Source code for dpgen.data.surf

#!/usr/bin/env python3 

import warnings
import os,json,shutil,re,glob,argparse
import numpy as np
import subprocess as sp
import dpgen.data.tools.hcp as hcp
import dpgen.data.tools.fcc as fcc
import dpgen.data.tools.diamond as diamond
import dpgen.data.tools.sc as sc
import dpgen.data.tools.bcc as bcc
from dpgen import dlog
from dpgen import ROOT_PATH
from dpgen.remote.decide_machine import  convert_mdata
from dpgen.dispatcher.Dispatcher import make_submission_compat
from dpgen.generator.lib.utils import symlink_user_forward_files
#-----PMG---------
from pymatgen.io.vasp import Poscar
from pymatgen.core import Structure, Element
from pymatgen.io.ase import AseAtomsAdaptor
#-----ASE-------
from ase.io import read
from ase.build import general_surface


[docs]def create_path (path) : path += '/' if os.path.isdir(path) : dirname = os.path.dirname(path) counter = 0 while True : bk_dirname = dirname + ".bk%03d" % counter if not os.path.isdir(bk_dirname) : shutil.move (dirname, bk_dirname) break counter += 1 os.makedirs (path) return path
[docs]def replace (file_name, pattern, subst) : file_handel = open (file_name, 'r') file_string = file_handel.read () file_handel.close () file_string = ( re.sub (pattern, subst, file_string) ) file_handel = open (file_name, 'w') file_handel.write (file_string) file_handel.close ()
""" 1 make unit cell copy to make super cell place element make vasp relax 1a vasp relax 2 scale system perturb system 3 make vasp md 3a vasp md 4 collect md data """ global_dirname_02 = '00.place_ele' global_dirname_03 = '01.scale_pert' global_dirname_04 = '02.md' max_layer_numb = 50
[docs]def out_dir_name(jdata) : super_cell = jdata['super_cell'] from_poscar= jdata.get('from_poscar',False) if from_poscar: from_poscar_path = jdata['from_poscar_path'] poscar_name = os.path.basename(from_poscar_path) cell_str = "%02d" % (super_cell[0]) for ii in range(1,len(super_cell)) : cell_str = cell_str + ("x%02d" % super_cell[ii]) return poscar_name + '.' + cell_str else: cell_type = jdata['cell_type'] elements = jdata['elements'] super_cell = jdata['super_cell'] ele_str = "surf." for ii in elements: ele_str = ele_str + ii.lower() cell_str = "%02d" % (super_cell[0]) for ii in range(1,len(super_cell)) : cell_str = cell_str + ("x%02d" % super_cell[ii]) return ele_str + '.' + cell_type + '.' + cell_str
[docs]def class_cell_type(jdata) : ct = jdata['cell_type'] if ct == "hcp" : cell_type = hcp elif ct == "fcc" : cell_type = fcc elif ct == "diamond" : cell_type = diamond elif ct == "sc" : cell_type = sc elif ct == "bcc" : cell_type = bcc else : raise RuntimeError("unknow cell type %s" % ct) return cell_type
[docs]def poscar_ele(poscar_in, poscar_out, eles, natoms) : ele_line = "" natom_line = "" for ii in eles : ele_line += str(ii) + " " for ii in natoms : natom_line += str(ii) + " " with open(poscar_in, 'r') as fin : lines = list(fin) lines[5] = ele_line + "\n" lines[6] = natom_line + "\n" with open(poscar_out, 'w') as fout : fout.write("".join(lines))
def _poscar_natoms(lines) : numb_atoms = 0 for ii in lines[6].split() : numb_atoms += int(ii) return numb_atoms
[docs]def poscar_natoms(poscar_in) : with open(poscar_in, 'r') as fin : lines = list(fin) return _poscar_natoms(lines)
[docs]def poscar_shuffle(poscar_in, poscar_out) : with open(poscar_in, 'r') as fin : lines = list(fin) numb_atoms = _poscar_natoms(lines) idx = np.arange(8, 8+numb_atoms) np.random.shuffle(idx) out_lines = lines[0:8] for ii in range(numb_atoms) : out_lines.append(lines[idx[ii]]) with open(poscar_out, 'w') as fout: fout.write("".join(out_lines))
