materialsvirtuallab / mlearn Goto Github PK
View Code? Open in Web Editor NEWBenchmark Suite for Machine Learning Interatomic Potentials for Materials
License: BSD 3-Clause "New" or "Revised" License
Benchmark Suite for Machine Learning Interatomic Potentials for Materials
License: BSD 3-Clause "New" or "Revised" License
Hello,
What is the unit of the virial stress Si data set? Also, for each structure, the virial stress has 6 members. How are these members organized?
Thank you for your help in advance.
Mo and Cu dataset is identical to Ni dataset.
Dependabot couldn't authenticate with https://pypi.python.org/simple/.
You can provide authentication details in your Dependabot dashboard by clicking into the account menu (in the top right) and selecting 'Config variables'.
Could you please update example scripts in 'notebooks'?
I downloaded this package on Dec. 4, 2019 & successfully installed various descriptors (except for MTP which needs a license from the author). However, when I tried to run the examples scripts in the directory 'notebooks', I found they were kind of obsolete. For instant, to import the GAP potential, it is written in the script as "from mlearn.potential.gap import GAPotential". While I checked the location of 'mlearn' package, and the folder has already been renamed from 'potential' to 'potentials'. There are more incompatibilities like this. Accordingly, the example scripts cannot even be ran.
As a beginner, I'm so grateful that your teams can share such a handy tool to run various machine learning potentials. But I think it is really important for us to follow the example scripts, since there is no full documentation with respect to this tool available online.
Many thanks!
Currently fitting a potential require the installation of specific packages, e.g., QUIP for GAP and LAMMPS for SNAP. mlearn codes work as python wrappers for those packages. We may need to find a way to share those packages, or tell the users where to find them and how to install them so that the users can use mlearn.
Can author give me some tips or turtorial about how to train a new model.
Hi,
I have a question regarding the scaling.data. For example here:
mlearn/mvl_models/Cu/nnp/scaling.data
Line 1 in 7606e21
What are the columns representing?
Hi,
I am trying to reproduce the result for the Cu dataset and I have several questions about it. In the input.nn file, in the atom_energy, it is "Cu -0.97". What is the meaning of it? The energy of each Cu atom is subtracted by 0.97 eV or Hartree? Also, I am wondering which standardization scheme is used for the fingerprints.
fp = (fp - fp_min)/(fp_max - fp_min) or something else?
Thanks!
Mingjie
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.