Added ABACUS pw interface for dpgen init_bulk #590
Conversation
|
It seems that dpdata installed in the test does not support abacus/md yet, leading to the failure of the test. |
|
Could you please merge the recent devel branch? Thanks! |
Codecov Report
@@ Coverage Diff @@
## devel #590 +/- ##
==========================================
+ Coverage 33.16% 33.91% +0.74%
==========================================
Files 86 86
Lines 14837 15425 +588
==========================================
+ Hits 4921 5231 +310
- Misses 9916 10194 +278
Continue to review full report at Codecov.
|
Thank you for reminding me of this! I've already merged the pr with devel branch. |
README.md
Outdated
| @@ -188,7 +188,12 @@ If you want to specify a structure as starting point for `init_bulk`, you may se | |||
| ```json | |||
| "from_poscar": true, | |||
| "from_poscar_path": "....../C_mp-47_conventional.POSCAR", | |||
| ``` | |||
| ``` | |||
There was a problem hiding this comment.
Please notice that the format is not correct.
There was a problem hiding this comment.
Sorry for that... I just fixed the bug.
dpgen/data/tools/hcp.py
Outdated
| def STRU_unit(latt): | ||
|
|
||
| box = gen_box() * latt/np.sqrt(2) | ||
| stru_data = {} | ||
| stru_data['cells'] = box | ||
| stru_data['atom_numbs'] = [2] | ||
| stru_data['coords'] = np.matmul(np.array([[0.0, 0.0, 0.0], [1./3, 1./3, 1./2]]), box) | ||
| stru_data['types'] = [0, 0] | ||
| return stru_data |
There was a problem hiding this comment.
Why not using dpdata for converting from POSCAR to STRU? You do not need to write support for abacus for each crystal type.
There was a problem hiding this comment.
😯 Sorry I did not realize it when I was writing the code. I will remove the STRU_unit codes and use the dpdata conversion code.
There was a problem hiding this comment.
I've removed the STRU_unit functions and replaced them with poscar_unit and dpdata conversion.
README.md
Outdated
| | relax_kpt | String | "....../KPT" | Path of `KPT` file for relaxation in stage 1. Only useful if `init_fp_style` is "ABACUS". | ||
| | md_kpt | String | "....../KPT" | Path of `KPT` file for MD simulations in stage 3. Only useful if `init_fp_style` is "ABACUS". |
There was a problem hiding this comment.
It is not good idea to explicitly set the number of kpoint, because the k-mesh should be adjusted according to the size of the system. For example different super_cell settings needs different number of kpoints.
A better and recommended way of setting kmesh is to set kspacing. you can check auto_test.lib.vasp.make_kspacing_kpoints to make a kmesh from a given kspacing.
There was a problem hiding this comment.
Sorry for the delay. I did not receive the notifications orz. I will solve the two issues as suggested.
There was a problem hiding this comment.
I have added relax_kspacing and md_kspacing in param.json for dpgen to generate the KPT files. I also reserved the previous relax_kpt and md_kpt parameters. Both methods are available.
dpgen/data/tools/diamond.py
Outdated
| def STRU_unit(latt): | ||
|
|
||
| box = gen_box() * latt | ||
| stru_data = {} | ||
| stru_data['cells'] = box | ||
| stru_data['atom_numbs'] = [2] | ||
| stru_data['coords'] = np.matmul(np.array([[0.125, 0.125, 0.125], [0.875, 0.875, 0.875]]), box) | ||
| stru_data['types'] = [0, 0] | ||
| return stru_data |
There was a problem hiding this comment.
This method should not be needed.
No description provided.