Error in rule simFeatures about 2020plus HOT 5 OPEN

You need to install scikit-learn, which was listed as one of the dependencies in the conda environment yaml file

from 2020plus.

After installing scikit-learn, I run the code again. But, another mistake occurs.

AN ERROR HAS OCCURRED: check the log file

Type: <class 'ImportError'>
Exception: cannot import name 'cross_validation'
Traceback:
File "/home/kjf/2020plus-1.2.2/2020plus.py", line 263, in
import src.classify.python.classifier
File "/home/kjf/2020plus-1.2.2/src/classify/python/classifier.py", line 2, in
from src.classify.python.dummy_clf import DummyClf
File "/home/kjf/2020plus-1.2.2/src/classify/python/dummy_clf.py", line 2, in
from src.classify.python.generic_classifier import GenericClassifier
File "/home/kjf/2020plus-1.2.2/src/classify/python/generic_classifier.py", line 9, in
from sklearn import cross_validation

[Fri Apr 5 21:37:49 2019]
Error in rule features:
jobid: 2
output: output/features.txt

RuleException:
CalledProcessError in line 282 of /home/kjf/2020plus-1.2.2/Snakefile:
Command 'set -euo pipefail; python which 2020plus.py features -s output/summary.txt --tsg-test output/tsg.txt -og-test output/oncogene.txt -o output/features.txt' returned non-zero exit status 1.
File "/home/kjf/2020plus-1.2.2/Snakefile", line 282, in __rule_features
File "/home/kjf/anaconda3/envs/2020plus/lib/python3.6/concurrent/futures/thread.py", line 56, in run

Could you please help with this problem? Thanks.

from 2020plus.

You need to install the correct version of scikit learn, which is already specified in the conda environment file (scikit-learn<0.20.0).

from 2020plus.

Thanks.
I think that there are a lot of problem with my environment and required packages.
So I try the command:
$ pip install -r requirements.txt
success

but fail with
$ pip install -r requirements_dev.txt

It warns that
Fail building wheel for probabilistic2020
and I check the packages by "$ conda list"
most of the packages required are in wrong version.
I am wondering how to turn each packages into correct version.

from 2020plus.

@MarkKJF Create a new conda environment, activate it and install the dependencies inside the conda environment.

from 2020plus.