-
Notifications
You must be signed in to change notification settings - Fork 95
Add Tensornet2 architecture #375
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
Conversation
|
It also includes some Tensornet optimizations first attempted in this PR: #320
where F is the embedding dimension and the leading dim is the number of nodes or number of edges. |
|
I see you merged this just after creating the 2.5 release. How soon will there be a release that can support AceFF-2.0? |
|
The plan is to release it this week, I just want to take a look at the PR
with the examples on Wednesday
Στις Δευ 5 Ιαν 2026, 21:10 ο χρήστης Peter Eastman ***@***.***>
έγραψε:
… *peastman* left a comment (torchmd/torchmd-net#375)
<#375 (comment)>
I see you merged this just after creating the 2.5 release. How soon will
there be a release that can support AceFF-2.0?
—
Reply to this email directly, view it on GitHub
<#375 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AB4RLAROV7BZCBRLXEVORF34FKZIZAVCNFSM6AAAAACQJS27WKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTOMJRG42DENBRG4>
.
You are receiving this because you modified the open/close state.Message
ID: ***@***.***>
|
|
The paper Is out, so release It, then we will do more updates.
Last time it took a month to release the paper in the archive, this time 1
day.
G
…On Mon, 5 Jan 2026, 21:25 Stefan Doerr, ***@***.***> wrote:
*stefdoerr* left a comment (torchmd/torchmd-net#375)
<#375 (comment)>
The plan is to release it this week, I just want to take a look at the PR
with the examples on Wednesday
Στις Δευ 5 Ιαν 2026, 21:10 ο χρήστης Peter Eastman ***@***.***>
έγραψε:
> *peastman* left a comment (torchmd/torchmd-net#375)
> <#375 (comment)>
>
> I see you merged this just after creating the 2.5 release. How soon will
> there be a release that can support AceFF-2.0?
>
> —
> Reply to this email directly, view it on GitHub
> <#375 (comment)>,
> or unsubscribe
> <
https://github.com/notifications/unsubscribe-auth/AB4RLAROV7BZCBRLXEVORF34FKZIZAVCNFSM6AAAAACQJS27WKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTOMJRG42DENBRG4>
> .
> You are receiving this because you modified the open/close state.Message
> ID: ***@***.***>
>
—
Reply to this email directly, view it on GitHub
<#375 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AB3KUOQEJTBGWRZRE5SPTQ34FLCDPAVCNFSM6AAAAACQJS27WKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTOMJRHE3DQNZUHE>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
|
Done, release building right now as 2.6.0. Should be available in 30ish minutes |
adds the new tensornet2 architecture for the aceff-2.0 pre-trained model.
the key addition is Aimnet2 style neutral charge equilibration and a coulomb interaction.
This PR adds the tensonet2 model and a new output model that does a coulomb interaction using the predicted charges.
Note that the parent branch is #373