Re: [Phonopy-users] Group Velocity

[Alejandro R. <ar...@ui...>]

Hi Togo,

Once again, thanks a lot, it works!

The interface for group velocity would be great!

Best,

Alejandro

On Fri, February 22, 2013 4:45 am, Atsushi Togo wrote:
> Hi Alejandro,
>
> Probably only I may be able to reply this because this is not
> comprehensive.
> You may need to set calculate_full_force_constants=True at
> phonon.set_post_process([[1, 0., 0.], [0., 1, 0.], [0., 0.,
> 1]],set_of_forces, calculate_full_force_constants=True)
>
> I'm now implementing interface for group velocity.
>
> Togo
>
> On Fri, Feb 22, 2013 at 6:47 PM, "Alejandro Rébola" <ar...@ui...>
> wrote:
>> Hi Sebastian,
>>
>> Thank you for your answer. I've been playing with it, but now I have new
>> question. I adapted the Si.py script in order to print the frequencies
>> for
>> my system. It works, and I get some frequencies that are the same I get
>> by
>> setting the mesh.conf file and running phonopy. However, if now I set
>> the
>> FC_SYMMETRY=10 (or to any other value) on mesh.conf and do the same (or
>> what I think it should be the same) in the script by adding a line:
>>
>> phonon.symmetrize_force_constants(10)
>>
>> then the frequencies I get are different. Do you know what could be
>> going
>> on? I attach the script and the mesh.conf file.
>> Thank you in advance,
>>
>> Alejandro
>>
>> On Thu, February 21, 2013 3:55 pm, Sebastian Christensen wrote:
>>> Hi Alejandro
>>>
>>> It sounds like you are using the command line interface to phonopy. I
>>> think you should use phonopy as a python module instead. See the manual
>>> on
>>> how to. Or even better study examples/<mailto:examples~@:/>Si/Si.py in
>>> your phonopy folder. My experience is that it is pretty easy to do even
>>> though you are not experienced python-user. And you have much more
>>> freedom
>>> to do what you want/need!
>>>
>>> By using phonopy as python-module you get the eigenvalues of your
>>> q-mesh
>>> in a numpy-array (which is much, much smaller than 1gb).
>>>
>>>
>>> Good luck with it!
>>>
>>> Cheers
>>> Sebastian
>>>
>>> Sent from Windows Mail
>>>
>>> From: Alejandro Rébola
>>> Sent: ‎21‎ ‎February‎ ‎2013
>>> ‎22‎:‎42
>>> To: Atsushi Togo
>>> CC: pho...@li...
>>> Subject: Re: [Phonopy-users] Group Velocity
>>>
>>> Hi Togo,
>>>
>>> Thanks, I'll try to test it. Right now, I'm using my own version that
>>> uses
>>> the grid generated by mesh.conf. However, I'm having a problem with the
>>> file sizes generated in yaml. To get converged results I need grids of
>>> the
>>> order of 40x40x40 q-points (I need the group velocities at each
>>> q-point).
>>> When I do this I get a ~ 1Gb file (my system has ~ 80 atoms in the unit
>>> cell, and I'm using a 1x1x1 supercell). However, if I also want the
>>> eigenvectors the output file becomes huge. Is there a way around this?
>>> For
>>> instance printing plain text outputs?
>>> Thank you,
>>>
>>> Alejandro
>>>
>>> On Thu, February 21, 2013 3:57 am, Atsushi Togo wrote:
>>>> Hi,
>>>>
>>>> It's under testing. Hopefully I will release soon, but if you want to
>>>> test it right now, please tell me. But I need test report.
>>>>
>>>> Togo
>>>>
>>>> On Thu, Feb 21, 2013 at 5:27 PM, "Alejandro Rébola"
>>>> <ar...@ui...>
>>>> wrote:
>>>>> Dear users,
>>>>>
>>>>> Does anybody know how to calculate the group velocity at an arbitrary
>>>>> q-point with Phonopy? I've been doing it by using the output
>>>>> contained
>>>>> in
>>>>> the mesh.yaml file, but it's rather messy because it requires to
>>>>> interpolate if the q point does not belong to the mesh. I see there's
>>>>> a
>>>>> class called GroupVelocity that apparently gets the group velocity at
>>>>> an
>>>>> arbitrary q-point, but I'm not sure how it works. Can anybody help
>>>>> me?
>>>>> Thank you in advance,
>>>>>
>>>>> Alejandro
>>>>>
>>>>>
>>>>> ------------------------------------------------------------------------------
>>>>> Everyone hates slow websites. So do we.
>>>>> Make your web apps faster with AppDynamics
>>>>> Download AppDynamics Lite for free today:
>>>>> http://p.sf.net/sfu/appdyn_d2d_feb
>>>>> _______________________________________________
>>>>> Phonopy-users mailing list
>>>>> Pho...@li...
>>>>> https://lists.sourceforge.net/lists/listinfo/phonopy-users
>>>>
>>>>
>>>>
>>>> --
>>>> Atsushi Togo
>>>> http://atztogo.users.sourceforge.net/
>>>> atz...@gm...
>>>>
>>>>
>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Everyone hates slow websites. So do we.
>>> Make your web apps faster with AppDynamics
>>> Download AppDynamics Lite for free today:
>>> http://p.sf.net/sfu/appdyn_d2d_feb
>>> _______________________________________________
>>> Phonopy-users mailing list
>>> Pho...@li...
>>> https://lists.sourceforge.net/lists/listinfo/phonopy-users
>>>
>>
>> ------------------------------------------------------------------------------
>> Everyone hates slow websites. So do we.
>> Make your web apps faster with AppDynamics
>> Download AppDynamics Lite for free today:
>> http://p.sf.net/sfu/appdyn_d2d_feb
>> _______________________________________________
>> Phonopy-users mailing list
>> Pho...@li...
>> https://lists.sourceforge.net/lists/listinfo/phonopy-users
>>
>
>
>
> --
> Atsushi Togo
> http://atztogo.users.sourceforge.net/
> atz...@gm...
>
> ------------------------------------------------------------------------------
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_d2d_feb
> _______________________________________________
> Phonopy-users mailing list
> Pho...@li...
> https://lists.sourceforge.net/lists/listinfo/phonopy-users
>