Source code for pyprocar.scriptDosplot

"""
Created on May 17 2020
@author: Pedram Tavadze
"""
# from .elkparser import ElkParser
from .splash import welcome
from .doscarplot import DosPlot
from .vaspxml import VaspXML
from .lobsterparser import LobsterDOSParser
from .qeparser import QEDOSParser

import numpy as np
import matplotlib.pyplot as plt

# import matplotlib
plt.rcParams["mathtext.default"] = "regular"
# Roman ['rm', 'cal', 'it', 'tt', 'sf',
#        'bf', 'default', 'bb', 'frak',
#        'circled', 'scr', 'regular']
plt.rcParams["font.family"] = "Arial"
plt.rc("font", size=18)  # controls default text sizes
plt.rc("axes", titlesize=22)  # fontsize of the axes title
plt.rc("axes", labelsize=22)  # fontsize of the x and y labels
plt.rc("xtick", labelsize=22)  # fontsize of the tick labels
plt.rc("ytick", labelsize=22)  # fontsize of the tick labels
# plt.rc('legend', fontsize=22)    # legend fontsize
# plt.rc('figure', titlesize=22)  # fontsize of the figure title


[docs]def dosplot( filename="vasprun.xml", mode="plain", interpolation_factor=None, orientation="horizontal", spin_colors=None, colors=None, spins=None, atoms=None, orbitals=None, elimit=None, dos_limit=None, cmap="jet", linewidth=1, vmax=None, vmin=None, grid=False, savefig=None, title=None, plot_total=True, code="vasp", labels=None, items={}, ax=None, plt_show=True, verbose=True, ): """ This function plots the density of states in different formats Parameters ---------- filename : str, optional (default ``'vasprun.xml'``) The most important argument needed dosplot is **filename**. **filename** defines the path to `vasprun.xml` from the density of states calculation. If plotting is being carried out in the directory of the calculation, one does not need to specify this argument. e.g. ``filename='~/SrVO3/DOS/vasprun.xml'`` mode : str, optional (default ``'plain'``) **mode** defines the mode of the plot. This parameter will be explained in details with exmaples in the tutorial. options are ``'plain'``, ``'parametric'``, ``'parametric_line'``, ``'stack'``, ``'stack_orbitals'``, ``'stack_species'``. e.g. ``mode='stack'`` interpolation_factor : int, optional (default ``None``) Number of points in energy axis is multiplied by this factor and interpolated using cubic spline. e.g. ``interpolation_factor=3`` orientation : str, optional (default ``horizontal'``) The orientation of the DOS plot. options are ``'horizontal', 'vertical'`` e.g. ``orientation='vertical'`` spin_colors : list str or tuples, (optional ``spin_colors=['blue', 'red']``) **spin_colors** represent the colors the different spin ploarizations are going to be represented in the DOS plot. These colors can be chosen from any type of color acceptable by matplotlib(string,rgb,html). e.g. ``spin_colors=['blue','red']``, ``spin_colors=[(0, 0, 1),(1, 0,0 )]``, ``spin_colors=['#0000ff','#ff0000']`` .. caution:: If the calculation is spin polarized one has to provide two colors even if one is plotting one spin. I disregard this cation if using default. colors : list str or tuples, optional (default, optional) ``colors`` defines the color of plots filling the area under the curve of Total density of states. This is only important in the ``mode=stack``, ``mode=stack_species``, ``mode=stack_orbitals``. To have a better sense of this parameter refer to the stack plots of SrVO\ :sub:`3`\. These colors can be chosen from any type of color acceptable by matplotlib(string,rgb,html). e.g. ``colors=['red', 'blue', 'green', 'magenta', 'cyan']`` spins : list int, optional ``spins`` defines plotting of different spins channels present in the calculation, If the calculation is spin non-polorized the spins will be set by default to ``spins=[0]``. if the calculation is spin polorized this parameter can be set to 0 or 1 or both. e.g. ``spins=[0, 1]`` atoms : list int, optional ``atoms`` define the projection of the atoms in the Density of States. In other words it selects only the contribution of the atoms provided. Atoms has to be a python list(or numpy array) containing the atom indices. Atom indices has to be order of the input files of DFT package. ``atoms`` is only relevant in ``mode='parametric'``, ``mode='parametric_line'``, ``mode='stack_orbitals'``. keep