[ase-users] Writing of the trajectory file for optmization on a big endian architecture

[Torstein Fjermestad]

Dear Jens Jørgen Mortensen,

I replaced the aff.py file with the new version you gave me.  Thereafter,
I ran the command
"python setup.py test" which returned no errors. I think the big endian
issue is solved now. However, I still have some problems related to the
optimization of the structure. I will describe these in a new thread.

Regards,
Torstein Fjermestad



On Fri, Jul 3, 2015 at 5:42 PM, Torstein Fjermestad <tfjermestad at gmail.com>
wrote:

> Dear ASE users,
>
> I am trying to do a geometry optmization with ASE. In order to monitor the
> progress of the calculation, I would like to write the trajectory file. To
> do this, I write the following:
>    BFGSLineSearch(trajectory='filename.traj')
>
>
> With the trajectory file specified, the calculation stops after a few
> seconds with the following error message:
>
> File "ase/ase/io/aff.py", line 102, in __init__
>     assert np.little_endian  # deal with this later
> AssertionError
>
> I am running on a machine with big endian architecture so I think the
> error is related to that.
>
> A simple workaround of this problem is simply to set trajectory=None. This
> is, however, not satisfactory because I want to know how the structure
> optimizes; not just the final structure.
>
> Does anyone know of a solution to this problem?
>
> Thanks in advance for your help.
>
> Regards,
> Torstein Fjermestad
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/ase-users/attachments/20150824/960a868f/attachment.html>