate = 0.5
criterion = nn.MSELoss()
optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)

# 训练循环
epochs_num = 1000 # 训练次数
for epoch in range(epochs_num): # 循环训练
    #前向传播
    y_predicted = model(x)
    loss = criterion(y_predicted, y)
    #反向传播
    loss.backward()
    #更新权重
    optimizer.step()
    optimizer.zero_grad() # 梯度清零

    if (epoch+1) % 10 == 0:
        print(f'epoch: {epoch+1}, loss = {loss.item():.3f}')

# 画图plot
predicted = model(x).detach().numpy()
plt.plot(x_numpy, y_numpy, 'ro')
plt.plot(x_numpy, predicted, 'b')
plt.show()

如果你很喜欢AI，那么请你坚持下去！----小菜鸡