def poscar_scale_direct (str_in, scale) : lines = str_in.copy() numb_atoms = _poscar_natoms(lines) pscale = float(lines[1]) pscale = pscale * scale lines[1] = str(pscale) + "\n" return lines def poscar_scale_cartesian (str_in, scale) : lines = str_in.copy() numb_atoms = _poscar_natoms(lines) # scale box for ii in range(2,5) : boxl = lines[ii].split() boxv = [float(ii) for ii in boxl] boxv = np.array(boxv) * scale lines[ii] = "%.16e %.16e %.16e\n" % (boxv[0], boxv[1], boxv[2]) # scale coord for ii in range(8, 8+numb_atoms) : cl = lines[ii].split() cv = [float(ii) for ii in cl] cv = np.array(cv) * scale lines[ii] = "%.16e %.16e %.16e\n" % (cv[0], cv[1], cv[2]) return lines def poscar_scale (poscar_in, poscar_out, scale) : with open(poscar_in, 'r') as fin : lines = list(fin) if 'D' == lines[7][0] or 'd' == lines[7][0]: lines = poscar_scale_direct(lines, scale) elif 'C' == lines[7][0] or 'c' == lines[7][0] : lines = poscar_scale_cartesian(lines, scale) else : raise RuntimeError("Unknow poscar style at line 7: %s" % lines[7]) with open(poscar_out, 'w') as fout: fout.write("".join(lines))
[docs]def poscar_elong (poscar_in, poscar_out, elong, shift_center=True) : with open(poscar_in, 'r') as fin : lines = list(fin) if lines[7][0].upper() != 'C' : raise RuntimeError("only works for Cartesian POSCAR") sboxz = lines[4].split() boxz = np.array([float(sboxz[0]), float(sboxz[1]), float(sboxz[2])]) boxzl = np.linalg.norm(boxz) elong_ratio = elong / boxzl boxz = boxz * (1. + elong_ratio) lines[4] = '%.16e %.16e %.16e\n' % (boxz[0],boxz[1],boxz[2]) if shift_center: poscar_str="".join(lines) st=Structure.from_str(poscar_str,fmt='poscar') cart_coords=st.cart_coords z_mean=cart_coords[:,2].mean() z_shift=st.lattice.c/2-z_mean cart_coords[:,2]=cart_coords[:,2]+z_shift nst=Structure(st.lattice,st.species,coords=cart_coords,coords_are_cartesian=True) nst.to('poscar',poscar_out) else: with open(poscar_out, 'w') as fout: fout.write("".join(lines))
[docs]def make_unit_cell (jdata) : from_poscar= jdata.get('from_poscar',False) if not from_poscar: latt = jdata['latt'] cell_type = class_cell_type(jdata) out_dir = jdata['out_dir'] path_uc = os.path.join(out_dir, global_dirname_02) cwd = os.getcwd() # for ii in scale : # path_work = create_path(os.path.join(path_uc, '%.3f' % ii)) path_work = create_path(path_uc) os.chdir(path_work) if not from_poscar: with open('POSCAR.unit', 'w') as fp: fp.write (cell_type.poscar_unit(latt)) os.chdir(cwd)
[docs]def make_super_cell_pymatgen (jdata) : make_unit_cell(jdata) out_dir = jdata['out_dir'] path_uc = os.path.join(out_dir, global_dirname_02) elements=[Element(ii) for ii in jdata['elements']] if 'vacuum_min' in jdata: vacuum_min=jdata['vacuum_min'] else: vacuum_min=max([float(ii.atomic_radius) for ii in elements]) from_poscar= jdata.get('from_poscar',False) if from_poscar: from_poscar_path = jdata['from_poscar_path'] poscar_name = os.path.basename(from_poscar_path) ss = Structure.from_file(poscar_name) else: from_path = path_uc from_file = os.path.join(from_path, 'POSCAR.unit') ss = Structure.from_file(from_file) # ase only support X type element for i in range(len(ss)): ss[i]='X' ss=AseAtomsAdaptor.get_atoms(ss) all_millers = jdata['millers'] path_sc = os.path.join(out_dir, global_dirname_02) user_layer_numb = None # set default value z_min = None if 'layer_numb' in jdata: user_layer_numb = jdata['layer_numb'] else: z_min = jdata['z_min'] super_cell = jdata['super_cell'] cwd = os.getcwd() path_work = (path_sc) path_work = os.path.abspath(path_work) os.chdir(path_work) for miller in all_millers: miller_str="" for ii in miller : miller_str += str(ii) path_cur_surf = create_path('surf-'+miller_str) os.chdir(path_cur_surf) #slabgen = SlabGenerator(ss, miller, z_min, 1e-3) if user_layer_numb: slab=general_surface.surface(ss,indices=miller,vacuum=vacuum_min,layers=user_layer_numb) else: # build slab according to z_min value for layer_numb in range( 1,max_layer_numb+1): slab=general_surface.surface(ss,indices=miller,vacuum=vacuum_min,layers=layer_numb) if slab.cell.lengths()[-1] >= z_min: break if layer_numb == max_layer_numb: raise RuntimeError("can't build the required slab") #all_slabs = slabgen.get_slabs() dlog.info(os.getcwd()) #dlog.info("Miller %s: The slab has %s termination, use the first one" %(str(miller), len(all_slabs))) #all_slabs[0].to('POSCAR', 'POSCAR') slab.write('POSCAR',vasp5=True) if super_cell[0] > 1 or super_cell[1] > 1 : st=Structure.from_file('POSCAR') st.make_supercell([super_cell[0], super_cell[1], 1]) st.to('POSCAR','POSCAR') os.chdir(path_work) os.chdir(cwd)
[docs]def make_combines (dim, natoms) : if dim == 1 : return [[natoms]] else : res = [] for ii in range(natoms+1) : rest = natoms - ii tmp_combines = make_combines(dim-1, rest) for jj in tmp_combines : jj.append(ii) if len(res) == 0 : res = tmp_combines else : res += tmp_combines return res
[docs]def place_element (jdata) : out_dir = jdata['out_dir'] super_cell = jdata['super_cell'] cell_type = class_cell_type(jdata) elements = jdata['elements'] from_poscar= jdata.get('from_poscar',False) path_sc = os.path.join(out_dir, global_dirname_02) path_pe = os.path.join(out_dir, global_dirname_02) path_sc = os.path.abspath(path_sc) path_pe = os.path.abspath(path_pe) assert(os.path.isdir(path_sc)) assert(os.path.isdir(path_pe)) cwd = os.getcwd() os.chdir(path_sc) surf_list = glob.glob('surf-*') surf_list.sort() os.chdir(cwd) for ss in surf_list: path_surf = os.path.join(path_sc, ss) pos_in = os.path.join(path_surf, 'POSCAR') natoms = poscar_natoms(pos_in) combines = np.array(make_combines(len(elements), natoms), dtype = int) for ii in combines : if any(ii == 0) : continue comb_name = "sys-" for idx,jj in enumerate(ii) : comb_name += "%04d" % jj if idx != len(ii)-1 : comb_name += "-" path_work = os.path.join(path_surf, comb_name) create_path(path_work) pos_out = os.path.join(path_work, 'POSCAR') if from_poscar: shutil.copy2( pos_in, pos_out) else: poscar_ele(pos_in, pos_out, elements, ii) poscar_shuffle(pos_out, pos_out)
[docs]def make_vasp_relax (jdata) : out_dir = jdata['out_dir'] potcars = jdata['potcars'] cwd = os.getcwd() work_dir = os.path.join(out_dir, global_dirname_02) assert (os.path.isdir(work_dir)) work_dir = os.path.abspath(work_dir) if os.path.isfile(os.path.join(work_dir, 'INCAR' )) : os.remove(os.path.join(work_dir, 'INCAR' )) if os.path.isfile(os.path.join(work_dir, 'POTCAR')) : os.remove(os.path.join(work_dir, 'POTCAR')) shutil.copy2( jdata['relax_incar'], os.path.join(work_dir, 'INCAR')) out_potcar = os.path.join(work_dir, 'POTCAR') with open(out_potcar, 'w') as outfile: for fname in potcars: with open(fname) as infile: outfile.write(infile.read()) os.chdir(work_dir) sys_list = glob.glob(os.path.join('surf-*', 'sys-*')) for ss in sys_list: os.chdir(ss) ln_src = os.path.relpath(os.path.join(work_dir,'INCAR')) os.symlink(ln_src, 'INCAR') ln_src = os.path.relpath(os.path.join(work_dir,'POTCAR')) os.symlink(ln_src, 'POTCAR') os.chdir(work_dir) os.chdir(cwd)
[docs]def poscar_scale_direct (str_in, scale) : lines = str_in.copy() numb_atoms = _poscar_natoms(lines) pscale = float(lines[1]) pscale = pscale * scale lines[1] = str(pscale) + "\n" return lines
