The documentation said:

Configure aenet paths
---------------------

To configure the paths to the aenet installation directory and
compiled binaries with the `--path/-p` option.

  $ aenet config --path <aenet_path>

But only -p worked, while --path reported error:

aenet: error: unrecognized arguments: --path

Error message was:

Traceback (most recent call last):
  File "/Users/jianzhou/.local/bin/aenet", line 30, in <module>
    args.run(args)
  File "/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/commandline/aenet_sconv.py", line 231, in run
    self.sconvert(infile=args.input_file,
  File "/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/commandline/aenet_sconv.py", line 465, in sconvert
    struc = self. read_input_structure(
  File "/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/commandline/aenet_sconv.py", line 290, in read_input_structure
    struc = structure.read_safely(infile, frmt=infrmt, **kwargs)
  File "/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/io/structure.py", line 87, in read_safely
    struc = read(filename, frmt=frmt, **kwargs)
  File "/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/io/structure.py", line 78, in read
    return formats[frmt].read(filename, **kwargs)
  File "/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/formats/vasp.py", line 214, in read
    if has_lxml:
NameError: name 'has_lxml' is not defined

I tried to debug it:

In the ../aenet/formats/vasp.py file, line 11

try:
       from lxml import etree
       has_lxml = True
except ImportError:
       import xml.etree.cElementTree as ET

As I didn't install lxml package, the variable has_lxml is not defined. Thus on the line 214:

if has_lxml:
             parser = etree.XMLParser(recover=True, huge_tree=True)
             tree = etree.parse(infile, parser)

It reported one error.

I copied the config file contents sent by Vahe via Slack. Two error happened:

1. Quotation formats were wrong.
2. Quotations were missing for "False" and "True".

Here is the copy from Vahe (via slack):

{
“aenet”: {
“root_path”: “/data/home/UrbanGroup/2022-09-similarity/aenet”,
“generate_x_path”: “/data/home/UrbanGroup/2022-09-similarity/aenet/generate.x”,
“train_x_path”: “/data/home/UrbanGroup/2022-09-similarity/aenet”,
“predict_x_path”: “/data/home/UrbanGroup/2022-09-similarity/aenet”,
“trnset2ascii_x_path”: “/data/home/UrbanGroup/2022-09-similarity/aenet/trnset2ASCII.x”
},
“matplotlib_rc_params”: {
“font.size”: 14,
“legend.frameon”: False,
“xtick.top”: True,
“xtick.direction”: “in”,
“xtick.minor.visible”: True,
“xtick.major.size”: 8,
“xtick.minor.size”: 4,
“ytick.right”: True,
“ytick.direction”: “in”,
“ytick.minor.visible”: True,
“ytick.major.size”: 8,
“ytick.minor.size”: 4
}
}

The installation with the command python setup.py install --user was successful. When run aenet in the command line, the error came up:

Traceback (most recent call last):
  File "/Users/jianzhou/.local/bin/aenet", line 27, in <module>
    cmd.discover(subparsers)
  File "/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/commandline/__init__.py", line 23, in discover
    __import__('aenet.commandline.' + t)
  File "/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/commandline/aenet_sfp.py", line 6, in <module>
    from ..trainset import TrnSet
  File "/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/trainset.py", line 18, in <module>
    from .io.structure import read_safely
  File "/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/io/structure.py", line 18, in <module>
    from ..formats import formats
  File "/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/formats/__init__.py", line 23, in <module>
    parser_module.append(__import__('aenet.formats.' + package))
  File "/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/formats/crystal.py", line 14, in <module>
    from ..staticdata import atomic_number, atomic_species
  File "/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/staticdata.py", line 24, in <module>
    with open(resource_filename(__name__, "data/species.json"), 'r') as fp:
FileNotFoundError: [Errno 2] No such file or directory: '/Users/jianzhou/.local/lib/python3.9/site-packages/aenet/data/species.json'

I guess the reason is aenet/data folder was not copied to the right path during installation. Maybe something is wrong with the setup.py file.

I encountered some issues while trying to run the example notebook. Specifically, the from_fortran_binary_file function fails to convert data.train into data.train.ascii unless I remove the --raw mask. However, even after successful conversion, the from_ascii_file function expects a different structure order for data.train.ascii.

Steps to Reproduce:

Run the example notebook.
Observe the failure of the from_fortran_binary_file function to convert data.train into data.train.ascii when the --raw mask is applied.
Remove the --raw mask and observe successful conversion.
Notice that the from_ascii_file function expects a different structure order for data.train.ascii after conversion.
Expected Behavior:
The from_fortran_binary_file function should be able to convert data.train into data.train.ascii even when the --raw mask is applied. Additionally, the from_ascii_file function should be compatible with the structure order of data.train.ascii after conversion.

Additional Information:
I suspect that there might be multiple versions of trainbin2ASCII.x available, and I may not be using the correct one. Is it possible that the issue arises because I am using an outdated version instead of the latest version of aenet-Pytorch?

Thank you for your assistance in resolving this issue.