pytorch实现线性回归

import numpy as np
import torch
x = np.linspace(0,20,100).astype(np.float32).reshape((-1,1))
y = x*4 + np.random.rand((*x.shape)).astype(np.float32).reshape((-1,1))


import torch.nn as nn
class LinearRegressionModel(nn.Module):

    def __init__(self, input_dim, output_dim):

        super(LinearRegressionModel, self).__init__() 
        self.linear = nn.Linear(input_dim, output_dim)

    def forward(self, x):

        out = self.linear(x)
        return out

input_dim = 1
output_dim = 1

net = LinearRegressionModel(input_dim,output_dim)
net = net.cuda()

criterion = nn.MSELoss()# Mean Squared Loss
l_rate = 0.001
optimizer = torch.optim.SGD(net.parameters(), lr = l_rate) #Stochastic Gradient Descent

epochs = 10000


for epoch in range(epochs):

    epoch +=1
    #increase the number of epochs by 1 every time
    inputs = torch.from_numpy(x)
    labels = torch.from_numpy(y)
    inputs = inputs.cuda()
    labels = labels.cuda()
    outputs = net(inputs)
    
    optimizer.zero_grad()
    loss = criterion(outputs, labels)
    loss.backward()
    optimizer.step()
    
    if epoch % 200 == 0:
        print('epoch {}, loss {}, w {}, b {},'.format(epoch,loss.item(), net.linear.weight.item(), net.linear.bias.item()))

转载于:https://www.jianshu.com/p/817d9b3137d1