[ase-users] Re: ASE for post processing ...

[Chaitanya Krishna]

Hi Jonas,

I get an error. This is what I get.

chaitanya at drona:~/tmp/asetest/ag01> ls
ag.py  CONTCAR  OUTCAR  read_force.py
chaitanya at drona:~/tmp/asetest/ag01> python read_force.py
Traceback (most recent call last):
  File "read_force.py", line 4, in <module>
    calc = Vasp(restart=True)
  File "/usr/local/lib/python2.5/site-packages/ase/calculators/vasp6.py",
line 115, in __init__
    self.atoms = ase.io.read('CONTCAR', format='vasp')
  File "/usr/local/lib/python2.5/site-packages/ase/io/__init__.py",
line 106, in read
    return read_vasp(filename)
  File "/usr/local/lib/python2.5/site-packages/ase/io/vasp.py", line
26, in read_vasp
    from ase.constraints import FixAtoms, fix_scaled
ImportError: cannot import name fix_scaled

By the way, why don't we set the vasp binary also in an environment
variable, instead of meddling with the code.

Cheers,
Chaitanya

On Tue, Sep 16, 2008 at 1:53 PM, Jonas Björk <j.bjork at liverpool.ac.uk> wrote:
> Hi Chaitanya,
>
> I have made a small script which returns the forces from an already done
> vasp calculation. This is done in an indirect way and I agree that the
> post processing should also be directly accessed in the interface. I
> will do some work implementing it.
>
> I attach the script (read_force.py) and the newest version of the
> interface (vasp6.py, just put in the ase/calculators directory) which
> you'll need to run the script. Please let me know if you have any
> problems running it.
>
> Best regards,
> Jonas
>
>
> On Tue, 2008-09-16 at 13:16 +0200, Chaitanya Krishna wrote:
>> Hi Jonas,
>>
>> Please find the CONTCAR attached.
>>
>> It will be of much greater if we can also have the post processing
>> features for VASP.
>>
>> Regards,
>> Chaitanya.
>>
>
>> On Tue, Sep 16, 2008 at 1:13 PM, Jonas Björk <j.bjork at liverpool.ac.uk> wrote:
>> > Hi Chaitanya,
>> >
>> > This is possible and shouldn't be to complicated to do. However as the
>> > interface looks now there's no direct way to do it. This will however be
>> > implemented at some point.
>> >
>> > So at the moment you'll need to use some indirect way of doing it. Could
>> > you attach the CONTCAR/POSCAR corresponding to the OUTCAR file, and I'll
>> > put a simple script taking out the forces for the atoms?
>> >
>> > Cheers,
>> > Jonas
>> >
>> >
>> > On Tue, 2008-09-16 at 11:45 +0200, Chaitanya Krishna wrote:
>> >> Hi Jonas,
>> >>
>> >> Just need a clarification on a few things about the ASE module for VASP.
>> >>
>> >> Suppose I want to use the vasp module from ase purely for post
>> >> processing and not to actually steer the calculation. That is what I
>> >> would like to do would be something like,
>> >>
>> >> - Read OUTCAR and define the system
>> >> - Get some properties that I need
>> >> - Do some further processing on what I get from OUTCAR.
>> >>
>> >> instead of the examples that we have for ASE like,
>> >>
>> >> - Define system
>> >> - Run calculation
>> >> - Get properties.
>> >>
>> >> If the above is possible can you send me a sample script, say just get
>> >> the forces on the atoms from OUTCAR. I am attaching an OUTCAR that I
>> >> am using right now. Say, I want all the forces on the atoms. Can you
>> >> send me a sample script that can do this.
>> >>
>> >> Regards,
>> >> Chaitanya.
>> >
>