[ase-users] Save "forces" from GULP/QuasiNewton

[Dong gi Kang]

Dear ASE users,

I would like to save “forces”, “positions” in every step of optimisation process from GULP calculations.
However, it only saves “positions” of atoms, no energy and forces even though force can be calculated.

(One atom calculation (gulp) with the same script gives an error: PropertyNotImplementedError: forces not present in this calculation)

With the command I could save force component from FHI-aims calculation but not from GULP.
Do I have to add more command to get the force component?


This is the saved .xyz data that converted from .traj file.
4
Properties=species:S:1:pos:R:3 pbc="F F F"
Zn       2.99999999      -0.00000001       0.00000001
Zn       0.00000000       3.00000000       0.00000000
O       -0.00000008      -0.00000004       3.00000002
O       -2.00000003      -0.00000005       0.00000005

  =================
~~~~#This is a part of the code that I used
    cluster.calc = calc
    dynamics = QuasiNewton(cluster) #trajectory='g_ZnO_n1.traj')

    traj = Trajectory('g_ZnO_n1.traj', 'w', cluster)
    dynamics.attach(traj.write, interval=1)             # <----- save results every one fs
    def forces():
        a = cluster.get_forces()
        print(a)
    dynamics.attach(forces, interval=5)
    dynamics.run(steps=600, fmax=0.0001)


    dynamics.attach(forces, interval=5) produces
BFGSLineSearch:   86[598] 19:55:54      -48.088229*      22.3339
BFGSLineSearch:   87[603] 19:55:55      -48.088130*      22.3339
BFGSLineSearch:   88[608] 19:55:55      -48.088037*      22.3340
BFGSLineSearch:   89[613] 19:55:55      -48.087941*      22.3340
BFGSLineSearch:   90[618] 19:55:56      -48.087845*      22.3340
[[ -0.657599  -1.870524   6.558049]
[ -0.        -0.        -0.      ]
[ -0.639623   0.210503   4.824076]
[-17.48983  -13.84042    1.165193]]
BFGSLineSearch:   91[623] 19:55:57      -48.087749*      22.3341
BFGSLineSearch:   92[628] 19:55:57      -48.087655*      22.3341


Many thanks,

Dong-gi Kang

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/ase-users/attachments/20201218/a1cc2b68/attachment.htm>