From c38034ffa368ae0199004fbc022eee875badf94e Mon Sep 17 00:00:00 2001
From: Peter Luenenschloss <peter.luenenschloss@ufz.de>
Date: Mon, 29 Apr 2019 19:41:35 +0200
Subject: [PATCH] made flagSoilMoistureByFrost parameter passing compatible to
call by metadata.csv.
---
funcs/functions.py | 81 +++++++++-------------------------------------
1 file changed, 16 insertions(+), 65 deletions(-)
diff --git a/funcs/functions.py b/funcs/functions.py
index 280397048..6f0cf24ea 100644
--- a/funcs/functions.py
+++ b/funcs/functions.py
@@ -129,92 +129,43 @@ def flagMad(data, flags, field, flagger, length, z, deriv, **kwargs):
return data, flags
-def flagSoilMoistureBySoilFrost(data, flags, field, flagger, time_stamp, tolerated_deviation, soil_temp_reference,
- reference_field=None, reference_flags=None, reference_flagger=None,
- reference_time_stamp=None, frost_level=0, **kwargs):
+def flagSoilMoistureBySoilFrost(data, flags, field, flagger, soil_temp_reference, tolerated_deviation,
+ frost_level=0, **kwargs):
"""Function flags Soil moisture measurements by evaluating the soil-frost-level in the moment of measurement.
Soil temperatures below "frost_level" are regarded as denoting frozen soil state.
:param data: The pandas dataframe holding the data-to-be flagged.
:param flags: A dataframe holding the flags/flag-entries of "data"
- :param field: Fieldname of the Soil moisture measurements in data.
- (Soil moisture measurement column should be accessible by "data[field]")
+ :param field: Fieldname of the Soil moisture measurements field in data.
:param flagger: A flagger - object.
- :param time_stamp: (1)A STRING, denoting the data fields name, that holds the timestamp
- series associated with the data,
- (2) Pass None or 'index', if the input data dataframe is indexed with a
- timestamp.
- (3) Pass an array-like thingy, holding timestamp/datetime
like thingies that refer to the data(including datestrings).
:param tolerated_deviation: An offset alias, denoting the maximal temporal deviation,
- the Soil frost states timestamp is allowed to have, relative to the
+ the soil frost states timestamp is allowed to have, relative to the
data point to-be-flagged.
- :param soil_temp_reference: (1) A STRING, denoting the fields name in data,
+ :param soil_temp_reference: A STRING, denoting the fields name in data,
that holds the data series of soil temperature values,
the to-be-flagged values shall be checked against.
- (2) A date indexed pandas.Series, holding the data series of soil
- temperature values, the to-be-flagged values shall be checked against.
- (3) A data frame (most likely refering to a loggers measurements), containing the
- temperature values, the to-be-flagged values shall be checked against,
- in one of its fields. (In this case, you have to pass
- reference_field and reference_time_stamp as well)
- :param reference_field: If a Dataframe is passed to soil_temp_reference, that parameter holds the
- Fieldname refereing to the Soil temperature measurements.
- :param reference_flag: If there are flags available for the reference series, pass them here
- :param reference_flagger: If the flagger of the reference series is not the same as the one used
- for the data-to-be-flagged, pass it here.
- :param reference_time_stamp:
:param frost_level: Value level, the flagger shall check against, when evaluating soil frost level.
"""
# TODO: (To ASK):HOW TO FLAG nan values in input frame? general question: what should a test test?
# TODO: -> nan values with reference values that show frost, are flagged bad, nan values with reference value nan
# TODO: as well, are not flagged (test not applicable-> no flag)
- # TODO: (To comment):PERFORMANCE COST OF NOT HARMONIZED
- # TODO: Index = None input option
+ # TODO: puffer zone for intermediate/fluktuating frost state
- # check and retrieve data series input:
- if isinstance(time_stamp, str):
- dataseries = pd.Series(data[field].values, index=pd.to_datetime(data[time_stamp].values))
- else:
- dataseries = pd.Series(data[field].values, index=pd.to_datetime(list(time_stamp)))
+ # retrieve data series input:
+ dataseries = pd.Series(data[field].values, index=pd.to_datetime(data.index))
- # check and retrieve reference input:
- #if reference is a string, it refers to data field
- if isinstance(soil_temp_reference, str):
- # if reference series is part of input data frame, evaluate input data flags:
- flag_mask = flagger.isFlagged(flags)[soil_temp_reference]
- # retrieve reference series
- refseries = pd.Series(data[soil_temp_reference].values,
- index=dataseries.index)
- # drop flagged values:
- refseries = refseries.loc[~np.array(flag_mask)]
-
- # if reference is a series, it represents the soil temperature series-to-refer-to:
- elif isinstance(soil_temp_reference, pd.Series):
- refseries = soil_temp_reference
- if reference_flags is not None:
- if reference_flagger is None:
- reference_flagger = flagger
- reference_flag_mask = reference_flagger.isFlagged(reference_flags)
- refseries = refseries.loc[~np.array(reference_flag_mask)]
-
- # if reference is a dataframe, it contains the soil temperature series to-refer-to:
- elif isinstance(soil_temp_reference, pd.DataFrame):
- if isinstance(reference_time_stamp, str):
- refseries = pd.Series(soil_temp_reference[reference_field].values,
- index=pd.to_datetime(soil_temp_reference[reference_time_stamp].values))
- else:
- refseries = pd.Series(soil_temp_reference[reference_field].values,
- index=pd.to_datetime(list(reference_time_stamp)))
-
- if reference_flags is not None:
- if reference_flagger is None:
- reference_flagger = flagger
- reference_flag_mask = reference_flagger.isFlagged(reference_flags)[reference_field]
- refseries = refseries.loc[~np.array(reference_flag_mask)]
+ # retrieve reference input:
+ #if reference is a string, it refers to data field
+ # if reference series is part of input data frame, evaluate input data flags:
+ flag_mask = flagger.isFlagged(flags)[soil_temp_reference]
+ # retrieve reference series
+ refseries = pd.Series(data[soil_temp_reference].values, index=pd.to_datetime(data.index))
+ # drop flagged values:
+ refseries = refseries.loc[~np.array(flag_mask)]
# make refseries index a datetime thingy
refseries.index = pd.to_datetime(refseries.index)
# drop nan values from reference series, since those are values you dont want to refer to.
--
GitLab