# requirements:
python >= 3.7
scikit-learn
numpy
# setup through Makefile:
make
# recurrence results
python3 test/test_birdmap.py
python3 test/test_abalone.py
# learn conditional optimization of other models:
# need a fit, predict method of the model "reg", well-fitted to sklearn.linear_model
# follow the methods in test/test_birdmap.py or test_abalone.py,
# set up schema, database and init-params of functions
# then use cond_regress.separation(.) to train conditional regressions.