Fully differentiable descriptors for atomistic systems
Project description
Libdescriptor
Supported Python versions: 3.8, 3.9, 3.10
Libdescriptor is a high performance descriptor library for providing access to fully differentiable descriptor functions.
While libdescriptor
is a general purpose descriptor library, it's API compatible with KIM models and associated projects.
This will also provide uniform access to various selected descriptors for KLIFF using Pybind11 ports.
For gradient calculations, Libdescriptor relies on Enzyme AD, which provides it with capability to trivially generate near analytical performance gradient functions.
Use of Enzyme AD enables Libdescriptor to not only provide gradients against coordinates, but against hyperparameters as well, thus opening way for better optimized descriptors.
This should enable rapid development, extension and deployment of various descriptors.
Installation
For AMD/Intel based Linux systems, we provide a precompiled binary package for libdescriptor. It can be installed using
pip install libdescriptor
Github Source: libdescriptor
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distributions
File details
Details for the file libdescriptor-0.2.3-py3-none-manylinux_2_5_x86_64.whl
.
File metadata
- Download URL: libdescriptor-0.2.3-py3-none-manylinux_2_5_x86_64.whl
- Upload date:
- Size: 14.4 MB
- Tags: Python 3, manylinux: glibc 2.5+ x86-64
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.2 CPython/3.9.15
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 880abe892571856ef0c79cb05e39420763c6faf28b63582eca1bd6e778f1534d |
|
MD5 | c56818fdcbeb688839041b40aa441d94 |
|
BLAKE2b-256 | 0ca32aa44bd0696d1cb4d9c4df1c2c689b8d568e41039eb7929ef2d4709b3ee0 |
File details
Details for the file libdescriptor-0.2.3-py3-none-macosx_11_0_arm64.whl
.
File metadata
- Download URL: libdescriptor-0.2.3-py3-none-macosx_11_0_arm64.whl
- Upload date:
- Size: 1.8 MB
- Tags: Python 3, macOS 11.0+ ARM64
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.2 CPython/3.9.15
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 1ff937a9eb459e63e3513c28046131be9eccf84e70a7fa531401d5bec13c3aa3 |
|
MD5 | 97157c99380730b5d4657f9b3c0a85fd |
|
BLAKE2b-256 | bf6bba18eb93ba6cf519244bb4f5378d14bab490e0e72ea73f3a1659b542e704 |