pytorch04

linearRegression

np.array 默认 dtype=float64 (可用a.float查看dtype)
此处需定义为 dtype=np.float32 否则torch.nn.linear()无法输出？默认训练32位，64位基本用不上

class LinearRegression(nn.Module):
def init(self,input_size,output_size):
super(LinearRegression,self).init()
self.linear = nn.Linear(input_size,output_size)
def forward(self,x):
out = self.linear(x)
return out
def a(self,x):
print("x is ",x)
return 0
outputs=model(inputs)

调用nn.Module中的__call__ （__call__的作用是把类实例当做函数。）__call__再调用forward函数。所以不用model.forward(inputs)
更多关于 nn.module

https://www.cnblogs.com/hellcat/p/8473412.html