Replies: 1 comment 1 reply
-
|
You don't provide a reference solution, then how to compute l2 relative error? Just delete |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
博士您好
我是一个接触tensorflow和deepxde的新手,在我学习deepxde的过程中,我遇到了一些我暂时解决不了的问题,在这个代码里,使用metrics=["l2 relative error"],出现错误,您能解释一下为什么吗?然后改如何改正呢?
`import deepxde as dde
import math
def pde(x,y):
dy_xx = dde.grad.hessian(y,x,i=0,j=0)
dy_yy = dde.grad.hessian(y,x,i=1,j=1)
return -dy_yy - dy_xx - 1
def boundary(_,on_boundary):
return on_boundary
geom = dde.geometry.Rectangle([0, 0], [math.pi, math.pi])
bc = dde.DirichletBC(geom,lambda x:0, boundary)
data = dde.data.PDE(geom,pde,bc,num_domain=1200,num_boundary=120,num_test=1500)
net = dde.maps.FNN([2]+[50]*4+[1],"tanh","Glorot uniform")
model = dde.Model(data,net)

model.compile("adam",lr=0.001,metrics=["l2 relative error"])
losshistory,train_state = model.train(epochs=10000)
dde.saveplot(losshistory,train_state,issave=True,isplot=True)`
这是程序的报错,但是我没有明白是为什么:
Beta Was this translation helpful? Give feedback.
All reactions