Problems with using command train

[Sean-Nolan6]

Hi
I am currently setting up Antrax on an Ubuntu 21.10. I am using conda as a virtual environment and have installed the Matlab python engine that came with Matlab R2019a. The issue I currently am having is that when using the command antrax train I am seeing an error of:

Traceback (most recent call last):
  File "/home/sean/anaconda3/envs/test/bin/antrax", line 8, in <module>
    sys.exit(main())
  File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/antrax/cli.py", line 659, in main
    """)
  File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/sigtools/modifiers.py", line 158, in __call__
    return self.func(*args, **kwargs)
  File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/clize/runner.py", line 363, in run
    ret = cli(*args)
  File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/clize/runner.py", line 220, in __call__
    return func(*posargs, **kwargs)
  File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/clize/runner.py", line 262, in _cli
    return func('{0} {1}'.format(name, command), *args)
  File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/clize/runner.py", line 220, in __call__
    return func(*posargs, **kwargs)
  File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/antrax/cli.py", line 447, in train
    c.train(examplesdir, ne=ne, aug_options=aug_options)
  File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/antrax/classifier.py", line 513, in train
    cw = class_weight.compute_class_weight('balanced', np.unique(FL.classes), FL.classes)
TypeError: compute_class_weight() takes 1 positional argument but 3 were given

It seems to be an error regarding the Matlab python engine version however I am using Matlab R2019a and pip list shows I have matlabengineforpython R2018a installed.
Any help fixing this would be appreciated.

[asafgal]

Hey Sean,
This is a python error, not related to the matlab engine.
Can you try this fix:

In the file anTraX/antrax/classifier.py, change line 513 to:

cw = class_weight.compute_class_weight(class_weight='balanced', classes =np.unique(FL.classes), y=FL.classes)

Then update the python code with pip install -e .
I am unable to test this myself currently, so let me know what happens!

[Sean-Nolan6]

Hi your first solution helped the first issue, but I'm currently trialing your other datasets to get a sense of how long my tests were. However when I try and run train on the V25 sample, it keeps coming up with the same error:

Epoch 1/3
124/1000 [==>...........................] - ETA: 5:54 - loss: 3.5632 - acc: 0.3478Traceback (most recent call last):
File "/home/sean/anaconda3/envs/test/bin/antrax", line 33, in
sys.exit(load_entry_point('antrax', 'console_scripts', 'antrax')())
File "/home/sean/Documents/AntTrax/anTraX/antrax/cli.py", line 659, in main
""")
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/sigtools/modifiers.py", line 158, in call
return self.func(*args, **kwargs)
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/clize/runner.py", line 363, in run
ret = cli(*args)
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/clize/runner.py", line 220, in call
return func(*posargs, **kwargs)
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/clize/runner.py", line 262, in _cli
return func('{0} {1}'.format(name, command), *args)
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/clize/runner.py", line 220, in call
return func(*posargs, **kwargs)
File "/home/sean/Documents/AntTrax/anTraX/antrax/cli.py", line 447, in train
c.train(examplesdir, ne=ne, aug_options=aug_options)
File "/home/sean/Documents/AntTrax/anTraX/antrax/classifier.py", line 533, in train
verbose=verbose)
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training.py", line 1296, in fit_generator
steps_name='steps_per_epoch')
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training_generator.py", line 221, in model_iteration
batch_data = _get_next_batch(generator)
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training_generator.py", line 363, in _get_next_batch
generator_output = next(generator)
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/tensorflow_core/python/keras/utils/data_utils.py", line 789, in get
six.reraise(*sys.exc_info())
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/six.py", line 719, in reraise
raise value
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/tensorflow_core/python/keras/utils/data_utils.py", line 783, in get
inputs = self.queue.get(block=True).get()
File "/home/sean/anaconda3/envs/test/lib/python3.7/multiprocessing/pool.py", line 657, in get
raise self._value
File "/home/sean/anaconda3/envs/test/lib/python3.7/multiprocessing/pool.py", line 121, in worker
result = (True, func(*args, **kwds))
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/tensorflow_core/python/keras/utils/data_utils.py", line 571, in get_index
return _SHARED_SEQUENCES[uid][i]
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/keras_preprocessing/image/iterator.py", line 65, in getitem
return self._get_batches_of_transformed_samples(index_array)
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/keras_preprocessing/image/iterator.py", line 230, in _get_batches_of_transformed_samples
interpolation=self.interpolation)
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/keras_preprocessing/image/utils.py", line 114, in load_img
img = pil_image.open(io.BytesIO(f.read()))
File "/home/sean/anaconda3/envs/test/lib/python3.7/site-packages/PIL/Image.py", line 3009, in open
"cannot identify image file %r" % (filename if filename else fp)
PIL.UnidentifiedImageError: cannot identify image file <_io.BytesIO object at 0x7f37b8629470>

[asafgal]

Both errors are due to compatibility issue with 3rd party packages.
To fix your last error, reinstall the pillow package:

pip uninstall pillow
pip install pillow==8.4.0

Thanks for bringing this to my attention, I will update the requirement file accordingly.

[Sean-Nolan6]

Uninstalling and installing pillow doesn't seem to have fixed the issue, any other ideas to what it might be?

[asafgal]

Can you verify the pillow version?
In a python console:

import PIL
print(PIL.__version__)

[Sean-Nolan6]

This is the error I get:
(test) sean@seans-ants:~$ import PIL
Traceback (most recent call last):
File "/usr/lib/command-not-found", line 28, in
from CommandNotFound import CommandNotFound
File "/usr/lib/python3/dist-packages/CommandNotFound/CommandNotFound.py", line 19, in
from CommandNotFound.db.db import SqliteDatabase
File "/usr/lib/python3/dist-packages/CommandNotFound/db/db.py", line 5, in
import apt_pkg
ImportError: /usr/local/MATLAB/MATLAB_Runtime/v96/sys/os/glnxa64/libstdc++.so.6: version `GLIBCXX_3.4.29' not found (required by /lib/x86_64-linux-gnu/libapt-pkg.so.6.0)

[asafgal]

The command I sent you are python command that need to be run inside a python interactive shell (open one by typing “python” while inside your antrax conda environment).

…

On 23 Feb 2022, at 22:14, Sean-Nolan6 ***@***.***> wrote: This is the error I get: (test) ***@***.***:~$ import PIL Traceback (most recent call last): File "/usr/lib/command-not-found", line 28, in from CommandNotFound import CommandNotFound File "/usr/lib/python3/dist-packages/CommandNotFound/CommandNotFound.py", line 19, in from CommandNotFound.db.db import SqliteDatabase File "/usr/lib/python3/dist-packages/CommandNotFound/db/db.py", line 5, in import apt_pkg ImportError: /usr/local/MATLAB/MATLAB_Runtime/v96/sys/os/glnxa64/libstdc++.so.6: version `GLIBCXX_3.4.29' not found (required by /lib/x86_64-linux-gnu/libapt-pkg.so.6.0) — Reply to this email directly, view it on GitHub <#35 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACERP5VPDS74BUENN6UVWPTU4U53NANCNFSM5LYZGGHA>. You are receiving this because you commented.