CIF File non-iterable NoneType object Error

[aydinmirac]

Dear All,

I train ALIGNN with cif files. To improve the performance, I tried to augment my cif files with AugLiChem library. Here is the snippet from the original file and the augmented cif file:

Original file:

data_image0
_chemical_formula_structural       H10C14S2N2O2
_chemical_formula_sum              "H10 C14 S2 N2 O2"
_cell_length_a       4.3258
_cell_length_b       8.982
_cell_length_c       8.4721
_cell_angle_alpha    90
_cell_angle_beta     90.594
_cell_angle_gamma    90

_space_group_name_H-M_alt    "P 1"
_space_group_IT_number       1

loop_
  _space_group_symop_operation_xyz
  'x, y, z'

loop_
  _atom_site_type_symbol
  _atom_site_label
  _atom_site_symmetry_multiplicity
  _atom_site_fract_x
  _atom_site_fract_y
  _atom_site_fract_z
  _atom_site_occupancy
  H   H1        1.0  0.83700  0.83300  0.55800  1.0000
  H   H2        1.0  0.16300  0.33300  0.44200  1.0000
.
.
.
(continues)

Augmented file:

# generated using pymatgen
data_H5C7SNO
_symmetry_space_group_name_H-M   'P 1'
_cell_length_a   4.32580000
_cell_length_b   8.98200000
_cell_length_c   8.47210000
_cell_angle_alpha   90.00000000
_cell_angle_beta   90.59400000
_cell_angle_gamma   90.00000000
_symmetry_Int_Tables_number   1
_chemical_formula_structural   H5C7SNO
_chemical_formula_sum   'H10 C14 S2 N2 O2'
_cell_volume   329.16012678
_cell_formula_units_Z   2
loop_
 _symmetry_equiv_pos_site_id
 _symmetry_equiv_pos_as_xyz
  1  'x, y, z'
loop_
 _atom_site_type_symbol
 _atom_site_label
 _atom_site_symmetry_multiplicity
 _atom_site_fract_x
 _atom_site_fract_y
 _atom_site_fract_z
 _atom_site_occupancy
  H  H0  1  0.83420779  0.82961480  0.54722879  1.0
  H  H1  1  0.15729856  0.32521300  0.43626670  1.0
.
.
.
(continues)

ALIGNN works fine with original cif files but whenever I try to train it with augmented file, I encounter the following error:

Using backend: pytorch
Traceback (most recent call last):
  File "/raid/apps/alignn/2021/bin/train_folder.py", line 195, in <module>
    train_for_folder(
  File "/raid/apps/alignn/2021/bin/train_folder.py", line 103, in train_for_folder
    atoms = Atoms.from_cif(file_path)
  File "/raid/apps/alignn/2021/lib/python3.8/site-packages/jarvis/core/atoms.py", line 537, in from_cif
    cif_atoms = cif_atoms.get_primitive_atoms
  File "/raid/apps/alignn/2021/lib/python3.8/site-packages/jarvis/core/atoms.py", line 710, in get_primitive_atoms
    return Spacegroup3D(self).primitive_atoms
  File "/raid/apps/alignn/2021/lib/python3.8/site-packages/jarvis/analysis/structure/spacegroup.py", line 240, in primitive_atoms
    lattice, scaled_positions, numbers = spglib.find_primitive(
TypeError: cannot unpack non-iterable NoneType object

I can not see a problem in augmented files. Do you have any suggestions?

Best regards,

[bdecost]

Hi @miracaydin1, can you please attach the full CIF file and share the versions you're using for at least jarvis-tools and spglib?

This looks like a potential issue with https://github.com/usnistgov/jarvis. Have you tried any other CIF parsers?

