def test_processing_for_selected_time_period(self):
cml = load_and_clean_example_cml()
# First as a comparisson the standard way
cml.process.wet_dry.std_dev(window_length=60, threshold=0.8)
assert cml.channel_1.data.wet.loc["2016-11-02 18:30"].values[0] == True
assert cml.channel_1.data.wet.loc["2016-11-02 14:00"].values[0] == True
assert cml.channel_1.data.wet.loc["2016-11-02 13:00"].values[
0] == False
# Then for a period from a starting point in time
t_start = "2016-11-02 14:30"
cml.channel_1.data.wet = False
cml.process.wet_dry.std_dev(window_length=60,
threshold=0.8,
t_start=t_start)
assert cml.channel_1.data.wet.loc["2016-11-02 18:30"].values[0] == True
assert cml.channel_1.data.wet.loc["2016-11-02 14:00"].values[
0] == False
assert cml.channel_1.data.wet.loc["2016-11-02 13:00"].values[
0] == False
# Then for a period to a end point in time
t_stop = "2016-11-02 15:00"
cml.channel_1.data.wet = False
cml.process.wet_dry.std_dev(window_length=60,
threshold=0.8,
t_stop=t_stop)
assert cml.channel_1.data.wet.loc["2016-11-02 18:30"].values[
0] == False
assert cml.channel_1.data.wet.loc["2016-11-02 14:00"].values[0] == True
assert cml.channel_1.data.wet.loc["2016-11-02 13:00"].values[
0] == False
def test_standrad_processing(self):
cml = load_and_clean_example_cml()
cml.process.wet_dry.std_dev(window_length=60, threshold=0.8)
assert (
cml.channel_1.data.wet.loc['2016-11-02 14:00'].values[0] == True)
assert (
cml.channel_1.data.wet.loc['2016-11-02 13:00'].values[0] == False)
assert (
cml.channel_2.data.wet.loc['2016-11-02 14:00'].values[0] == True)
assert (
cml.channel_2.data.wet.loc['2016-11-02 13:00'].values[0] == False)
cml.process.wet_dry.std_dev(window_length=30, threshold=0.8)
assert (
cml.channel_1.data.wet.loc['2016-11-02 14:00'].values[0] == False)
assert (
cml.channel_1.data.wet.loc['2016-11-02 13:00'].values[0] == False)
assert (
cml.channel_1.data.wet.loc['2016-11-02 14:30'].values[0] == True)
assert (
cml.channel_2.data.wet.loc['2016-11-02 14:00'].values[0] == False)
assert (
cml.channel_2.data.wet.loc['2016-11-02 13:00'].values[0] == False)
assert (
cml.channel_2.data.wet.loc['2016-11-02 14:30'].values[0] == True)
def test_standrad_processing(self):
cml = load_and_clean_example_cml()
cml.process.wet_dry.std_dev(window_length=60, threshold=0.8)
assert (cml.channel_1.data.wet.loc['2016-11-02 14:00'].values[0]
== True)
assert (cml.channel_1.data.wet.loc['2016-11-02 13:00'].values[0]
== False)
assert (cml.channel_2.data.wet.loc['2016-11-02 14:00'].values[0]
== True)
assert (cml.channel_2.data.wet.loc['2016-11-02 13:00'].values[0]
== False)
cml.process.wet_dry.std_dev(window_length=30, threshold=0.8)
assert (cml.channel_1.data.wet.loc['2016-11-02 14:00'].values[0]
== False)
assert (cml.channel_1.data.wet.loc['2016-11-02 13:00'].values[0]
== False)
assert (cml.channel_1.data.wet.loc['2016-11-02 14:30'].values[0]
== True)
assert (cml.channel_2.data.wet.loc['2016-11-02 14:00'].values[0]
== False)
assert (cml.channel_2.data.wet.loc['2016-11-02 13:00'].values[0]
== False)
assert (cml.channel_2.data.wet.loc['2016-11-02 14:30'].values[0]
== True)
# Test if the end result, the rain rate, is the same when using the
# Processor and the functions
# This test only works correctly if the CML uses vertical polarization
# since the default in the function is 'H'
assert cml.channel_1.metadata['polarization'] == 'V'
cml.process.wet_dry.std_dev(window_length=30, threshold=0.8)
cml.process.baseline.linear()
cml.process.baseline.calc_A()
cml.process.A_R.calc_R()
R_from_processor = cml.channel_1.data.R.copy()
R_from_function = (
pycml.processing.A_R_relation.A_R_relation
.calc_R_from_A(
A=cml.channel_1.data.A,
L=cml.metadata['length'],
f_GHz=cml.channel_1.metadata['frequency'] / 1e9,
pol=cml.channel_1.metadata['polarization']))
np.testing.assert_almost_equal(R_from_processor, R_from_function)