pytporch(3) 保存和提取 &&批处理

保存和提取

net = torch.nn.Sequential(
torch.nn.Linear(2, 10),
torch.nn.ReLu(),
torch.nn.Linear(10, 2)
)

plt.ion()
plt.show()

optimizers = torch.optim.SGD(net.parameters(), lr=0.02)
loss_func = torch.nn.CrossEntropyLoss()
#计算的是概率

for t in range(100):
     out = net(x)#这里输出的不是类别0/1 也不是概率 是上计算出来的值

     loss = loss_func(out, y)

     optimizers.zero_grad()
     loss.backward()
     optimizers.step()
    
torch.save(net, 'net.pkl')#整个网络
torch.save(net.state_dict(), 'net_params.pkl') #保存的是参数


def restore_net():
    net2 = torch.load('net.pkl')


def restore_params():
    #提取参数时候要先建立和上面相同的神经网络
    net3 = torch.nn.Senquential(
        torch.nn.Linear(1, 10),
        torch.nn.ReLu(),
        torch.nn.Linear(10, 1)
    )
    net3.load_state_dict(torch.load('net_params.pkl'))

 

批处理

import  torch
import torch.utils.data as Data

if __name__ == '__main__':
    BATCH_SIZE = 5
    x = torch.linspace(1, 10, 10)
    y = torch.linspace(10, 1, 10)

    torch_dataset = Data.TensorDataset(x, y)
    loader = Data.DataLoader(
        dataset=torch_dataset,
        batch_size=BATCH_SIZE,
        shuffle=True,
        num_workers=2,#每次提取loader中的数据都是有两个线程
    )

    for epoch in range (3):
        for step,(batch_x, batch_y) in enumerate(loader):
            #training
            print('Epoch', epoch, '| Step', step, '|batch x:', batch_x.numpy(), '|batch y:', batch_y.numpy())