This is the user forum of the DFT FLAPW code FLEUR.
It is meant to be used as a place to asks questions to other users and the developers, to provide feedback and suggestions.
The documentation of the code can be found at the FLEUR homepage.
[b][/b]
[i][/i]
[u][/u]
[code][/code]
[quote][/quote]
[spoiler][/spoiler]
[url][/url]
[img][/img]
[video][/video]
Smileys
smile
smile2
spook
alien
zunge
rose
shy
clown
devil
death
sick
heart
idee
frage
blush
mad
sad
wink
frown
crazy
grin
hmm
laugh
mund
oh
rolling_eyes
oh2
shocked
cool
[pre][/pre]
Farben
[rot][/rot]
[blau][/blau]
[gruen][/gruen]
[orange][/orange]
[lila][/lila]
[weiss][/weiss]
[schwarz][/schwarz]
JensB
Posts: 8 | Last online: 06.30.2021
Name
Jens Broeder
Date registered
05.18.2021
Sex
not specified
    • JensB has written a new post "hdf" 06.02.2021

      Which .hdf file do you mean? There are several. And what is too large?

      In general there are tools to delete certain entries from hdf files and create a new file. If you want to store certain things in there longterm.
      In case you mean the cdn.hdf? Since this contains the full charge density history and scales with the system size this is usually the largest.
      If you check the command line options of fleur with fleur -h there is an option '-last_extra' which will create a separate cdn_last.hdf file containing only the last charge denisty, which is usual the only thing store from the density.

      If you mean the banddos.hdf file, this depends on your number of kpoints, energy window, how many projections and so on.

    • JensB has written a new post "波段输出" 06.01.2021

      To modify the plot you can also play with

      markersize_min (default 0.5)
      markersize_scaling (default 5.0)
      alpha (default 1.0)?

    • JensB has written a new post "波段输出" 06.01.2021

      I checked for a test bandstructure of Fe it works, see attached files:


      from masci_tools.io.parsers.hdf5 import HDF5Reader
      from masci_tools.io.parsers.hdf5.recipes import FleurBands

      from masci_tools.vis.fleur import plot_fleur_bands

      filepath='banddos.hdf'

      with HDF5Reader(filepath) as h5reader:
      data, attributes = h5reader.read(recipe=FleurBands)

      # We are interested in each state's d-projection in the MT sphere of the 1st atom type.
      weightName = "MT:1d"
      # Some other weights:
      weightName1 = "MT:1s" # s-projection in the 1st atom type's MT sphere.
      weightName2 = "MT:1p" # p-projection in the 1st atom type's MT sphere.


      # Plot the bandstructure and save to a file bandstructure.png
      # If we just want a unweighted bandstrcture, we can omitt the weight argument here
      ax = plot_fleur_bands(data, attributes, weight=weightName, limits={'y': (-10, 5)}, show=False, save_plots=True, saveas='bandstr1', cmap='Blues')
      ax = plot_fleur_bands(data, attributes, weight=weightName1, limits={'y': (-10, 5)}, axis=ax, show=False, save_plots=True, saveas='bandstr2', cmap='Reds')
      ax = plot_fleur_bands(data, attributes, weight=weightName2, limits={'y': (-10, 5)}, axis=ax, show=False, save_plots=True, saveas='bandstr3', cmap='Greens')


      since you get a matplotlib axis back, you can add text, or legends to it. you can also save it again. To get the same colors as the DOS, you may want to define the coloring of the DOS plot and make this in accordance of the color maps you use. If you type in a wrong color map or color, you will get an error message showing you all supported color maps. You could also define your own I assume.

      @gregor: it only makes sense if you have large differences in weights. Also maybe you want to highlight a defect state or something else.

    • JensB has written a new post "波段输出" 05.27.2021

      I just saw that this is indeed part of an open issue: https:// github.com/JuDFTteam/masci-tools/issues/32
      Feel free to contribute :-)

    • JensB has written a new post "波段输出" 05.27.2021

      Hi,
      If I understand correctly, you complain that the default colors of the DOS plot and Band structure plot are not consistent and that only one weight is visualized at a time and the default color is a blue color map.

      For the plot source code look at (remove spaces):
      (https:// github.com/JuDFTteam/masci-tools/blob/291bbbdedd3d99d4eeda2a92c000d3d7e178f626/masci_tools/vis/fleur.py
      and https:// github.com/JuDFTteam/masci-tools/blob/291bbbdedd3d99d4eeda2a92c000d3d7e178f626/masci_tools/vis/plot_methods.py#L1769
      )
      From the source code I get that there is a 'cmap' key you can provide, which you can try but it looks pretty hard coded to me.
      If this does not work open an issue on the masci-tools repo.
      Also plotting several weights in one go may be a good feature request.

      So far you might be able to do it like this:
      All plot function (at least the static ones without bokeh) return an axis and take an axis as input so that one can plot further plots in them.
      So to plot several you can do the following (I have not tried it):

      from masci_tools.io.parsers.hdf5 import HDF5Reader
      from masci_tools.io.parsers.hdf5.recipes import FleurBands

      from masci_tools.vis.fleur import plot_fleur_bands

      filepath='banddos.hdf'

      with HDF5Reader(filepath) as h5reader:
      data, attributes = h5reader.read(recipe=FleurBands)

      # We are interested in each state's d-projection in the MT sphere of the 1st atom type.
      weightName = "MT:1d"
      # Some other weights:
      weightName1 = "MT:1s" # s-projection in the 1st atom type's MT sphere.
      weightName2 = "MT:1p" # p-projection in the 1st atom type's MT sphere.
      weightName3 = "MT:1f" # f-projection in the 1st atom type's MT sphere.
      weightName4 = "MT:2s" # s-projection in the 2nd atom type's MT sphere (if available).