in mind that python counting starts from zero. e.g. for SrVO\ :sub:`3`\ we are choosing only the oxygen atoms. ``atoms=[2, 3, 4]``, keep in mind that python counting starts from zero, for a **POSCAR** similar to following:: Sr1 V1 O3 1.0 3.900891 0.000000 0.000000 0.000000 3.900891 0.000000 0.000000 0.000000 3.900891 Sr V O 1 1 3 direct 0.500000 0.500000 0.500000 Sr atom 0 0.000000 0.000000 0.000000 V atom 1 0.000000 0.500000 0.000000 O atom 2 0.000000 0.000000 0.500000 O atom 3 0.500000 0.000000 0.000000 O atom 4 if nothing is specified this parameter will consider all the atoms present. orbitals : list int, optional ``orbitals`` define the projection of orbitals in the density of States. In other words it selects only the contribution of the orbitals provided. Orbitals has to be a python list(or numpy array) containing the Orbital indices. Orbitals indices has to be order of the input files of DFT package. The following table represents the indecies for different orbitals in **VASP**. +-----+-----+----+----+-----+-----+-----+-----+-------+ | s | py | pz | px | dxy | dyz | dz2 | dxz | x2-y2 | +-----+-----+----+----+-----+-----+-----+-----+-------+ | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | +-----+-----+----+----+-----+-----+-----+-----+-------+ ``orbitals`` is only relavent in ``mode='parametric'``, ``mode='parametric_line'``, ``mode='stack_species'``. e.g. ``orbitals=[1,2,3]`` will only select the p orbitals while ``orbitals=[4,5,6,7,8]`` will select the d orbitals. If nothing is specified pyprocar will select all the present orbitals. elimit : list float, optional Energy window limit asked to plot. ``elimit`` has to be a two element python list(or numpy array). e.g. ``elimit=[-2, 2]`` The default is set to the minimum and maximum of the energy window. dos_limit : list float, optional ``dos_limit`` defines the density of states axis limits on the graph. It is automatically set to select 10% higher than the maximum of density of states in the specified energy window. e.g. ``dos_limit=[0, 30]`` cmap : str , optional (default 'jet') The color map used for color coding the projections. ``cmap`` is only relevant in ``mode='parametric'``. a full list of color maps in matplotlib are provided in this web page. `https://matplotlib.org/2.0.1/users/colormaps.html <https://matplotlib.org/2.0.1/users/colormaps.html>`_ e.g. ``cmap='plasma'`` linewidth : str, optional (default 1) The line width with which the total DOS is ploted e.g. linewidth=2 vmax : float, optional The maximum value in the color bar. ``cmap`` is only relevant in ``mode='parametric'``. e.g. ``vmax=1.0`` vmin : float, optional The maximum value in the color bar. ``cmap`` is only relevant in ``mode='parametric'``. e.g. ``vmin=-1.0`` grid : bool, optional (default Flase) Defines If a grid is plotted in the plot. The entry should be python boolian. e.g. ``grid=True`` savefig : str , optional (default None) ``savefig`` defines the file that the plot is going to be saved in. ``savefig`` accepts all the formats accepted by matplotlib such as png, pdf, jpg, ... If not provided the plot will be shown in the interactive matplotlib mode. e.g. ``savefig='DOS.png'``, ``savefig='DOS.pdf'`` title : str, optional Defines the plot title asked to be added above the plot. If ``title`` is not defined, PyProcar will not add any title. e.g. ``title="Total Density of States SrVO_$3$"``. One can use LaTex format as well. plot_total : bool, optional (default ``True``) If the total density of states is plotted as well as other options. The entry should be python boolian. e.g. ``plot_total=True`` code : str, optional (default ``'vasp'``) Defines the Density Functional Theory code used for the calculation. The default of this argument is vasp, so if the cal is done in vasp one does not need to define this argumnet. e.g. ``code=vasp``, ``code=elk``, ``code=abinit`` labels : list str, optional ``labels`` define the legends plotted in defining each spin. e.g. ``labels=['Oxygen-Up','Oxygen-Down']``, ``labels=['Oxygen-'+r'$\\uparrow$','Oxygen-'+r'$\\downarrow$']`` Side means the string will be treated as raw string. This has to be used if LaTex formating is used. No default is used in the ``mode=plain``, ``mode=parametric``, ``mode=parametric_line``. In ``mode=stack``, `ack_species``, ``mode=stack_orbitals`` the labels are generated automatically based on the other parameters such as atoms and orbitals. items : dict, optional ``items`` is only relavent for ``mode='stack'``. stack will plot the items defined with stacked filled areas under curve. For clarification visit the examples in the tutorial. ``items`` need to be provided as a python dictionary, with keys being specific species and values being projections of ``orbitals``. The following examples can clarify the python lingo. e.g. ``items={'Sr':[0],'O':[1,2,3],'V':[4,5,6,7,8]}`` or ``items=dict(Sr=[0],O=[1,2,3],V=[4,5,6,7,8])``. The two examples are equivalent to each other. This will plot the following curves stacked on top of each other. projection of s orbital in Sr, projection of p orbitals in O and projection of d orbitals in V. The default is set to take every atom and every orbital. Which will be equivalent to ``mode='stack_species'`` ax : matplotlib ax object, optional ``ax`` is a matplotlib axes. In case one wants to put plot generated from this plot in a different figure and treat the output as a subplot in a larger plot. e.g. :: >>> # Creates a figure with 3 rows and 2 colomuns >>> fig, axs = plt.subplots(3, 2) >>> x = np.linspace(-np.pi, np.pi, 1000) >>> y = np.sin(x) >>> axs[0, 0].plot(x, y) >>> pyprocar.dosplot(mode='plain',ax=axs[2, 2]),elimit=[-2,2]) >>> plt.show() plt_show : bool, optional (default ``True``) whether to show the generated plot or skip to the saving. e.g. ``plt_show=True`` Returns ------- fig : matplotlib figure The generated figure ax : matplotlib ax object The generated ax for this density of states. If one chooses ``plt_show=False``, one can modify the plot using this returned object. e.g. :: >>> fig, ax = pyprocar.dosplot(mode='plain', plt_show=False) >>> ax.set_ylim(-2,2) >>> fig.show() """ if mode not in [ 'plain', 'parametric_line', 'parametric', 'stack_species', 'stack_orbitals', 'stack' ]: raise ValueError( "Mode should be choosed from ['plain', 'parametric_line','parametric','stack_species','stack_orbitals','stack']" ) if verbose: welcome() # Verbose section print("Script initiated") print("code : ", code) print("File name : ", filename) print("mode : ", mode) print("spins : ", spins) print("atoms list : ", atoms) print("orbs. list : ", orbitals) print("energy range : ", elimit) print("colormap : ", cmap) print("vmax : ", vmax) print("vmin : ", vmin) print("grid enabled : ", grid) print("savefig : ", savefig) print("title : ", title) total = plot_total code = code.lower() if orientation[0].lower() == 'h': orientation = 'horizontal' elif orientation[0].lower() == 'v': orientation = 'vertical' if code == "vasp": vaspxml = VaspXML(filename=filename, dos_interpolation_factor=interpolation_factor) dos_plot = DosPlot(dos=vaspxml.dos, structure=vaspxml.structure) if atoms is None: atoms = list(np.arange(vaspxml.structure.natoms, dtype=int)) if spins is None: spins = list(np.arange(len(vaspxml.dos.total))) if orbitals is None: orbitals = list( np.arange(len(vaspxml.dos.projected[0][0]), dtype=int)) if elimit is None: elimit = [vaspxml.dos.energies.min(), vaspxml.dos.energies.max()] elif code == "lobster": vaspxml = LobsterDOSParser( filename="DOSCAR.lobster", dos_interpolation_factor=interpolation_factor) dos_plot = DosPlot(dos=vaspxml.dos, structure=vaspxml.structure) if atoms is None: atoms = list(np.arange(vaspxml.structure.natoms, dtype=int)) if spins is None: spins = list(np.arange(len(vaspxml.dos.total))) if orbitals is None: orbitals = list( np.arange(len(vaspxml.dos.projected[0][0]), dtype=int)) if elimit is None: elimit = [vaspxml.dos.energies.min(), vaspxml.dos.energies.max()] elif code == "qe": vaspxml = QEDOSParser(nscfin="nscf.in", pdosin="pdos.in", outfile="scf.out", dos_interpolation_factor=interpolation_factor) dos_plot = DosPlot(dos=vaspxml.dos, structure=vaspxml.structure) if atoms is None: atoms = list(np.arange(vaspxml.structure.natoms, dtype=int)) if spins is None: spins = list(np.arange(len(vaspxml.dos.total))) if orbitals is None: orbitals = list( np.arange(len(vaspxml.dos.projected[0][0]), dtype=int)) if elimit is None: elimit = [vaspxml.dos.energies.min(), vaspxml.dos.energies.max()] if mode == "plain": fig, ax1 = dos_plot.plot_total( spins=spins, spin_colors=spin_colors, ax=ax, orientation=orientation, labels=labels, linewidth=linewidth, ) elif mode == "parametric_line": if not total: fig, ax1 = dos_plot.plot_parametric_line( atoms=atoms, spins=spins, orbitals=orbitals, spin_colors=spin_colors, ax=ax, orientation=orientation, labels=labels, linewidth=linewidth, ) else: fig, ax1 = dos_plot.plot_total( spins=spins, spin_colors=[(0, 0, 0), (0, 0, 0)], ax=ax, orientation=orientation, linewidth=linewidth, ) _, ax1 = dos_plot.plot_parametric_line( atoms=atoms, spins=spins, orbitals=orbitals, spin_colors=spin_colors, ax=ax1, orientation=orientation, labels=labels, linewidth=linewidth, ) elif mode == "parametric": if not total: fig, ax1 = dos_plot.plot_parametric( atoms=atoms, spins=spins, orbitals=orbitals, spin_colors=spin_colors, cmap=cmap, vmin=vmin, vmax=vmax, elimit=elimit, ax=ax, orientation=orientation, labels=labels, ) else: fig, ax1 = dos_plot.plot_parametric( atoms=atoms, spins=spins, orbitals=orbitals, spin_colors=spin_colors, cmap=cmap, vmin=vmin, vmax=vmax, elimit=elimit, ax=ax, orientation=orientation, labels=labels, ) _, ax1 = dos_plot.plot_total( spins=spins, spin_colors=[(0, 0, 0), (0, 0, 0)], ax=ax1, orientation=orientation, linewidth=linewidth, ) elif mode == "stack_species": if not total: fig, ax1 = dos_plot.plot_stack_species( spins=spins, orbitals=orbitals, spin_colors=spin_colors, colors=colors, elimit=elimit, figsize=(12, 6), ax=ax, orientation=orientation, ) else: fig, ax1 = dos_plot.plot_stack_species( spins=spins, orbitals=orbitals, spin_colors=spin_colors, colors=colors, elimit=elimit, figsize=(12, 6), ax=ax, orientation=orientation, ) _, ax1 = dos_plot.plot_total( spins=spins, spin_colors=[(0, 0, 0), (0, 0, 0)], ax=ax1, orientation=orientation, linewidth=linewidth, ) elif mode == "stack_orbitals": if not total: fig, ax1 = dos_plot.plot_stack_orbitals( spins=spins, atoms=atoms, spin_colors=spin_colors, colors=colors, elimit=elimit, figsize=(12, 6), ax=ax, orientation=orientation, ) else: fig, ax1 = dos_plot.plot_stack_orbitals( spins=spins, atoms=atoms, spin_colors=spin_colors, colors=colors, elimit=elimit, figsize=(12, 6), ax=ax, orientation=orientation, ) _, ax1 = dos_plot.plot_total( spins=spins, spin_colors=[(0, 0, 0), (0, 0, 0)], ax=ax1, orientation=orientation, linewidth=linewidth, ) elif mode == "stack": if not total: fig, ax1 = dos_plot.plot_stack( items=items, spins=spins, spin_colors=spin_colors, colors=colors, elimit=elimit, figsize=(12, 6), ax=ax, orientation=orientation, ) else: fig, ax1 = dos_plot.plot_stack( items=items, spins=spins, spin_colors=colors, colors=colors, elimit=elimit, figsize=(12, 6), ax=ax, orientation=orientation, ) _, ax1 = dos_plot.plot_total( spins=spins, spin_colors=[(0, 0, 0), (0, 0, 0)], ax=ax1, orientation=orientation, linewidth=linewidth, ) cond1 = vaspxml.dos.energies >= elimit[0] cond2 = vaspxml.dos.energies <= elimit[1] cond = np.all([cond1, cond2], axis=0) if dos_limit is not None: ylim = dos_limit else: if len(spins) > 1: ylim = [ vaspxml.dos.total[1, cond].max() * -1.1, vaspxml.dos.total[0, cond].max() * 1.1 ] else: ylim = [0, vaspxml.dos.total[spins[0], cond].max() * 1.1] if orientation == "horizontal": ax1.set_xlabel(r"$E-E_f$ [eV]") ax1.set_ylabel("Density of States [a.u.]") ax1.set_xlim(elimit) ax1.set_ylim(ylim) elif orientation == "vertical": ax1.set_ylabel(r"$E-E_f$ [eV]") ax1.set_xlabel("Density of States [a.u.]") ax1.set_ylim(elimit) ax1.set_xlim(ylim) # we use ylim because the plot is vertical ax1.axhline(color="black", linestyle="--") ax1.axvline(color="black", linestyle="--") # fig.tight_layout() if grid: ax1.grid() if labels or "stack" in mode: ax1.legend() if title: ax1.set_title(title, fontsize=17) if savefig: fig.savefig(savefig, bbox_inches="tight") plt.close( ) # Added by Nicholas Pike to close memory issue of looping and creating many figures return None, None else: plt.show(block=plt_show) return fig, ax1
# # ## if __name__ == "__main__": ## bandsplot(mode='parametric',elimit=[-6,6],orbitals=[4,5,6,7,8],vmin=0,vmax=1, code='elk') ## knames=['$\Gamma$', '$X$', '$M$', '$\Gamma$', '$R$','$X$'], ## kticks=[0, 8, 16, 24, 38, 49])