-
Notifications
You must be signed in to change notification settings - Fork 22
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add classical optimisation problems to testing suite & examples #391
Comments
Do you mean throwing these functions to the optimisers or going through the PyBOP problem class? I think the former is more adequate. The other question I have is, should the examples be with the other examples? If we just use these functions for testing, we can define them in the tests folder. Otherwise, if we want to use them in the examples, we will have to define them inside PyBOP. |
The former would be my preferred method as well. I think an easy entry point is inheriting the BaseCost for these functions.
That's a good question, I think it would be nice to offer these to users. In #481, I've created an |
If they inherit from BaseCost we could incude them in asubdirectory in the costs folder |
Since they won't have a problem class, they won't be able to be used exactly like the other cost classes. This limitation could be captured in their docstrings, but the experimental directory does provide a clear warning to users for the beginning. Either works with me, as long as we are clear what the limitations are :) For reference the standalone cost class is a good starting place for this: https://github.com/pybop-team/PyBOP/blob/develop/examples/standalone/cost.py |
I am keen on implementing this, but probably won't have much time in the next few weeks. If someone else is keen, please go ahead. Otherwise I will give it a go once I have a bit of time. |
Feature description
In addition to our current battery focused identification and optimisation tasks, we should add tests and examples with classical benchmark optimisation tasks. An initial list of potential functions is available here
Motivation
This can help validate our workflows outside of battery focused tasks and provide benchmarks against other inference packages. Likewise, this could help the MCMC integration tests in #340 by having a higher performance forward model.
Possible implementation
No response
Additional context
No response
The text was updated successfully, but these errors were encountered: