pytorch 多元回归

拟合函数y = 0.90 + 0.50 * x + 3.00 * x^2 + 2.40 * x^3

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

#%%
# 定义一个多变量函数
w_target = np.array([0.5, 3, 2.4])
b_target = np.array([0.9]) 

#%%

f_des = 'y = {:.2f} + {:.2f} * x + {:.2f} * x^2 + {:.2f} * x^3'.format(
    b_target[0], w_target[0], w_target[1], w_target[2]) 
print(f_des)
print(f'y = {b_target[0]} + {w_target[0]} * x + {w_target[1]} * x^2 + {w_target[2]} * x^3')
#%%
# 画出函数图像
x_sample = np.arange(-3, 3.1, 0.1)
y_sample = b_target[0] + w_target[0] * x_sample + w_target[1] * x_sample ** 2 +w_target[2] * x_sample ** 3
plt.plot(x_sample, y_sample, label='real curve')
plt.legend()

构建数据集，需要x和y，同时是一个三次多项式，所以我们取了x,x²,x³

# 构造数据x和y
# x是一个矩阵[[x,x2,x3]]
# y是函数结果 [[y]]
x_train = np.stack([x_sample ** i for i in range(1, 4)], axis=1)
x_train = torch.from_numpy(x_train).float() 
y_train = torch.from_numpy(y_sample).float().unsqueeze(1) 
# 初始化权重 
w = torch.randn((3,1), requires_grad=True)
b = torch.zeros((1), requires_grad=True)
# 定义模型
def multi_linear(x):
    return torch.mm(x, w) + b
# 定义损失函数
def get_loss(y_pred,y):
    return torch.mean((y_pred-y)**2)
    
# 进行100次参数更新
for e in range(100):
    y_pred = multi_linear(x_train)
    loss = get_loss(y_pred, y_train)
    
    w.grad.data.zero_()
    b.grad.data.zero_()
    loss.backward()
    
    # 更新参数
    w.data = w.data - 0.001 * w.grad.data
    b.data = b.data - 0.001 * b.grad.data
    if (e + 1) % 20 == 0:
        print('epoch {}, Loss: {:.5f}'.format(e+1, loss.item()))

# 画出图像
y_pred = multi_linear(x_train)
plt.plot(x_train.data.numpy()[:, 0], y_pred.data.numpy(), label='fitting curve',
color='r')
plt.plot(x_train.data.numpy()[:, 0], y_sample, label='real curve', color='b')
plt.legend()

可以看到，经过100次更新之后，可以看到拟合的线和真实的线已经完全重合了