from sklearn.externals import joblib
from sklearn.model_selection import KFold
from sklearn.neural_network import MLPRegressor
from sklearn.preprocessing import StandardScaler
import env
import util.logging
from data_scripts import pcsml_data_loader as dl
from modeling import preprocessing, score_util
from modeling.preprocessing import make_one_hot_pipeline
util.logging.setup_default(env.result_path)
log = logging.getLogger(__name__)
log.info("Running...")
log.info("Env:\n%s", env.dump())
# load the data frame (a sample of the sample to make debugging faster...)
# df: DataFrame = dl.load_df_corn_pkl_smpl_25_20171018().sample(2000)
df: DataFrame = dl.load_df_corn_pkl_smpl_25_20171018()
logging.debug("data shape: %s", df.shape)
y = df.pop('Dry_Yield')
X = df
###
# transform
###
X, label_cols = preprocessing.shape_gis_pps(X)
def test_dump():
os.environ['FOO'] = 'BAR'
f = io.StringIO()
env.dump(f)
assert "= ENV =" in f.getvalue()
assert "FOO=BAR" in f.getvalue()