[bdecost]

a possible workaround might be to parse the structures using pymatgen and convert to jarvis Atoms, since your augmented file seems to have been written by the pymatgen CIF io module

[knc6]

Can you try to install cif2cell pip install cif2cell==2.0.0a3 ? jarvis-tools by default will use it and hopefully the issue should be resolved https://github.com/usnistgov/jarvis/blob/master/jarvis/core/atoms.py#L304 .
Also as @bdecost mentioned please feel free to send the cif file via email or attach here.

[aydinmirac]

Hi @bdecost,

I installed the latest version of ALIGNN (2022.11.06). The latest version includes spglib=1.16.2 and jarvis-tools=2021.10.03

As I said, I augmented my current cif files with AugLichem (https://github.com/BaratiLab/AugLiChem). As I understand, this library uses pymatgen.

Here are another original and augmented CIF files. I attached the full files.

Original file:

data_image0
_chemical_formula_structural       H10C14S2N2O2
_chemical_formula_sum              "H10 C14 S2 N2 O2"
_cell_length_a       4.3258
_cell_length_b       8.982
_cell_length_c       8.4721
_cell_angle_alpha    90
_cell_angle_beta     90.594
_cell_angle_gamma    90

_space_group_name_H-M_alt    "P 1"
_space_group_IT_number       1

loop_
  _space_group_symop_operation_xyz
  'x, y, z'

loop_
  _atom_site_type_symbol
  _atom_site_label
  _atom_site_symmetry_multiplicity
  _atom_site_fract_x
  _atom_site_fract_y
  _atom_site_fract_z
  _atom_site_occupancy
  H   H1        1.0  0.83700  0.83300  0.55800  1.0000
  H   H2        1.0  0.16300  0.33300  0.44200  1.0000
  H   H3        1.0  0.44200  0.88800  0.82500  1.0000
  H   H4        1.0  0.55800  0.38800  0.17500  1.0000
  H   H5        1.0  0.17000  0.75200  0.00700  1.0000
  H   H6        1.0  0.83000  0.25200  0.99300  1.0000
  H   H7        1.0  0.09300  0.49900  0.98000  1.0000
  H   H8        1.0  0.90700  0.99900  0.02000  1.0000
  H   H9        1.0  0.36800  0.35600  0.77600  1.0000
  H   H10       1.0  0.63200  0.85600  0.22400  1.0000
  C   C1        1.0  0.86010  0.62540  0.49070  1.0000
  C   C2        1.0  0.13990  0.12540  0.50930  1.0000
  C   C3        1.0  0.58150  0.70380  0.69700  1.0000
  C   C4        1.0  0.41850  0.20380  0.30300  1.0000
  C   C5        1.0  0.42910  0.78120  0.81420  1.0000
  C   C6        1.0  0.57090  0.28120  0.18580  1.0000
  C   C7        1.0  0.25580  0.69710  0.92090  1.0000
  C   C8        1.0  0.74420  0.19710  0.07910  1.0000
  C   C9        1.0  0.24010  0.54200  0.90470  1.0000
  C   C10       1.0  0.75990  0.04200  0.09530  1.0000
  C   C11       1.0  0.38720  0.46420  0.78700  1.0000
  C   C12       1.0  0.61280  0.96420  0.21300  1.0000
  C   C13       1.0  0.55200  0.54980  0.68570  1.0000
  C   C14       1.0  0.44800  0.04980  0.31430  1.0000
  S   S1        1.0  0.08567  0.61384  0.33595  1.0000
  S   S2        1.0  0.91433  0.11384  0.66405  1.0000
  N   N1        1.0  0.77000  0.74630  0.57380  1.0000
  N   N2        1.0  0.23000  0.24630  0.42620  1.0000
  O   O1        1.0  0.72840  0.50336  0.55790  1.0000
  O   O2        1.0  0.27160  0.00336  0.44210  1.0000

Augmented file:

