代码实现:
import torch.nn
#3*1的矩阵
x_data=torch.Tensor([[1.0],
[2.0],
[3.0]])
#3*1的矩阵
y_data=torch.Tensor([[2.0],[4.0],[6.0]])
#继承类Module,自动会实现反向计算图
class LinearModel(torch.nn.Module):
#构造方法
def __init__(self):
super(LinearModel, self).__init__()
self.linear=torch.nn.Linear(1,1) #类(),构造了一个对象。
def forward(self, x):
y_pred=self.linear(x) #对象(),实现了一个可调用的对象
return y_pred
model=LinearModel()
criterion=torch.nn.MSELoss(size_average=False)
optimizer=torch.optim.SGD(model.parameters(),lr=0.01)
for epoch in range(1000):
y_pred=model(x_data)
loss=criterion(y_pred,y_data)
print(epoch,loss.item())
optimizer.zero_grad()
loss.backward()
optimizer.step()
print('w=',model.linear.weight.item())
print('b=',model.linear.bias.item())
x_test=torch.Tensor([[4.0]])
y_test=model(x_test)
print('y_pred=',y_test.data)
输出:
说明
- 类Linear
- 2.可调用对象
class Footbar:
def __init__(self):
pass
def __call__(self,*args,**kwargs):
print("hello"+str(args[0]))
def func(*args,**kwargs):
print(args)
print(kwargs)
func(1,2,4,3,x=3,y=5)
footbar=Footbar()
footbar(1,2,3)
输出:
(1, 2, 4, 3)
{'x': 3, 'y': 5}
hello1
3.MSELoss和优化器