What am I doing wrong when training a classifier on my own dataset?

[AlexeyAL1e]

I use the commands as stated in this instructions and formed a dataset of 20 images and two personalities, as this is just for the test. But I am getting this error:

src/classifier.py:59: SyntaxWarning: assertion is always true, perhaps remove parentheses?                                                                 
assert(len(cls.image_paths)>0, 'There must be at least one image for each class in the dataset')                                                         
[1]    22199 illegal hardware instruction (core dumped)  python src/classifier.py TRAIN ~/test_data_160/  ~/models/my_test_class.pkl 

Here is what my training command looks like:

python src/classifier.py TRAIN  \
~/test_data_160/  \
~/models/model-20180402-114759.pb  \
~/models/my_test_class.pkl --batch_size 20

Do I understand correctly that I should get the file my_test_class.pkl in the models directory after training? Previously, I did not train the model on the LFW data.

requirements:

• python 3.6
• tensorflow==1.7

[adhoc-researcher]

Did you figure out what went wrong? Getting the same error.