Bug parsing coordinates from rlnBoxFile

[mgflast]

Hey, I just wanted to mention that in data_sequence.py _initialize_data, when starting a refine run and parsing coordinates from a rlnBoxFile TSV X Y Z file, the following error is raised when the coordinate file contains only a single row of values:

File "/cephfs/mlast/_compute/IsoNet2/build/conda_env/lib/python3.10/site-packages/IsoNet/models/network.py", line 168, in prepare_train_dataset
self.train_dataset = Train_sets_n2n(training_params['star_file'],method=training_params['method'],
File "/cephfs/mlast/_compute/IsoNet2/build/conda_env/lib/python3.10/site-packages/IsoNet/models/data_sequence.py", line 128, in init
self._initialize_data()
File "/cephfs/mlast/_compute/IsoNet2/build/conda_env/lib/python3.10/site-packages/IsoNet/models/data_sequence.py", line 152, in _initialize_data
coords = np.loadtxt(row['rlnBoxFile'], dtype=int)[:, [2, 1, 0]]
IndexError: too many indices for array: array is 1-dimensional, but 2 were indexed

Which isn't a big issue but it happened towards the very end of a really long preprocessing job so it was a bit of a shame 🥲

Preprocess tomograms:  96%|█████████████████████████████████▍ | 1615/1688 [6:32:49<17:45, 14.59s/it]
Traceback (most recent call last):
  File "/cephfs/mlast/_compute/IsoNet2/build/conda_env/bin/isonet.py", line 7, in <module>
    sys.exit(main())
  File "/cephfs/mlast/_compute/IsoNet2/build/conda_env/lib/python3.10/site-packages/IsoNet/bin/isonet.py", line 1064, in main
    fire.Fire(ISONET)
  File "/lmb/home/matteoall/.local/lib/python3.10/site-packages/fire/core.py", line 141, in Fire
    component_trace = _Fire(component, args, parsed_flag_args, context, name)
  File "/lmb/home/matteoall/.local/lib/python3.10/site-packages/fire/core.py", line 475, in _Fire
    component, remaining_args = _CallAndUpdateTrace(
  File "/lmb/home/matteoall/.local/lib/python3.10/site-packages/fire/core.py", line 691, in _CallAndUpdateTrace
    component = fn(*varargs, **kwargs)
  File "/cephfs/mlast/_compute/IsoNet2/build/conda_env/lib/python3.10/site-packages/IsoNet/bin/isonet.py", line 743, in refine
    network.prepare_train_dataset(training_params)
  File "/cephfs/mlast/_compute/IsoNet2/build/conda_env/lib/python3.10/site-packages/IsoNet/models/network.py", line 168, in prepare_train_dataset
    self.train_dataset = Train_sets_n2n(training_params['star_file'],method=training_params['method'],
  File "/cephfs/mlast/_compute/IsoNet2/build/conda_env/lib/python3.10/site-packages/IsoNet/models/data_sequence.py", line 128, in __init__
    self._initialize_data()
  File "/cephfs/mlast/_compute/IsoNet2/build/conda_env/lib/python3.10/site-packages/IsoNet/models/data_sequence.py", line 152, in _initialize_data
    coords = np.loadtxt(row['rlnBoxFile'], dtype=int)[:, [2, 1, 0]]
IndexError: too many indices for array: array is 1-dimensional, but 2 were indexed

[procyontao]

OK. This will be fixed, and it is incredible that you are processing 1688 tomograms all together.