flare++
Major features:
- Bayesian force fields based on sparse Gaussian process regression.
- Multielement many-body descriptors based on the atomic cluster expansion.
- Mapping to efficient parametric models.
- Coupling to LAMMPS for large-scale molecular dynamics simulations.
Check out our preprint introducing flare++ here.
Demo and Instructions for Use
An introductory tutorial in Google Colab is available here. The tutorial shows how to run flare++ on energy and force data, demoing "offline" training on the MD17 dataset and "online" on-the-fly training of a simple aluminum force field. A video walkthrough of the tutorial, including detailed discussion of expected outputs, is available here.
The tutorial takes a few minutes to run on a normal desktop computer or laptop (excluding installation time).
Installation guide
The easiest way to install is with pip:
pip install flare_pp
This will take a few minutes on a normal desktop computer or laptop.
If you're installing on Harvard's compute cluster, make sure to load the following modules first:
module load cmake/3.17.3-fasrc01 python/3.6.3-fasrc01 gcc/9.3.0-fasrc01
Compiling LAMMPS
System requirements
Software dependencies
- cmake>=3.14.5 (to compile the C++ source code)
- flare (for on-the-fly training)
Operating systems
flare++ is tested with Github Actions on a Linux operating system (Ubuntu 20.04.3). You can find a summary of recent builds here.
We expect flare++ to be compatible with Mac and Windows operating systems, but can't guarantee this. If you run into issues running the code on Mac or Windows, please post to the issue board.
Hardware requirements
There are no non-standard hardware requirements to download the software and train simple models—the introductory tutorial can be run on a single cpu. To train large models (10k+ sparse environments), we recommend using a compute node with at least 100GB of RAM.
Documentation
Preliminary documentation of the C++ source code can be accessed here.
1 maintainer
Metadata
1 maintainer
Release files for flare-pp 0.1.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| flare_pp-0.1.1.tar.gz | 386.7 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| flare_pp-0.1.1-cp36-cp36m-macosx_10_14_x86_64.whl | CPython 3.6 | CPython 3.6 pymalloc | macOS 10.14+ x86-64 | Details |
Total release size: 892.2 kB
Release files / flare_pp-0.1.1.tar.gz
| Download URL | flare_pp-0.1.1.tar.gz |
|---|---|
| Size | 386.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
1bc56999b915568984f49498cf40f963238d74e1be1c2a5058cc132fcfb8367a
|
|
BLAKE2b-256 checksum How to use checksums |
7a51d18f03a81cca7b894fc698ead3f74d342b3c6e322425688a04e14bc550a3
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.1.1 pkginfo/1.5.0.1 requests/2.23.0 setuptools/46.1.3.post20200330 requests-toolbelt/0.9.1 tqdm/4.46.0 CPython/3.6.10
|
Release files / flare_pp-0.1.1-cp36-cp36m-macosx_10_14_x86_64.whl
| Download URL | flare_pp-0.1.1-cp36-cp36m-macosx_10_14_x86_64.whl |
|---|---|
| Size | 505.5 kB |
| Tags | CPython 3.6 CPython 3.6 pymalloc macOS 10.14+ x86-64 |
|
SHA-256 checksum How to use checksums |
0c6e109226ff61bd9f90d9349683978bd0c6330447ed94324b7ae9e2fe6dd94f
|
|
BLAKE2b-256 checksum How to use checksums |
02643ca762bb61047c0bed6da13e6d6c2768eb832f1ecb0929af2a7addd2b382
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.1.1 pkginfo/1.5.0.1 requests/2.23.0 setuptools/46.1.3.post20200330 requests-toolbelt/0.9.1 tqdm/4.46.0 CPython/3.6.10
|
