Comments (6)
Hey,
and sorry for commenting this so late (our everyday life here changed drastically due to Corona..). That is excellent news, sounds great! I am looking forward to trying it out.
I have one minor suggestion: I originally intended to keep only the core functionality within the libnnp
directory. This way compiling the basic parts of the package will be possible without many dependencies. Interfaces to external software would then go to the libnnpif
("if" for interface) directory. Then again, I am beginning to question my original idea to put ALL interfaces to libnnpif
.. it makes no sense that a user who wants to build the LAMMPS interface also needs to fulfill all requirements for the future CabanaMD link. So, I am adding to the task list something for me to think about:
- Find a clever way to organize interfaces with external software.
All the best,
Andi
from n2p2.
My only new thought in finally opening the PR to close this is that perhaps libnnpif
could be renamed libnnpif-lammps
(or both LAMMPS and Cabana interfaces named something new)
from n2p2.
I finally merged #49 with some minor changes:
- Renamed the source files (i.e.
ElementCabana
instead ofElement_Cabana
) - Added documentation (see here), I included also a link to your paper on arXiv.
My plan is to create a new release version (via Github releases, which will automatically create an archive and a DOI on Zenodo) as soon as you finish work on CabanaMD with NNP (ECP-copa/CabanaMD#61, ECP-copa/CabanaMD#37). This way it may be easier for users and also for CI tests.. do you think this is a good idea or do you suggest something different? Also, I will take version numbers more serious in the future. I am thinking of integrating a CabanaMD build also in Travis to make sure I don't break things without noticing.
Let me know if you need some changes or want me to do some testing!
from n2p2.
Great! I merged and removed all n2p2 code from CabanaMD.
I agree; tagged versions are best. I will update my CI once the n2p2 tag is available. I assume you would also prefer a release of CabanaMD? I have one more update in functionality before I would like to do one, but that should be done soon.
At the very least any failures due to interface should show up in CabanaMD for now
from n2p2.
Maybe best to close this and open a similar CI/tag issue here?
from n2p2.
Yes, I will add some basic testing when a tag is available, no need to hurry. Created an issue: #65. Closing this one, thanks again for your efforts!
from n2p2.
Related Issues (20)
- Difference between nnp-comp2 and lammps interface predictions
- On the convergence of nnp training
- nnp-atomenv usage?
- undefined reference to `nnp::Log::registerCFilePointer(_IO_FILE** const&)' while building with lammps
- Can't build nnp-train: /bin/ld: ../../lib/libnnptrain.a(KalmanFilter.o): undefined reference to symbol 'dtrsm_'
- Compute stress/atom in lammps interface
- Training stops suddenly
- Angular Symmetry Functions
- LAMMPS NNP interface on macOS HOT 1
- nnp-predict HOT 2
- installation problem HOT 2
- Excessive time spent on MPI Bcast in nnp-train HOT 1
- Training on data with very low (<0.05 eV/A) forces - large force errors
- Problem with using n2p2 trained network in cp2k HOT 1
- N2P2/Lammps interface for performing 4G-HDNNP MD simulations. HOT 2
- difficulty to compile interface to LAMMPS (version 3 Nov 2022) HOT 2
- Cannot declare multiple elements
- Compute Stress n2p2/lammps interface
- Include an "inner cutoff" that suppresses the symmetry function values at short distances.
- Data set normalization
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from n2p2.