error running Convtasnet on Librimix

[nobel861017]

Hi, has anyone been able to run egs/librimix/ConvTasNet/run.sh successfully?
I encountered problems at stage 2.

Stage 1: Generating csv files including wav path and duration
Results from the following experiment will be stored in exp/train_convtasnet_9feb1466
Stage 2: Training
{'data': {'n_src': 2,
          'sample_rate': 8000,
          'segment': 3,
          'task': 'sep_clean',
          'train_dir': '/home/gerber68/Data/Libri2Mix/wav8k/min/train-100',
          'valid_dir': '/home/gerber68/Data/Libri2Mix/wav8k/min/dev'},
 'filterbank': {'kernel_size': 16, 'n_filters': 512, 'stride': 8},
 'main_args': {'exp_dir': 'exp/train_convtasnet_9feb1466', 'help': None},
 'masknet': {'bn_chan': 128,
             'hid_chan': 512,
             'mask_act': 'relu',
             'n_blocks': 8,
             'n_repeats': 3,
             'skip_chan': 128},
 'optim': {'lr': 0.001, 'optimizer': 'adam', 'weight_decay': 0.0},
 'positional arguments': {},
 'training': {'batch_size': 24,
              'early_stop': True,
              'epochs': 200,
              'half_lr': True,
              'num_workers': 4}}
Traceback (most recent call last):
  File "train.py", line 145, in <module>
    main(arg_dic)
  File "train.py", line 28, in main
    train_set = LibriMix(
  File "/home/gerber68/asteroid/asteroid/data/librimix_dataset.py", line 66, in __init__
    self.df = pd.read_csv(self.csv_path)
  File "/home/gerber68/.local/lib/python3.8/site-packages/pandas/io/parsers.py", line 686, in read_csv
    return _read(filepath_or_buffer, kwds)
  File "/home/gerber68/.local/lib/python3.8/site-packages/pandas/io/parsers.py", line 452, in _read
    parser = TextFileReader(fp_or_buf, **kwds)
  File "/home/gerber68/.local/lib/python3.8/site-packages/pandas/io/parsers.py", line 946, in __init__
    self._make_engine(self.engine)
  File "/home/gerber68/.local/lib/python3.8/site-packages/pandas/io/parsers.py", line 1178, in _make_engine
    self._engine = CParserWrapper(self.f, **self.options)
  File "/home/gerber68/.local/lib/python3.8/site-packages/pandas/io/parsers.py", line 2008, in __init__
    self._reader = parsers.TextReader(src, **kwds)
  File "pandas/_libs/parsers.pyx", line 537, in pandas._libs.parsers.TextReader.__cinit__
  File "pandas/_libs/parsers.pyx", line 711, in pandas._libs.parsers.TextReader._get_header
  File "pandas/_libs/parsers.pyx", line 905, in pandas._libs.parsers.TextReader._tokenize_rows
  File "pandas/_libs/parsers.pyx", line 2042, in pandas._libs.parsers.raise_parser_error
pandas.errors.ParserError: Error tokenizing data. C error: Calling read(nbytes) on source failed. Try engine='pyt
hon'.

I checked the code of create_local_metadata.py, and found out that the variable md_dirs in function create_local_metadata() is an empty list. Hence, causing self.df = pd.read_csv(self.csv_path) in librimix_dataset.py reading the wrong csv path.

Does anyone know what's going on?

[nobel861017]

I think I figured out the problem.
In line 18 of create_local_metadata.py, I printed out librimix_dir.

/home/gerber68/Data//Librin_srcMix

I solved the problem by changing line 13 in prepare_data.sh to changed_n_src=$n_src.

[LiuSiQi-TJ]

did you donwload metadata in librimix? you need to set correct path in perpare_data.sh, --librimix_dir, i change changed_n_src to 2 if n_src is 1