      # Plot the bandstructure and save to a file bandstructure.png
      # If we just want a unweighted bandstrcture, we can omitt the weight argument here
      ax = plot_fleur_bands(data, attributes, weight=weightName, limits={'y': (-10, 5)}, show=False, save_plots=True)# cmap='blues'
      ax = plot_fleur_bands(data, attributes, weight=weightName1, limits={'y': (-10, 5)}, axis=ax, show=False, save_plots=True)# cmap='reds'
      ax = plot_fleur_bands(data, attributes, weight=weightName2, limits={'y': (-10, 5)}, axis=ax, show=False, save_plots=True)# cmap='greens'
      ax = plot_fleur_bands(data, attributes, weight=weightName3, limits={'y': (-10, 5)}, axis=ax, show=False, save_plots=True)# cmap='blues'
      plot_fleur_bands(data, attributes, weight=weightName4, limits={'y': (-10, 5)}, axis=ax, show=False, save_plots=True)# cmap='blues'

    • JensB has written a new post "Geometry extraction?" 05.25.2021

      Hi Terry,


      The way you can/we do this is to go over the out.xml bzw, the inp.xml information. During relaxations everytime you have a new struct-relax.xsf you also have a new inp.xml. We need the atomic groups information to know to which atoms the forces belong to. The symmetries alone won't help you much. Therefore for a real trajectory, the struct-relax.xsf alone won't help you.

      The fleur parsers in masci-tools have parsers for fleur files see (remove the space in the link):
      https:// masci-tools.readthedocs.io/en/latest/user_guide/fleur_parser.html#parser-for-the-fleur-inp-xml-file

      This includes functions 'get_relaxation_information' and 'get_structure_data' which gets you besides the structure (geometry), the atom group information and what is relaxed and what not. From this you can easily create a new ASE, aiida, or pymatgen structure object. The forces we also parse. This is still a bit low level, but we do not have yet implemented putting that back together to a full trajectory.

      @Gregor: actually extraction a trajectory of the positions together with the forces is an open todo issue for the relaxation worklfow in aiid-fleur. For AiiDA the tricky thing will be to match the fleur atomtypes back to the original positions (pre inpgen).

      @Terry: if you like to contribute you could implement putting this information (forces and structure) together to a full real trajectory in python using the information returned from the parsers, given a list of out.xml files and ggf force files. We would help you of course.

    • JensB has written a new post "Long run or multiple restarts?" 05.20.2021

      Hi Terry,
      To add to Gregors answer: I think in the tutorials itmax was really small and convergence parameters were not so strict, just that things run fast.

      Per default we use a itmax of 30 or 80. Depending on the system sizes and how much resources you would waste if the systems do not converge.
      From our High-throughput projects for small unit cells we know that around 60% are converged within 30 iterations, 80% within 80 iterations. Everything that is still converging beyond this is converging slowly, stuck, or not at all. Of course this strongly depends on the system and which features of fleur you use. Magnetic calculations need more iterations and LDA+U calculations even more, simple because more quantities have to reach convergence. And of course there is a dependence on the mixing scheme deployed.

      Deleting the mixing history from time to time (i.e after 30, or 80 iterations) may help to get a system which was 'stuck' before to converge nicely again.

      Overall, one wants to minimize restarts of course, because they amount to more data and more data transfer, but at the same time you want to minimize the computational time in systems not converging, because else you end up spending 80% or more of your CPU budget on these. While as Gregor pointed out, systems which do not move at all are easy to detect, other convergence behavior may be more challenging.
      The long term ideal solution to this is to improve our automatic workflows to detect systems not (slow) converging or stuck early on and implement solutions how to deal with them or stop them.

    • JensB has written a new post "XPS spectra calculated by FLEUR" 05.18.2021

      Hi Hidemaru,

      Thanks for you interested in this. To add to Daniels answer:
      The theoretical spectra are formed by placing peakfuntions at the calculated shifted core-level positions.
      The type of the peakfunction (voigts or some asymmetric profiles) depends on the element and core-level. The peakfunction parameters are fitted within the evaluation, since besides the influence from the experimental apparatus we cannot calculate the lifetime broadenings, screening and other things accurately within simple DFT.

      What is currently a patent application is the 'data-driven chemical evaluation procedure' of XPS spectra.
      We have a small program to do such evaluations. It is not public (at this time), but you can request it here (remove space in link):
      www. inventionstore.de/angebot/5654/

      which is probably free if you are working in academia.

      For further references how it is done and what is currently possible I refer you to my PhD thesis (remove space in link):
      https:// juser.fz-juelich.de/record/891865/
      (Papers on this are not out yet)

      If you have further questions feel free to ask.
      Best regards and sorry for the late reply,
      Jens

Recipient
JensB
Subject:


text:

Sign up, to leave a comment


Xobor Einfach ein eigenes Xobor Forum erstellen
Datenschutz