by256 / icsg3d Goto Github PK

Great work, this is really exciting. However, I am having trouble generating new samples using the pre-trained models as I get the following value error when running a command such as "python3 generate.py --name perovskites --nsamples 1000 --base CeCrO3 --var 0.5":

ValueError: Shape must be rank 5 but is rank 4 for '{{node lambda/Tile}} = Tile[T=DT_FLOAT, Tmultiples=DT_INT32](Placeholder, lambda/Tile/multiples)' with input shapes: [?,1,1,1,10], [4].

It seems to be a problem with the encoder model of the LatticeDFCVAE class.

Do you know what the cause of this could be and how it could be fixed? Thanks

When trying the next command, the terminal gives this :

"Defaulting to user installation because normal site-packages is not writeable
Collecting absl-py == 0.8.1
Downloading absl-py-0.8.1.tar.gz (103 kB)

Preparing metadata (setup.py) ... error
error: subprocess-exited-with-error

x python setup. py egg_info did not run successfully.
exit code: 1
[6 lines of output]
Traceback (most recent call last):
File "", line 2, in
File "", line 34, in
File "/tmp/pip-install -_ 84sfsvx/absl-py_312bfbd450b64cef8ea93aac18552242/setup.py", line 34, in
raise RuntimeError('Python version 2.7 or 3.4+ is required. ')
RuntimeError: Python version 2.7 or 3.4+ is required.
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed

x Encountered error while generating package metadata.
See above for output."

note: This is an issue with the package mentioned above, not pip.
hint: See above for details."

Is there anyway to fix that ?
THX

I tried to fetch the pre-trained unet and failed. Here is my steps. I have bought one pack from GitHub with 50 G a month.

git clone [email protected]:by256/icsg3d.git
cd icsg3d
git lfs install
git lfs track ".h5"
git lfs track ".hdf5"
git lfs fetch origin master

Error:
fetch: Fetching reference refs/heads/master
batch response: This repository is over its data quota. Account responsible for LFS bandwidth should purchase more data packs to restore access.
error: failed to fetch some objects from 'https://github.com/by256/icsg3d.git/info/lfs'

Hi there,

Thanks for your help! Now I can generate the materials of heusler and perovskite, which is great!

I generated 1000 samples and tried to understand them. Below are two examples of heusler and perovskie materials, respectively.

ex mat1:
# generated using pymatgen data_MgAgO3 _symmetry_space_group_name_H-M 'P 1' _cell_length_a 4.24242926 _cell_length_b 3.87654877 _cell_length_c 3.97564435 _cell_angle_alpha 90.00000000 _cell_angle_beta 90.00000000 _cell_angle_gamma 90.00000000 _symmetry_Int_Tables_number 1 _chemical_formula_structural MgAgO3 _chemical_formula_sum 'Mg1 Ag1 O3' _cell_volume 65.38338299 _cell_formula_units_Z 1 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 Ag Ag0 1 0.00018641 0.00516879 0.01218479 1 O O1 1 -0.00928536 0.47925205 0.51363987 1 O O2 1 0.43859372 0.00218750 0.53234377 1 O O3 1 0.50947471 0.50731383 -0.00132978 1 Mg Mg4 1 0.46758641 0.45861037 0.47057847 1

ex mat2:
# generated using pymatgen data_GdCrO2 _symmetry_space_group_name_H-M 'P 1' _cell_length_a 3.51839519 _cell_length_b 3.70698476 _cell_length_c 3.81973004 _cell_angle_alpha 90.00000000 _cell_angle_beta 90.00000000 _cell_angle_gamma 90.00000000 _symmetry_Int_Tables_number 1 _chemical_formula_structural GdCrO2 _chemical_formula_sum 'Gd1 Cr1 O2' _cell_volume 49.81935364 _cell_formula_units_Z 1 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 Gd Gd0 1 0.00339958 0.00629250 0.01095058 1 O O1 1 0.49762579 -0.00460633 0.49799109 1 Cr Cr2 1 0.49284538 0.50370067 0.49876646 1 O O3 1 0.50543748 0.41581251 0.10943751 1

ex mat1 is perovskite and ex mat2 is heulser.

I am not an expert in material science. When I used pymatgen to analyze their space groups, it gave me 'Pm' for ex mat2 and 'P1' for ex mat1. Does this mean the generated two materials are potential materials for heulser and perovskite? After DFT optimization, potentially they will be heulser and perovskite?

Thank you!
Yong