[ase-users] ase-db output limit and sorting

[Tristan Maxson]

To be clear,  you are saying that ase-db discards extra results (past the
limit) BEFORE sorting rather than after?...This seems like a bug if so.  I
will look into it and see if its easily fixable if this is the case.

Thank you,
Tristan Maxson

On Fri, Oct 24, 2014 at 6:52 AM, Peter Bjerre Jensen <pbjen at dtu.dk> wrote:

>  Hi
>
> According to the help of ase-db (ase-db --help) one can specify the number
> of entries to show with the keyword limit; however I do not think it works
> very well in combination with the sort function.
>
> For example I have created a database following the ase-ga toturial (
> https://wiki.fysik.dtu.dk/ase/tutorials/ga/ga_fcc_alloys.html ) and
> generated a database where the 5 best structures according to raw_score are
> these:
>
> id|age|user |formula |pbc|volume|   mass|   hof|raw_score
> 153|22s|pbjen|Ni2Pt2  |---| 1.000|507.547|-0.107|    0.107
> 127|34s|pbjen|Cu2Pt2  |---| 1.000|517.252|-0.110|    0.110
>  75|54s|pbjen|Cu2Pt2  |---| 1.000|517.252|-0.110|    0.110
> 181| 9s|pbjen|Cu2Pt2  |---| 1.000|517.252|-0.110|    0.110
> 139|29s|pbjen|Ni3Pt   |---| 1.000|371.160|-0.118|    0.118
>
> This was obtained by using "ase-db fcc_alloys.db --limit 0
> -c+hof,raw_score -s raw_score" to show all entries and sort by
> raw_score(=-hof); which should be as high as possible for the best
> candidates.
>
> However, if I now do "ase-db fcc_alloys.db --limit 5 -c+hof,raw_score -s
> raw_score" I obtain the top of the list (which are the unrelaxed structures
> from the tutorial with no defined raw_score):
>
> id| age|user |formula |pbc|volume|   mass|hof|raw_score
>  1|248s|pbjen|        |---| 1.000|  0.000|   |
>  2|248s|pbjen|AgAl2Pt |---| 1.000|356.911|   |
>  3|248s|pbjen|Ag3Cu   |---| 1.000|387.150|   |
>  4|247s|pbjen|AgAlAuCu|---| 1.000|395.362|   |
>  5|247s|pbjen|AgAlAuPd|---| 1.000|438.236|   |
>
> I think this is very unpractical, and I think it would make most sense to
> show the last entries if one specifies a limit.
>
> I think a simple solution could be to add an extra keyword specifying the
> sorting to be reverse?
>
> Best regards
> Peter
>
>              Peter Bjerre Jensen   PhD Student   DTU Energy Conversion
>
>
> Technical University of Denmark
>     Department of Energy Conversion and Storage   Fysikvej, Building 307
> 2800 Kgs. Lyngby   Direct +45 45253204   pbjen at dtu.dk   www.ecs.dtu.dk
>
>
>
> --
> Peter Bjerre Jensen
> Gyngemose Parkvej 6C ST TH
> 2860 Søborg
> Tlf. 21 26 13 90
>
>
> _______________________________________________
> ase-users mailing list
> ase-users at listserv.fysik.dtu.dk
> https://listserv.fysik.dtu.dk/mailman/listinfo/ase-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/ase-users/attachments/20141024/32ca4da2/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: DTU_email_logo_01.gif
Type: image/gif
Size: 1055 bytes
Desc: not available
URL: <http://listserv.fysik.dtu.dk/pipermail/ase-users/attachments/20141024/32ca4da2/attachment.gif>