The component of y is missing.

[ngaiheibrian]

Hello Sir, I am new to deepxde and try to apply it on my project. However, my code was failed to run.

It showed:

raise ValueError("The component of y is missing.")
ValueError: The component of y is missing.

I would be grateful if you can point out which part of my code is wrong.

Thanks

"""Backend supported: tensorflow.compat.v1, tensorflow, pytorch, paddle, jax"""
import deepxde as dde
import matplotlib.pyplot as plt
import numpy as np
from scipy import integrate

Import tf if using backend tensorflow.compat.v1 or tensorflow

from deepxde.backend import tf

Import torch if using backend pytorch

import torch

Import paddle if using backend paddle

import paddle

Import jax.numpy if using backend jax

import jax.numpy as jnp

def ode_system(x,y):
m=0.2591
k=100
damping_ratio=0.05
c=damping_ratio
time_applied=1
ui=193
u=0

y1  =  y[:,0:1]
y2  =  y[:,1:2] 
y3  =  y[:,2:3] 
y4  =  y[:,3:4]
y5  =  y[:,4:5]

dy1_t= dde.grad.jacobian(y,x,i=0, j=0)
dy2_t= dde.grad.jacobian(y,x,i=1, j=0)
dy3_t= dde.grad.jacobian(y,x,i=2, j=0)
dy4_t= dde.grad.jacobian(y,x,i=3, j=0)
dy5_t= dde.grad.jacobian(y,x,i=4, j=0)

dy1_tt =dde.grad.hessian(y, x, i=0, j=0)
dy2_tt =dde.grad.hessian(y, x, i=1, j=0)
dy3_tt =dde.grad.hessian(y, x, i=2, j=0)
dy4_tt =dde.grad.hessian(y, x, i=3, j=0)
dy5_tt =dde.grad.hessian(y, x, i=4, j=0)

if x[:,0:1]<=time_applied:
    u=ui*tf.sin(2*np.pi*x[:, 0:1])

else:
    u=0









return [
        m*dy1_tt+c*dy1_t+k*(2*y1-y2)+m*u,
        m*dy2_tt+c*dy2_t+k*(-y1+2*y2-y3)+m*u,
        m*dy3_tt+c*dy3_t+k*(-y2+2*y3-y4)+m*u,
        m*dy4_tt+c*dy4_t+k*(-y3+2*y4-y5)+m*u,
        m*dy5_tt+c*dy5_t+k*(-y4+y5)+m*u
        ]

def boundary(_, on_initial):
return on_initial

def func(x):
"""
y1 = sin(x)
y2 = cos(x)
"""
return np.hstack(x)

geom = dde.geometry.TimeDomain(0, 20)
ic1 = dde.icbc.IC(geom, lambda x: 0, boundary, component=0)
ic2 = dde.icbc.IC(geom, lambda x: 0, boundary, component=1)
ic3 = dde.icbc.IC(geom, lambda x: 0, boundary, component=2)
ic4 = dde.icbc.IC(geom, lambda x: 0, boundary, component=3)
ic5 = dde.icbc.IC(geom, lambda x: 0, boundary, component=4)

data = dde.data.PDE(geom, ode_system, [ic1, ic2, ic3, ic4, ic5], 35, 2, solution=func, num_test=100)

layer_size = [1] + [50] * 3 + [5]
activation = "tanh"
initializer = "Glorot uniform"
net = dde.nn.FNN(layer_size, activation, initializer)

model = dde.Model(data, net)
model.compile("adam", lr=0.001, metrics=["l2 relative error"])
losshistory, train_state = model.train(iterations=20000)

dde.saveplot(losshistory, train_state, issave=True, isplot=True)

[ghost]

You have to put your code inside markdown code block.
See https://deepxde.readthedocs.io/en/latest/modules/deepxde.html#deepxde.gradients.hessian

dy1_tt = dde.grad.hessian(y, x, component=0, i=0, j=0)
dy2_tt = dde.grad.hessian(y, x, component=1, i=0, j=0)
dy3_tt = dde.grad.hessian(y, x, component=2, i=0, j=0)
dy4_tt = dde.grad.hessian(y, x, component=3, i=0, j=0)
dy5_tt = dde.grad.hessian(y, x, component=4, i=0, j=0)

u = torch.where(torch.less(x[:, 0:1], time_applied), ui * torch.sin(2 * np.pi * x[:, 0:1]), 0)

[ngaiheibrian]

Thanks a lot, I finally sorted the problem!!

[ngaiheibrian]

And guys, I have another question. As I am working on a project that I need to train a model to predict a bridge motion which contains over 80k node and it is impossible to write the program using the method ie dy1 dy2..... , is there any tips for me to build a model using deepxde?

[lululxvi]

#865