Machine learning nonparametric force fields (MFF)
An example tutorial jupyter notebook can be found in the tutorials folder.
Table of Contents
Background on MFF
The MFF package uses Gaussian process regression to extract non-parametric 2- and 3- body force fields from ab-initio calculations. For a detailed description of the theory behind Gaussian process regression to predict forces and/or energies, and an explanation of the mapping technique used, please refer to [1].
For an example use of the MFF package to build 3-body force fields for Ni nanoclusters, please see [2].
Pip Installation
To install MFF with pip, simply run the following in a Python 3.6 or 3.7 environment:
pip install mff
Source Installation
If the pip installation fails, try the following: Clone from source and enter the folder:
git clone https://github.com/kcl-tscm/mff.git
cd mff
If you don't have it, install virtualenv:
pip install virtualenv
Create a virtual environment using a python 3.6 installation
virtualenv --python=/usr/bin/python3.6 <path/to/new/virtualenv/>
Activate the new virtual environment
source <path/to/new/virtualenv/bin/activate>
To install from source run the following command:
python setup.py install
Or, to build in place for development, run:
python setup.py develop
Examples
Refer to the two files in the Tutorial folder for working jupyter notebooks showing most of the functionalities of this package.
Maintainers
- Claudio Zeni (claudio.zeni@kcl.ac.uk),
- Aldo Glielmo (aldo.glielmo@kcl.ac.uk),
- Ádám Fekete (adam.fekete@kcl.ac.uk).
References
[1] A. Glielmo, C. Zeni, A. De Vita, Efficient non-parametric n-body force fields from machine learning (https://arxiv.org/abs/1801.04823)
[2] C .Zeni, K. Rossi, A. Glielmo, A. Fekete, N. Gaston, F. Baletto, A. De Vita Building machine learning force fields for nanoclusters (https://arxiv.org/abs/1802.01417)
Metadata
Release files for mff 1.1.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| mff-1.1.2.tar.gz | 84.3 kB | Without provenance | Details for mff-1.1.2.tar.gz |
Release files / mff-1.1.2.tar.gz
| Download URL | mff-1.1.2.tar.gz |
|---|---|
| Size | 84.3 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
25ba7fa00b864098971ce7f4956e6eeec9a2f276d7ce7e72ffe0d82e29e0c6cc
|
|
BLAKE2b-256 checksum How to use checksums |
49cad45ea4c20468b96c082af94efed3c77640c5f6639311cbc9d62f8f5bfaaa
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.1.1 pkginfo/1.5.0.1 requests/2.22.0 setuptools/45.2.0 requests-toolbelt/0.9.1 tqdm/4.42.1 CPython/3.6.7
|
Security for mff 1.1.2
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.