Errors encountered during setup

[flackbash]

Dear Michael,

Thank you for making the code to WEXEA public!

I encountered a few minor problems while following the steps in the README to setup WEXEA. I'll outline them here along with my fixes in case anyone else needs it (and for personal reference).

1) Tensorflow 2.4 cannot be installed anymore (using pip). I installed Tensorflow 2.8.4 instead, which conflicts with the specified protobuf version (3.6.0), so I'm using protobuf 3.9.2.

2) When running parser_3.py, I get the error No module named stanza. stanza should probably be added to the requirements.

3) Also when running parser_3.py, I get the error

    sutime_path = config['sutime']
    KeyError: 'sutime'

I added the path to the sutime directory within the WEXEA repository to `configs/config.json`.

[flackbash]

When trying to run parser_4.py, I'm encountering the following error:

2024-10-09 11:40:20.630443: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcuda.so.1'; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory
2024-10-09 11:40:20.630463: W tensorflow/stream_executor/cuda/cuda_driver.cc:269] failed call to cuInit: UNKNOWN ERROR (303)
2024-10-09 11:40:20.630482: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (wolga): /proc/driver/nvidia/version does not exist
2024-10-09 11:40:20.630665: I tensorflow/core/platform/cpu_feature_guard.cc:151] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Traceback (most recent call last):
  File "/local/data/prangen/src/wexea-new-code/env/lib/python3.10/site-packages/tensorflow/python/training/py_checkpoint_reader.py", line 92, in NewCheckpointReader
    return CheckpointReader(compat.as_bytes(filepattern))
RuntimeError: Unsuccessful TensorSliceReader constructor: Failed to find any matching files for 

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/local/data/prangen/src/wexea-new-code/src/parser_4.py", line 329, in <module>
    for var_name, _ in tf.train.list_variables(checkpoint_dir):
  File "/local/data/prangen/src/wexea-new-code/env/lib/python3.10/site-packages/tensorflow/python/training/checkpoint_utils.py", line 109, in list_variables
    reader = load_checkpoint(ckpt_dir_or_file)
  File "/local/data/prangen/src/wexea-new-code/env/lib/python3.10/site-packages/tensorflow/python/training/checkpoint_utils.py", line 64, in load_checkpoint
    return py_checkpoint_reader.NewCheckpointReader(filename)
  File "/local/data/prangen/src/wexea-new-code/env/lib/python3.10/site-packages/tensorflow/python/training/py_checkpoint_reader.py", line 96, in NewCheckpointReader
    error_translator(e)
  File "/local/data/prangen/src/wexea-new-code/env/lib/python3.10/site-packages/tensorflow/python/training/py_checkpoint_reader.py", line 31, in error_translator
    raise errors_impl.NotFoundError(None, None, error_message)
tensorflow.python.framework.errors_impl.NotFoundError: Unsuccessful TensorSliceReader constructor: Failed to find any matching files for

From the code it looks like I need to set "original_el_model" in configs/config.json, however, I don't know which path this should point to. For now, I'm therefore stuck with using parser_4_greedy.py.

Any help with this is appreciated!