multislice

A Multislice interface to build decks, launch jobs and postprocess multislice simulations

Table of Contents

Example

Create a periodic simulation from data in 'Silicon/Si110', submit it and monitor its state :

multi=Multislice('Silicon/Si110',mulslice=False,NxNy=512,repeat=[6,6,40],Nhk=2)
p = multi.run()
multi.check_simu_state()

Once terminated, get the beams intensities as function of thickness, forcing to repostprocess in case it was already done in an older simulation :

multi.beam_vs_thickness(bOpt='f')

Multislice

Multislice(self, name, mulslice=False, tail='', data=None, TDS=False, T=300, n_TDS=16, keV=200, repeat=[2, 2, 1], NxNy=[512, 512], slice_thick=1.0, hk=[(0, 0)], Nhk=0, prev=None, opt='d', fopt='w', ppopt='uwP', v=1, ssh=None)

DATA PARAMETERS :

  • name : path to the simulation folder
  • tail : trailing string in the naming convention of outputs
    • name pattern = name_tail_program_name
  • data : data file names
    • if None and mulslice simu : all *[a-zA-Z].dat and stacked alphabetically
    • if None and autoslic simu : first *.xyz found
    • Otherwise paths to filename(s) of the data files to use (must be in name)

MULTISLICE simulation parameters :

  • mulslice : False(autoslic), True(mulslice)=>periodic
  • keV : float - wavelength in keV
  • repeat : 3-int-list - super cell size in the x,y,z directions
  • NxNy : 2-int-list or int : sampling in x and y (same sampling in x and y if only int provided)
  • slice_thick : float - slice thickness (A)
  • hk : list of tuple - beams to record as function of depth
  • Nhk: int - set hk on a grid of multiples of the supercell size. Prevails over hk argument if >0. default(0)

Thermal parameters:

  • TDS : bool - include thermal vibrations (as dictated by wobble parameter in .dat )
  • T : Temperature in K if TDS is True
  • nTDS : Number of runs to average over

RUN/OUTPUT options:

  • v : verbose - str or bool(all if True) or int(verbose level)
    • n(naming pattern),c(cell params),t(thickness),r(run cmd)
    • d(data),D(Decks),R(full run cmd)
  • opt : d(save_deck), s(save_obj), r(do_run), p(do_pp)
  • fopt : f(force rerun), w(warn rerun). If the simulation was previously completed :
    • 'w' is on (default case) : the user will be asked to confirm whether he wants to rerun it,
    • 'w' is off (fopt='') : The simulation is not rerun
    • 'f' in fopt : The simulation is rerun
  • ppopt : u(update),w(wait), I(image) B(beam) P(pattern) A(azim_avg)
  • ssh : ip address or alias of the machine on which to run the job

make_decks

Multislice.make_decks(self, save=True, datpath=None, prev=None)

create the decks from the information provided

save

Multislice.save(self)

save this multislice object

run

Multislice.run(self, v=1, fopt='w', ssh_alias=None, hostpath=None)

run the simulation with temsim - fopt : f(force rerun), w(warn ask rerun already done) - ssh : name of the host to run the job

postprocess

Multislice.postprocess(self, ppopt='uwP', ssh_alias='', tol=0.0001, figpath=None, opt='p')

Performs postprocessing with predefined options :

  • ppopt:u(update), w(wait till done) I(image), B(beam) P(pattern)
  • figpath : Directory to place the figures if saving with automatic naming (default datpath)

image

Multislice.image(self, opt='I', cmap='jet', **kwargs)

Displays the 2D image out of simulation - opt : I(intensity)

beam_vs_thickness

Multislice.beam_vs_thickness(self, bOpt='', iBs=[], tol=0.01, **kwargs)
  • bOpt : O(include Origin),p(print Imax),n or f(force new)
  • kwargs : see help(plot_beam_thickness)

get_beams

Multislice.get_beams(self, iBs=[], tol=0.01, bOpt='')

get the beams as recorded during:

  • iBs : selected beam indices : default=Ibeams.max()>tol
  • bOpt : O(include Origin),p(print Imax),n or f(force new) hk,t,re,im,Ib = beams

pattern

Multislice.pattern(self, Iopt='Incsl', out=0, tol=0.0001, Nmax=None, rings=[], **kwargs)

Displays the 2D diffraction pattern out of simulation - Iopt : I(intensity), c(crop I[r]<tol), n(normalize), s(fftshift), l(logscale) - Nmax : crop display beyond Nmax - rings : list or array - of resolutions for rings to display - kwargs : see stddisp returns : [qx,qy,I]

azim_avg

Multislice.azim_avg(self, out=0, **kwargs)

Display the average azimuthal diffraction pattern intensities

Multislice.print_datafiles(self, data=None)

show data files *.dat or .xyz depending on mulslice

Multislice.print_decks(self, decks=None)

show decks *.in

Multislice.print_log(self)

print the logfile of running multislice .log

check_simu_state

Multislice.check_simu_state(self, ssh_alias=None, v=False)

see completion state of a simulation

sweep_var

sweep_var(name, param, vals, df=None, ssh='', tail='', do_prev=0, **kwargs)

runs a set of similar simulations with one varying parameter - name : path to the simulation folder - param,vals : the parameters and values to sweep - df : - pd.Dataframe to update(since parsed as a reference) - int create and save the new dataframe if 1 - do_prev : Used for iterative fourier transform - kwargs : see help(Multislice)

make_xyz

make_xyz(name, pattern, lat_vec, n=[0, 0, 1], fmt='%.4f', dopt='scp')

Creates the.xyz file from a given compound and orientation - name : Full path to the file to save - pattern : Nx6 ndarray - Z,x,y,z,occ,wobble format - lat_vec : 3x3 ndarray - lattice vectors [a1,a2,a3] - n : beam direction axis - dopt : p(print file)