Replies: 1 comment 2 replies
-
It is correct. It is designed in this way.
There is no guarantee of the order and values of x during training. To integrate, you can check https://deepxde.readthedocs.io/en/latest/demos/pinn_forward.html#integro-differential-equations |
Beta Was this translation helpful? Give feedback.
2 replies
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.
-
Hello Lu.
I am trying to develop a SBINN using the DeepXDE library. In my system I need to integrate one of the variables. For this I extract the values of the time tensor t and variable G_PL as numpys arrays. However I noticed that the length of t is not as expected. It seems that the length of t inside the ODE system is number of initial conditions + the length of the observed time * (number of observed variables +1), with this last 1 appearing from anchors.
I do not understand the meaning of this and would like to know which portion of t and G_PL I can use to integrate. If there is another way to integrate a state variable (from current time - window until current time), that would also work for me.
Thank you in advance.
Beta Was this translation helpful? Give feedback.
All reactions