https://github.com/ireaml/defects_workflow
Code used to build the defects dataset for the publication "Machine-learning structural reconstructions for accelerated point defect calculations"
https://github.com/ireaml/defects_workflow
Last synced: 5 months ago
JSON representation
Code used to build the defects dataset for the publication "Machine-learning structural reconstructions for accelerated point defect calculations"
- Host: GitHub
- URL: https://github.com/ireaml/defects_workflow
- Owner: ireaml
- Created: 2024-01-15T15:54:40.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-01-22T09:43:34.000Z (over 2 years ago)
- Last Synced: 2024-01-22T12:51:38.943Z (over 2 years ago)
- Language: Python
- Homepage:
- Size: 81.1 KB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# `defects_workflow`
Workflow to run defect calculations with aiida.
Currently, it automates the following steps:
1. Relaxation of host structure (from mp-id or user defined structure)
2. Defect generation
* Defect charge states are determined based on the most common oxidation states for the element (following
the strategy implemented in [defectivator](https://github.com/alexsquires/defectivator)
by Dr Alex Squires)
3. Screening of symmetry inequivalent interstitials.
This is done by relaxing the neutral state of all the symmetry inequivalent
configurations for a given interstitial. The following cases are filtered out:
* Configurations that lead to the same final structures (only one is used for later calculations)
* Configurations very high in energy compared to the most stable one (e.g. if > 1 eV)
4. Structure searching using shakenbreak and submission of calculations
# Installation
1. Crate conda environment (python 3.10)
2. Install `aiida-core` using the [system-wide installation](https://aiida.readthedocs.io/projects/aiida-core/en/latest/intro/install_system.html#intro-get-started-system-wide-install) and using `pip` rather than `conda`.
3. Install other dependencies, including `aiida-archer2-scheduler` (to use the HPC `archer2`),
`parsevasp`, `aiida-vasp`, `aiida-user-addons` and `defectivator`:
```
git clone git@github.com:SMTG-UCL/aiida-archer2-scheduler.git
cd aiida-archer2-scheduler
pip install -e ./
reentry scan -r aiida
```
```
git clone https://github.com/aiida-vasp/parsevasp.git
cd parsevasp
git checkout develop
cd ../
pip install -e ./parsevasp
```
```
git clone https://github.com/aiida-vasp/aiida-vasp.git
cd aiida-vasp
git checkout develop
cd ../
pip install -e ./aiida-vasp
```
```
git clone https://github.com/SMTG-UCL/aiida-user-addons.git
cd aiida-user-addons
git checkout dev
cd ../
pip install -e ./aiida-user-addons
```
```
git clone https://github.com/alexsquires/defectivator.git
cd defectivator
git checkout dev
cd ../
pip install ./defectivator
```
Run `pip install reentry` and `reentry scan -r aiida`
4. Configure `aiida-vasp` (potcars)
5. Install `shakenbreak`
```
git clone https://github.com/SMTG-UCL/shakenbreak.git
pip install .
```
6. Install `defects_workflow`
```
git clone https://github.com/ireaml/defects_workflow.git
pip install .
```
7. Add ab-initio codes to aiida profile