# generated using pymatgen
data_H5C7SNO
_symmetry_space_group_name_H-M   'P 1'
_cell_length_a   4.32580000
_cell_length_b   8.98200000
_cell_length_c   8.47210000
_cell_angle_alpha   90.00000000
_cell_angle_beta   90.59400000
_cell_angle_gamma   90.00000000
_symmetry_Int_Tables_number   1
_chemical_formula_structural   H5C7SNO
_chemical_formula_sum   'H10 C14 S2 N2 O2'
_cell_volume   329.16012678
_cell_formula_units_Z   2
loop_
 _symmetry_equiv_pos_site_id
 _symmetry_equiv_pos_as_xyz
  1  'x, y, z'
loop_
 _atom_site_type_symbol
 _atom_site_label
 _atom_site_symmetry_multiplicity
 _atom_site_fract_x
 _atom_site_fract_y
 _atom_site_fract_z
 _atom_site_occupancy
  H  H0  1  0.83420779  0.82961480  0.54722879  1.0
  H  H1  1  0.15729856  0.32521300  0.43626670  1.0
  H  H2  1  0.44203270  0.87915717  0.82747758  1.0
  H  H3  1  0.55019103  0.38184614  0.18319968  1.0
  H  H4  1  0.17514122  0.75619369  0.00136765  1.0
  H  H5  1  0.83553055  0.24418767  0.99477299  1.0
  H  H6  1  0.09609248  0.49658218  0.98098265  1.0
  H  H7  1  0.89645711  0.00334474  0.01640165  1.0
  H  H8  1  0.36614542  0.34817874  0.77735952  1.0
  H  H9  1  0.62119004  0.84995189  0.22153261  1.0
  C  C10  1  0.84699770  0.61782486  0.49019123  1.0
  C  C11  1  0.14011910  0.13232519  0.51033348  1.0
  C  C12  1  0.59049700  0.70509916  0.69616131  1.0
  C  C13  1  0.42211903  0.20489082  0.30292231  1.0
  C  C14  1  0.42748631  0.78231425  0.81204122  1.0
  C  C15  1  0.57196073  0.28928330  0.18326057  1.0
  C  C16  1  0.25796312  0.69827142  0.92115724  1.0
  C  C17  1  0.74538351  0.19733637  0.07035087  1.0
  C  C18  1  0.25245680  0.54381040  0.90293595  1.0
  C  C19  1  0.76391822  0.04290571  0.09075583  1.0
  C  C20  1  0.38889531  0.46937274  0.79015752  1.0
  C  C21  1  0.61097097  0.96269025  0.21394540  1.0
  C  C22  1  0.55416748  0.55167165  0.68179287  1.0
  C  C23  1  0.42629221  0.04735866  0.31177248  1.0
  S  S24  1  0.07796286  0.61757333  0.32790173  1.0
  S  S25  1  0.89490833  0.11503795  0.66120287  1.0
  N  N26  1  0.75230103  0.74561484  0.57227018  1.0
  N  N27  1  0.22614752  0.23622076  0.43074555  1.0
  O  O28  1  0.71180393  0.49850082  0.56406201  1.0
  O  O29  1  0.27605354  0.00896080  0.44516716  1.0

Hi @knc6, let me try to install that package and inform you as soon as possible.

[aydinmirac]

Hi @knc6, I installed the package.

Then I checked "config.json" file. Giving the exact number of the length of dataset gives error. I reduced the number of "n_val", "n_test" and "n_train" values

{
        "version": "1859f483ea2b41f6163b845eb689f0f1afdd1e8f",
        "dataset": "user_data",
        "target": "target",
        "atom_features": "cgcnn",
        "neighbor_strategy": "k-nearest",
        "random_seed": 123,
        "classification_threshold": null,
        "n_val": 5350,
        "n_test": 5350,
        "n_train": 42000,
        "train_ratio": 0.8,
        "val_ratio": 0.1,
        "test_ratio": 0.1,
        "target_multiplication_factor": null,
        "epochs": 300,

After reducing the values by around 10 molecules. The model started to work. I think these numbers are not automatically detected by the length of dataset. Assigning these values manually can cause error sometimes if you give a value more then the length of dataset.