Version: Next

Au–Ag Alloy Example

This example demonstrates active learning for an Au–Ag alloy and is located in pwact/example/auag_pwmat/. It first uses init_bulk to construct the initial training set and trains a model on that dataset. Structures generated by perturbation during init_bulk are then used as initial configurations, and active-learning sampling is controlled by a user-provided lammp.in file.

tip

Active-learning sampling with a user-provided lammp.in control file is supported in pwact >= 0.4.

The example below uses PWmat for DFT calculations. Corresponding configurations are also provided for VASP in pwact/example/auag_vasp/ and CP2K in pwact/example/auag_cp2k/. Mcloud users can find the examples under /share/public/PWMLFF_test_data/pwact_examples/25-pwact-demo.

Note that the DFT settings supplied with the example are intended only to test the workflow and do not guarantee calculation accuracy.

Prepare Pretraining Data with init_bulk

Command: Enter the pwact/example/auag_pwmat/init_bulk directory:

pwact init_bulk init_param.json resource.json

init_bulk Directory Structure

The init_bulk directory has the same structure as in the si_pwmat example.

Active Learning with run

Command: After init_bulk completes, enter the examples/auag_pwmat/run_iter_lmps directory:

pwact run param.json resource.json

Active-Learning Directory

The active-learning directory has the same structure as in the si_pwmat example. The only difference is the MD subdirectory name: for example, md.000.sys.000/md.000.sys.000.t.000.p.000 becomes md.000.sys.000/md.000.sys.000.lmps.000, where lmps.000 identifies the selected lammps.in file.

example
├──param.json
├──resource.json
├──scf_etot.input
├──iter.0000
│    ├──00.train
│    │   ├──...
...
│    ├──01.explore
│    │   ├──md
│    │   │    ├──md.000.sys.000
│    │   │    │     ├──md.000.sys.000.lmps.000
│    │   │    │     ├──...
│    │   │    │     └──model_devi_distribution.png
│    │   │    ├──md.000.sys.001
│    │   │    ├──md.001.sys.000
│    │   │    └──md.001.sys.003
│    │   └──select
│    │        ├──accurate.csv
│    │        ├──candidate.csv
│    │        ├──candidate_delete.csv
│    │        ├──fail.csv
│    │        ├──error_traj.log
│    │        ├──select_summary.txt
│    │        ├──model_devi_distribution-md.000.sys.000.png
│    │        └──...
│    └──02.label
│    │   ├──scf
│    │   │    ├──md.000.sys.000
│    │   │    │    ├──md.000.sys.000.t.001
│    │   │    │    │    ├──820-scf
│    │   │    │    │    │    ├──820.config
│    │   │    │    │    │    ├──etot.input
│    │   │    │    │    │    ├──REPORT
│    │   │    │    │    │    └──OUT.MLMD
│    │   │    │    │    ├──200-scf
│    │   │    │    │    └──...
│    │   │    │    └──md.000.sys.000.t.000
│    │   │    │         └──...
│    │   │    ├──md.000.sys.001
│    │   │    ├──...
│    │   │    ├──md.001.sys.000
│    │   └──result
│    │        └──train.xyz

├──iter.0001
│    └──...
├──...