inwonyeu / interphon Goto Github PK

Dear In Won Yeu,

During the use of your software I faced with another problem. If I try to sequentially initialize two instances of PreProcess with different POSCARS that have different number of atoms, I get the following error:

ValueError Traceback (most recent call last)
Input In [3], in <cell line: 15>()
13 pre.set_user_arg(user_args)
14 pre.set_unit_cell(path_us_2)
---> 15 pre.set_super_cell(path_ss_2)

File ~\PycharmProjects\interphon\InterPhon\core\pre_process.py:109, in PreProcess.set_super_cell(self, out_file, comment, write_file, code_name)
93 def set_super_cell(self, out_file: FilePath,
94 comment: str = 'Supercell',
95 write_file: bool = True,
96 code_name: str = 'vasp') -> File:
97 """
98 Set a SuperCell instance from the information stored in its UnitCell and UserArgument instances.
99
(...)
107 :type code_name: str
108 """
--> 109 self.super_cell.set_super_cell(self.unit_cell, self.user_arg)
110 self.super_cell.set_super_ind_true(self.unit_cell, self.user_arg)
111 if write_file is True:

File ~\PycharmProjects\interphon\InterPhon\core\super_cell.py:72, in SuperCell.set_super_cell(self, unit_cell, user_arg)
69 if value:
70 _enlarge = _enlarge * user_arg.enlargement[ind]
---> 72 self.atom_type = []
73 for atom_type in unit_cell.atom_type:
74 self.atom_type.extend([atom_type for _ in range(_enlarge)])

File ~\PycharmProjects\interphon\InterPhon\core\unit_cell.py:119, in UnitCell.atom_type(self, _atom_type)
117 self.__atom_type = _atom_type
118 else:
--> 119 raise ValueError("The length of atom_type should be the total number of atoms")

ValueError: The length of atom_type should be the total number of atoms

How to reproduce:

Just run the following code where (path_us_1, path_ss_1) and (path_us_2, path_ss_2) are the paths to the Unitcell/Supercell having different number of atoms. I used a Jupyter Notebook, python 3.9.

from InterPhon.core import PreProcess

user_args = {'dft_code': 'vasp',
'displacement': 0.05,
'enlargement': "1 1 1",
'periodicity': "1 1 0"}

pre = PreProcess()
pre.set_user_arg(user_args)
pre.set_unit_cell(path_us_1)
pre.set_super_cell(path_ss_1)

pre = PreProcess()
pre.set_user_arg(user_args)
pre.set_unit_cell(path_us_1)
pre.set_super_cell(path_ss_2)

Hallo,

THank you for providing the code. Is there in future a plan to release for 3D system. When I do for bulk system it says it has not been implemented.

Also, one question in this code at T=0 the code calculates ZPE. Is that right? ZPE= 1/2 * h_bar * (sum over all positive frequencies))

Thanks,
Asif

Dear In Won Yeu,

If I try to create POSCAR files with atom displacements with any 'periodicity' tag except '1 1 0' (for instance, '1 1 1') the program fails with an error:

File InterPhon\core\pre_process.py:139, in PreProcess.write_displace_cell(self, out_file, code_name, sym_flag)
137 if sym_flag:
138 self.sym = Symmetry2D(self.unit_cell, self.super_cell, self.user_arg)
--> 139 _, _, _ = self.sym.search_point_group()
140 except error.Cannot_Search_Point_Group as e:
141 print("look-up table: ", e.value)

File InterPhon\util\symmetry.py:151, in Symmetry2D.search_point_group(self)
149 rot_ind = []
150 for ind, rot in enumerate(W_candidate):
--> 151 G_rotate = np.dot(np.transpose(rot), np.dot(G_metric, rot))
152 if np.allclose(G_metric, G_rotate, atol=1e-06):
153 rot_ind.append(ind)

File <array_function internals>:180, in dot(*args, **kwargs)

ValueError: shapes (3,3) and (2,2) not aligned: 3 (dim 1) != 2 (dim 0)

Is it expected that InterPhon can operate only with 2D PBC?

Thank you in advance.