# NeuralPDE debuging: is it posible to get the loss of a point in the domain?
**URL:** https://discourse.julialang.org/t/neuralpde-debuging-is-it-posible-to-get-the-loss-of-a-point-in-the-domain/105427
**Category:** Modelling & Simulations
**Created:** [October 26, 2023, 9:51am UTC](https://discourse.julialang.org/t/neuralpde-debuging-is-it-posible-to-get-the-loss-of-a-point-in-the-domain/105427 "2023-10-26T09:51:53Z")
**Posts on this page:** 3
**Page:** 1
### Author:  [@Knud\_Sorensen](https://discourse.julialang.org/u/Knud_Sorensen)
#### Post date: [October 26, 2023, 9:51am UTC](https://discourse.julialang.org/t/neuralpde-debuging-is-it-posible-to-get-the-loss-of-a-point-in-the-domain/105427/1 "2023-10-26T09:51:53Z")
Hi.
I am solving some differential equations with NeuralPDE and the solutions don’t look like expected. So, I really want to plot the loss at every point in the domain to see what is going on. Is there any way to get this loss information out for each point, maybe from symbolic\_discretize or something else ?
---
### Author:  [@Knud\_Sorensen](https://discourse.julialang.org/u/Knud_Sorensen)
#### Post date: [October 27, 2023, 11:42am UTC](https://discourse.julialang.org/t/neuralpde-debuging-is-it-posible-to-get-the-loss-of-a-point-in-the-domain/105427/2 "2023-10-27T11:42:11Z")
When I look inside the structure from symbolic\_discretize
I see
symbolic\_bc\_loss\_functions
symbolic\_pde\_loss\_functions
And insider .loss\_functions I find.
additional\_loss\_function
bc\_loss\_functions
datafree\_bc\_loss\_functions
datafree\_pde\_loss\_functions
full\_loss\_function
pde\_loss\_functions
Is there any documentation on how to use those loss functions ?
I can find examples with bc\_loss\_functions and pde\_loss\_functions
but not the others.
---
### Author:  [@ChrisRackauckas](https://discourse.julialang.org/u/ChrisRackauckas)
#### Post date: [April 7, 2024, 1:32pm UTC](https://discourse.julialang.org/t/neuralpde-debuging-is-it-posible-to-get-the-loss-of-a-point-in-the-domain/105427/3 "2024-04-07T13:32:27Z")
We are redoing the lowering process to give the symbolic loss function
>
>
> The current setup was written before we had all the useful tools present in Symb…olics.jl, this PR takes advantage of these and the PDE helper functions present in PDEBase
When that’s done this should be better documented and easier to do.