-
Notifications
You must be signed in to change notification settings - Fork 1
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
45 add basic structure for energy classes #46
Conversation
Thinking now about extending this to forces: Should the forces structure not be an exact mirror of energies? If so, how can we encourage or ensure that they are always updated/extended simultaneously? |
Are they tho? I am guessing that forces are matrices (or vectors?), while energies are scalars, no? What do you mean "updated/extended simultaneously"? We can do a quick chatting in a Discord room to understand this point too. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Reviewed outputs.py
, can do the rest if needed.
No real conceptual comments, just specifics.
You are reviewing the wrong changes. The ones in #43 are the ones you want to take a look. |
Sorry for being unclear. I just meant that besides the total force, there are various force contributions defined currently, e.g., "free" and "t0". These directly correspond to some energy contributions (although the schema is not nearly as built out as for energies. In principle, there could be a force contribution/definition for every energy, since they are directly related. My question was, why not enforce there to always be both terms added, for each type of contribution? And if this is desired how can we try to achieve this in practice? For the classical contributions I will build this mirroring regardless, since energy and force contributions coming from particular interactions are commonly stored. |
@JFRudzinski if you branch off from a non-merged branch, pls specify this in the PR or set the target branch to the proper base... |
Ah! Got it, Sorry about that! |
I don't think that we should enforce this across the board for 2 reasons:
With that said, I think it's a good idea to provide an optional functionality. Would recommend something numpy-based as implementation. |
I don't really understand what you are saying tbh. I am not talking about enforcing the storage of such quantities. I was talking about schema development. I think "enforcing" is a bit too strong of a word, but I had in mind the case that someone wants to store some energy contribution and implements a class for it, but doesn't create the corresponding force contribution because they personally do not store this in there simulations. But then another user comes along and wants to store this force contribution and now has to extend the schema themselves, whereas if the original person had been encouraged to add a corresponding force contribution it would have made everyone's life a lot easier. Of course there might be some energy definitions that it doesn't make sense to define a corresponding force. In that case, obviously it should not be done, but I think in most cases it is possible at least in principle. |
7bfc4e9
to
d7cd51f
Compare
To be honest, I need to see or talk about an example to fully understand the situation. But I'd wildly suggest to use But again, I need to see the example in a board or in a piece of paper. |
Ow, it sounded like you meant the actual storage, hence my reply. As for schema extension, while the scenario you mention is indeed suboptimal, how would you enforce your suggestion? |
@JFRudzinski bear in mind I merged something which created the sub-folder, |
Yes, got it, thank you 👍 |
Added extraction of full_shape of the PhyiscalProperty.value from variables and tensor order
Added functionality to see if a property is_scf_converged
d7cd51f
to
2f025fb
Compare
@JFRudzinski to avoid confusion I am closing this in favor of #91 You can add (in the description) a "Closes #numberofissue" so Github closes automatically the issue(s) related with #91 |
@ndaelman-hu @JosePizarro3 @Bernadette-Mohr Here is a first proposal for a basic structure for various energies. Please provide feedback at your convenience (you can exclusively look at
energies.py
)