[docs]def poscar_scale_cartesian (str_in, scale) : lines = str_in.copy() numb_atoms = _poscar_natoms(lines) # scale box for ii in range(2,5) : boxl = lines[ii].split() boxv = [float(ii) for ii in boxl] boxv = np.array(boxv) * scale lines[ii] = "%.16e %.16e %.16e\n" % (boxv[0], boxv[1], boxv[2]) # scale coord for ii in range(8, 8+numb_atoms) : cl = lines[ii].split() cv = [float(ii) for ii in cl] cv = np.array(cv) * scale lines[ii] = "%.16e %.16e %.16e\n" % (cv[0], cv[1], cv[2]) return lines
[docs]def poscar_scale (poscar_in, poscar_out, scale) : with open(poscar_in, 'r') as fin : lines = list(fin) if 'D' == lines[7][0] or 'd' == lines[7][0]: lines = poscar_scale_direct(lines, scale) elif 'C' == lines[7][0] or 'c' == lines[7][0] : lines = poscar_scale_cartesian(lines, scale) else : raise RuntimeError("Unknow poscar style at line 7: %s" % lines[7]) poscar=Poscar.from_string("".join(lines)) with open(poscar_out, 'w') as fout: fout.write(poscar.get_string(direct=False))
[docs]def make_scale(jdata): out_dir = jdata['out_dir'] scale = jdata['scale'] skip_relax = jdata['skip_relax'] cwd = os.getcwd() init_path = os.path.join(out_dir, global_dirname_02) init_path = os.path.abspath(init_path) work_path = os.path.join(out_dir, global_dirname_03) os.chdir(init_path) init_sys = glob.glob(os.path.join('surf-*', 'sys-*')) init_sys.sort() os.chdir(cwd) create_path(work_path) for ii in init_sys : for jj in scale : if skip_relax : pos_src = os.path.join(os.path.join(init_path, ii), 'POSCAR') assert(os.path.isfile(pos_src)) else : try: pos_src = os.path.join(os.path.join(init_path, ii), 'CONTCAR') assert(os.path.isfile(pos_src)) except Exception: raise RuntimeError("not file %s, vasp relaxation should be run before scale poscar") scale_path = os.path.join(work_path, ii) scale_path = os.path.join(scale_path, "scale-%.3f" % jj) create_path(scale_path) os.chdir(scale_path) poscar_scale(pos_src, 'POSCAR', jj) os.chdir(cwd)
[docs]def pert_scaled(jdata) : out_dir = jdata['out_dir'] scale = jdata['scale'] pert_box = jdata['pert_box'] pert_atom = jdata['pert_atom'] pert_numb = jdata['pert_numb'] vacuum_max = jdata['vacuum_max'] vacuum_resol = jdata.get('vacuum_resol',[]) if vacuum_resol: if len(vacuum_resol)==1: elongs = np.arange(vacuum_resol[0], vacuum_max, vacuum_resol[0]) elif len(vacuum_resol)==2: mid_point = jdata.get('mid_point') head_elongs = np.arange(vacuum_resol[0], mid_point, vacuum_resol[0]).tolist() tail_elongs = np.arange(mid_point, vacuum_max, vacuum_resol[1]).tolist() elongs = np.unique(head_elongs+tail_elongs).tolist() else: raise RuntimeError("the length of vacuum_resol must equal 1 or 2") else: vacuum_num = jdata['vacuum_numb'] head_ratio = jdata['head_ratio'] mid_point = jdata['mid_point'] head_numb = int(vacuum_num*head_ratio) tail_numb = vacuum_num - head_numb head_elongs = np.linspace(0,mid_point,head_numb).tolist() tail_elongs = np.linspace(mid_point,vacuum_max,tail_numb+1).tolist() elongs = np.unique(head_elongs+tail_elongs).tolist() cwd = os.getcwd() path_sp = os.path.join(out_dir, global_dirname_03) assert(os.path.isdir(path_sp)) path_sp = os.path.abspath(path_sp) os.chdir(path_sp) sys_pe = glob.glob(os.path.join('surf-*', 'sys-*')) sys_pe.sort() os.chdir(cwd) pert_cmd = "python "+os.path.join(ROOT_PATH, 'data/tools/create_random_disturb.py') pert_cmd += ' -etmax %f -ofmt vasp POSCAR %d %f > /dev/null' %(pert_box, pert_numb, pert_atom) for ii in sys_pe : for jj in scale : path_scale = path_sp path_scale = os.path.join(path_scale, ii) path_scale = os.path.join(path_scale, 'scale-%.3f' % jj) assert(os.path.isdir(path_scale)) os.chdir(path_scale) dlog.info(os.getcwd()) poscar_in = os.path.join(path_scale, 'POSCAR') assert(os.path.isfile(poscar_in)) for ll in elongs: path_elong = path_scale path_elong = os.path.join(path_elong, 'elong-%3.3f' % ll) create_path(path_elong) os.chdir(path_elong) poscar_elong(poscar_in, 'POSCAR', ll) sp.check_call(pert_cmd, shell = True) for kk in range(pert_numb) : pos_in = 'POSCAR%d.vasp' % (kk+1) dir_out = '%06d' % (kk+1) create_path(dir_out) pos_out = os.path.join(dir_out, 'POSCAR') poscar_shuffle(pos_in, pos_out) os.remove(pos_in) kk = -1 pos_in = 'POSCAR' dir_out = '%06d' % (kk+1) create_path(dir_out) pos_out = os.path.join(dir_out, 'POSCAR') poscar_shuffle(pos_in, pos_out) os.chdir(cwd)
