def test_TemporalDayofYear(sample_input_data):
ct = pp.TemporalDayofYear(variable=config.TEMPORAL_DAY, ref_feature=config.DATETIME_INDEX)
data_t = ct.transform(sample_input_data)
var = config.TEMPORAL_DAY
assert len(data_t[var].unique()) == 366
assert data_t[var].min() == 1
assert data_t[var].max() == 366
assert type(data_t) == pd.core.frame.DataFrame
def test_SunAzimuth(sample_input_data):
ct0 = pp.TemporalHour(variable=config.TEMPORAL_HOUR, ref_feature=config.DATETIME_INDEX)
ct1 = pp.TemporalDayofYear(variable=config.TEMPORAL_DAY, ref_feature=config.DATETIME_INDEX)
ct2 = pp.SolarElevAngle(var_name=config.SOLAR_ANGLE, day=config.TEMPORAL_DAY, hour=config.TEMPORAL_HOUR)
ct3 = pp.SunAzimuth(var_name=config.SUN_AZIMUTH, day=config.TEMPORAL_DAY, hour=config.TEMPORAL_HOUR)
data_t = ct0.transform(sample_input_data)
data_t = ct1.transform(data_t)
data_t = ct2.transform(data_t)
data_t = ct3.transform(data_t)
var = config.SUN_AZIMUTH
assert data_t[var] is not None
assert data_t[var].max() <= 180
assert data_t[var].min() >= -180
assert type(data_t) == pd.core.frame.DataFrame
def test_SolarElevAngle(sample_input_data):
ct0 = pp.TemporalHour(variable=config.TEMPORAL_HOUR, ref_feature=config.DATETIME_INDEX)
ct1 = pp.TemporalDayofYear(variable=config.TEMPORAL_DAY, ref_feature=config.DATETIME_INDEX)
ct2 = pp.SolarElevAngle(var_name=config.SOLAR_ANGLE, day=config.TEMPORAL_DAY, hour=config.TEMPORAL_HOUR)
data_t = ct0.transform(sample_input_data)
data_t = ct1.transform(data_t)
data_t = ct2.transform(data_t)
var = config.SOLAR_ANGLE
assert data_t[var] is not None
#in summer it doesn't go over 53 degrees
#in winter doesn't drop below -53 degrees
assert (data_t[var].max() > 50) & (data_t[var].max() < 55)
assert (data_t[var].min() < -50) & (data_t[var].min() > -55)
assert type(data_t) == pd.core.frame.DataFrame
def test_DropFeatures(sample_input_data):
ct0 = pp.TemporalHour(variable=config.TEMPORAL_HOUR, ref_feature=config.DATETIME_INDEX)
ct1 = pp.TemporalDayofYear(variable=config.TEMPORAL_DAY, ref_feature=config.DATETIME_INDEX)
ct2 = pp.SolarElevAngle(var_name=config.SOLAR_ANGLE, day=config.TEMPORAL_DAY, hour=config.TEMPORAL_HOUR)
ct3 = pp.SunAzimuth(var_name=config.SUN_AZIMUTH, day=config.TEMPORAL_DAY, hour=config.TEMPORAL_HOUR)
ct4 = pp.TheoreticalRadiation(var_name=config.THEOR_SRAD, day=config.TEMPORAL_DAY, hour=config.TEMPORAL_HOUR)
ct5 = pp.DropUnnecessaryFeatures(variables_to_drop=config.DROP_FEATURES)
data_t = ct0.transform(sample_input_data)
data_t = ct1.transform(data_t)
data_t = ct2.transform(data_t)
data_t = ct3.transform(data_t)
data_t = ct4.transform(data_t)
data_t = ct5.transform(data_t)
assert config.DROP_FEATURES not in data_t.columns.to_list()
assert type(data_t) == pd.core.frame.DataFrame
def test_TheoreticalIrradiance(sample_input_data):
ct0 = pp.TemporalHour(variable=config.TEMPORAL_HOUR, ref_feature=config.DATETIME_INDEX)
ct1 = pp.TemporalDayofYear(variable=config.TEMPORAL_DAY, ref_feature=config.DATETIME_INDEX)
ct2 = pp.SolarElevAngle(var_name=config.SOLAR_ANGLE, day=config.TEMPORAL_DAY, hour=config.TEMPORAL_HOUR)
ct3 = pp.SunAzimuth(var_name=config.SUN_AZIMUTH, day=config.TEMPORAL_DAY, hour=config.TEMPORAL_HOUR)
ct4 = pp.TheoreticalRadiation(var_name=config.THEOR_SRAD, day=config.TEMPORAL_DAY, hour=config.TEMPORAL_HOUR)
data_t = ct0.transform(sample_input_data)
data_t = ct1.transform(data_t)
data_t = ct2.transform(data_t)
data_t = ct3.transform(data_t)
data_t = ct4.transform(data_t)
var = config.THEOR_SRAD
assert data_t[var] is not None
assert len(sample_input_data) == len(data_t[var])
assert type(data_t) == pd.core.frame.DataFrame
from sklearn.pipeline import Pipeline
from sklearn.ensemble import RandomForestRegressor
from sklearn.preprocessing import StandardScaler
# from sklearn.preprocessing import MinMaxScaler #select one of the two
from regression_model.processing import preprocessors as pp
from regression_model.config import config
# from regression_model.processing import features
import logging
_logger = logging.getLogger(__name__)
energy_pipe = Pipeline(
[
('wind_disc', pp.WindDiscretizer(variables=config.WIND_DISCRETE)),
('num_to_binary', pp.DiscretizerNumericalIntoBinary(boundaries=config.BINARY_BOUNDARIES, variables=config.NUM_TO_BINARY)),
('temporal_hour', pp.TemporalHour(variable=config.TEMPORAL_HOUR, ref_feature=config.DATETIME_INDEX)),
('temporal_dayofyear', pp.TemporalDayofYear(variable=config.TEMPORAL_DAY, ref_feature=config.DATETIME_INDEX)),
('solar_angle', pp.SolarElevAngle(var_name=config.SOLAR_ANGLE, day=config.TEMPORAL_DAY, hour=config.TEMPORAL_HOUR)),
('sun_azimuth', pp.SunAzimuth(var_name=config.SUN_AZIMUTH, day=config.TEMPORAL_DAY, hour=config.TEMPORAL_HOUR)),
('theor_solar_radiation', pp.TheoreticalRadiation(var_name=config.THEOR_SRAD, day=config.TEMPORAL_DAY, hour=config.TEMPORAL_HOUR)),
('drop_features', pp.DropUnnecessaryFeatures(variables_to_drop=config.DROP_FEATURES)),
("scaler", StandardScaler()),
("random_forest", RandomForestRegressor(n_estimators=300, min_samples_split=5, min_samples_leaf=1, max_features='sqrt',max_depth=30, bootstrap=False, n_jobs= -1, random_state=config.RANDOM_STATE))
]
)