def test_adabl_blh_estimation():
dataFile = paths.file_defaultlidardata()
modelFile = paths.file_trainedmodel()
scalerFile = paths.file_trainedscaler()
blh_kabl = adabl_blh_estimation(
dataFile, modelFile, scalerFile, storeInNetcdf=False
)
assert blh_kabl.shape==(288,) and np.isnan(blh_kabl).sum()==0
def test_adabl_qualitymetrics():
dataFile = paths.file_defaultlidardata()
modelFile = paths.file_trainedmodel()
scalerFile = paths.file_trainedscaler()
scores = adabl_qualitymetrics(dataFile, modelFile, scalerFile)
everythingOK = all([
scores[0] < 1000.0, # errl2
scores[1] < 1000.0, # errl1
scores[2] < 3000.0, # errl0
scores[3] > 0.4, # corr
scores[4] < 12, # chrono
scores[5] == 0, # n_invalid
])
assert everythingOK
rcs_1 = rcss["rcs_1"]
blh_mnf = rcss["pbl"]
# Estimation with KABL
# ----------------------
params = utils.get_default_params()
params["n_clusters"] = 3
params["predictors"] = {"day": ["rcs_1"], "night": ["rcs_1", "rcs_2"]}
params["n_profiles"] = 1
params["init"] = "advanced"
blh_kabl = core.blh_estimation(lidarFile, storeInNetcdf=False, params=params)
# Estimation with ADABL
#-----------------------
modelFile = paths.file_trainedmodel()
scalerFile = paths.file_trainedscaler()
blh_adabl = adabl.adabl_blh_estimation(lidarFile,
modelFile,
scalerFile,
storeInNetcdf=False)
# Estimation with RS
#--------------------
rsdata = nc.Dataset(rsFile)
t_rs = rsdata.variables['time']
blh_rs = rsdata.variables['BEST_BLH']
rsdex = np.where(np.logical_and(t_rs > t_values[0], t_rs < t_values[-1]))[0]