Got an error when training on CPU

[volunteers]

I use this for training:
./train-local --config-file $(pwd)/models/MobileNet/config_mobilenet_technical.json --samples-file $(pwd)/data/TID2013/tid_labels_train.json --image-dir ~/Downloads/image-quality-assessment-master/training_data/distorted_images/

Got error message like:

Using TensorFlow backend.
Traceback (most recent call last):
File "/usr/lib/python3.6/runpy.py", line 193, in _run_module_as_main
"main", mod_spec)
File "/usr/lib/python3.6/runpy.py", line 85, in _run_code
exec(code, run_globals)
File "/src/trainer/train.py", line 6, in
from sklearn.model_selection import train_test_split
ModuleNotFoundError: No module named 'sklearn'

I am pretty sure sklearn installed and I can even run
>> from sklearn.model_selection import train_test_split in python directly.
I tried both on MacOS and Ubuntu, got the same error.

[skiry]

Try to put scikit-learn in the requirements.txt file from the src directory before building the docker.

[volunteers]

put scikit-learn in the requirements.txt solved the problem.