def _vasp_check_fin (ii) : if os.path.isfile(os.path.join(ii, 'OUTCAR')) : with open(os.path.join(ii, 'OUTCAR'), 'r') as fp : content = fp.read() count = content.count('Elapse') if count != 1 : return False else : return False return True
[docs]def run_vasp_relax(jdata, mdata): fp_command = mdata['fp_command'] fp_group_size = mdata['fp_group_size'] fp_resources = mdata['fp_resources'] # machine_type = mdata['fp_machine']['machine_type'] work_dir = os.path.join(jdata['out_dir'], global_dirname_02) forward_files = ["POSCAR", "INCAR", "POTCAR"] backward_files = ["OUTCAR","CONTCAR"] forward_common_files = [] work_path_list = glob.glob(os.path.join(work_dir, "surf-*")) task_format = {"fp" : "sys-*"} for work_path in work_path_list : symlink_user_forward_files(mdata=mdata, task_type="fp", work_path=work_path, task_format=task_format) user_forward_files = mdata.get("fp" + "_user_forward_files", []) forward_files += [os.path.basename(file) for file in user_forward_files] backward_files += mdata.get("fp" + "_user_backward_files", []) #if 'cvasp' in mdata['fp_resources']: # if mdata['fp_resources']['cvasp']: # forward_common_files=['cvasp.py'] relax_tasks = glob.glob(os.path.join(work_dir, "surf-*/","sys-*")) relax_tasks.sort() #dlog.info("work_dir",work_dir) #dlog.info("relax_tasks",relax_tasks) if len(relax_tasks) == 0: return relax_run_tasks = [] for ii in relax_tasks : if not _vasp_check_fin(ii): relax_run_tasks.append(ii) run_tasks = [ii.replace(work_dir+"/", "") for ii in relax_run_tasks] #dlog.info(run_tasks) make_submission_compat(mdata['fp_machine'], fp_resources, [fp_command], work_dir, run_tasks, fp_group_size, forward_common_files, forward_files, backward_files, api_version=mdata.get("api_version", "0.9"))
[docs]def gen_init_surf(args): try: import ruamel from monty.serialization import loadfn,dumpfn warnings.simplefilter('ignore', ruamel.yaml.error.MantissaNoDotYAML1_1Warning) jdata=loadfn(args.PARAM) if args.MACHINE is not None: mdata=loadfn(args.MACHINE) except Exception: with open (args.PARAM, 'r') as fp : jdata = json.load (fp) if args.MACHINE is not None: with open (args.MACHINE, "r") as fp: mdata = json.load(fp) out_dir = out_dir_name(jdata) jdata['out_dir'] = out_dir dlog.info ("# working dir %s" % out_dir) if args.MACHINE is not None: # Decide a proper machine mdata = convert_mdata(mdata, ["fp"]) # disp = make_dispatcher(mdata["fp_machine"]) #stage = args.STAGE stage_list = [int(i) for i in jdata['stages']] for stage in stage_list: if stage == 1 : create_path(out_dir) make_super_cell_pymatgen(jdata) place_element(jdata) make_vasp_relax(jdata) if args.MACHINE is not None: run_vasp_relax(jdata, mdata) elif stage == 2 : make_scale(jdata) pert_scaled(jdata) else : raise RuntimeError("unknown stage %d" % stage)
if __name__ == "__main__": parser = argparse.ArgumentParser( description="Generating initial data for surface systems.") parser.add_argument('PARAM', type=str, help="parameter file, json/yaml format") parser.add_argument('MACHINE', type=str,default=None,nargs="?", help="machine file, json/yaml format") args = parser.parse_args() gen_init_surf(args)