diff --git a/docs/FunctionDescriptions.md b/docs/FunctionDescriptions.md
index 66b78a567d5b87329f47880c1a38983b3f78dc0b..2d53066540da9e74a8b93696cb3afdfa934fb5f4 100644
--- a/docs/FunctionDescriptions.md
+++ b/docs/FunctionDescriptions.md
@@ -25,7 +25,10 @@ Main documentation of the implemented functions, their purpose and parameters an
- [machinelearning](#machinelearning)
- [harmonize](#harmonize)
- [deharmonize](#deharmonize)
- - [harmonize_shift2Grid](#harmonize_shift2Grid)
+ - [harmonize_shift2Grid](#harmonize_shift2grid)
+ - [harmonize_aggregate2Grid](#harmonize_aggregate2grid)
+ - [harmonize_linear2Grid](#harmonize_linear2grid)
+ - [harmonize_interpolate2Grid](#harmonize_interpolate2grid)
## range
@@ -774,7 +777,7 @@ and than:
(According to the flagging order of the current flagger.)
-## harmonize_shift2Grid
+## harmonize_shift2grid
```
harmonize_shift2Grid(freq, shift_method='nearest_shift', drop_flags=None)
@@ -819,7 +822,7 @@ In detail, the process includes:
if there is one available in the succeeding sampling interval. If not, BAD/np.nan - flag gets assigned.
* `"nearest_shift"`: every grid point gets assigned the closest flag/datapoint in its range. ( range = +/- `freq`/2 ).
-## harmonize_aggregate2Grid
+## harmonize_aggregate2grid
```
harmonize_aggregate2Grid(freq, agg_func, agg_method='nearest_agg', flag_agg_func=max, drop_flags=None)
@@ -876,7 +879,7 @@ In detail, the process includes:
aggregated with the function passed to agg_method and assigned to it.
-## harmonize_linear2Grid
+## harmonize_linear2grid
```
harmonize_linear2Grid(freq, flag_assignment_method='nearest_agg', flag_agg_func=max, drop_flags=None)
@@ -929,7 +932,7 @@ Linear interpolation of an inserted equidistant frequency grid of sampling rate
aggregated with the function passed to `agg_func` and assigned to it.
-## harmonize_interpolate2Grid
+## harmonize_interpolate2grid
```
harmonize_interpolate2Grid(freq, interpolation_method, interpolation_order=1, flag_assignment_method='nearest_agg',
diff --git a/saqc/funcs/harm_functions.py b/saqc/funcs/harm_functions.py
index 9f96141d1ef884d06db4dca6743fd84f5be4a03b..d5bb8f577946ca246b114420b4ff71a62f2059c6 100644
--- a/saqc/funcs/harm_functions.py
+++ b/saqc/funcs/harm_functions.py
@@ -829,7 +829,8 @@ def linear2Grid(data, field, flagger, freq, flag_assignment_method='nearest_agg'
@register('harmonize_interpolate2Grid')
-def interpolate2Grid(data, field, flagger, freq, interpolation_method, interpolation_order=1, flag_assignment_method='nearest_agg', flag_agg_func=max, drop_flags=None, **kwargs):
+def interpolate2Grid(data, field, flagger, freq, interpolation_method, interpolation_order=1,
+ flag_assignment_method='nearest_agg', flag_agg_func=max, drop_flags=None, **kwargs):
return harmonize(
data,
field,
@@ -842,4 +843,61 @@ def interpolate2Grid(data, field, flagger, freq, interpolation_method, interpola
drop_flags=drop_flags,
**kwargs)
-#def aggregate(data, field, flagger, freq_base, freq_target, agg_func, **kwargs):
\ No newline at end of file
+
+def aggregate(data, field, flagger, source_freq, target_freq, agg_func=np.mean, sample_func=np.mean,
+ invalid_flags=None, max_invalid=np.inf, **kwargs):
+
+ # define the "fastest possible" aggregator
+ if sample_func is None:
+ if max_invalid < np.inf:
+
+ def aggregator(x):
+ if x.isna().sum() < max_invalid:
+ return agg_func(x)
+ else:
+ return np.nan
+ else:
+
+ def aggregator(x):
+ return agg_func(x)
+ else:
+
+ dummy_resampler = pd.Series(np.nan, index=[pd.Timedelta('1min')]).resample('1min')
+ if hasattr(dummy_resampler, sample_func.__name__):
+
+ sample_func_name = sample_func.__name__
+ if max_invalid < np.inf:
+ def aggregator(x):
+ y = getattr(x.resample(source_freq), sample_func_name)()
+ if y.isna().sum() < max_invalid:
+ return agg_func(y)
+ else:
+ return np.nan
+
+ else:
+ def aggregator(x):
+ return agg_func(getattr(x.resample(source_freq), sample_func_name)())
+
+ else:
+ if max_invalid < np.inf:
+ def aggregator(x):
+ y = x.resample(source_freq).apply(sample_func)
+ if y.isna().sum() < max_invalid:
+ return agg_func(y)
+ else:
+ return np.nan
+ else:
+ def aggregator(x):
+ return agg_func(x.resample(source_freq).apply(sample_func))
+
+ return harmonize(
+ data,
+ field,
+ flagger,
+ target_freq,
+ inter_method='bagg',
+ reshape_method='bagg',
+ inter_agg=aggregator,
+ reshape_agg=max,
+ drop_flags=invalid_flags,
+ **kwargs)
diff --git a/test/funcs/test_harm_funcs.py b/test/funcs/test_harm_funcs.py
index 2419370e319243a54417d18225c9aba99586a48d..825f442651cade79ecdf64184e317bbe34458ef3 100644
--- a/test/funcs/test_harm_funcs.py
+++ b/test/funcs/test_harm_funcs.py
@@ -18,7 +18,8 @@ from saqc.funcs.harm_functions import (
linear2Grid,
interpolate2Grid,
shift2Grid,
- aggregate2Grid
+ aggregate2Grid,
+ aggregate
)
@@ -334,20 +335,10 @@ def test_wrapper(data, flagger):
field = data.columns[0]
freq = '15min'
flagger = flagger.initFlags(data)
+ aggregate(data, field, flagger, '15min', '30min', agg_func=np.sum, sample_func=np.mean)
+
linear2Grid(data, field, flagger, freq, flag_assignment_method='nearest_agg', flag_agg_func=max,
drop_flags=None)
aggregate2Grid(data, field, flagger, freq, agg_func=sum, agg_method='nearest_agg',
flag_agg_func=max, drop_flags=None)
shift2Grid(data, field, flagger, freq, shift_method='nearest_shift', drop_flags=None)
-
-if __name__ == "__main__":
- dat = data()
- flagger = TESTFLAGGER[1]
- test_gridInterpolation(data(), 'polynomial')
- flagger2 = TESTFLAGGER[2]
- flagger = flagger.initFlags(dat)
- flagger2 = flagger.initFlags(dat2)
- dat_out, flagger = interpolate2Grid(dat, 'data', flagger, '15min', interpolation_method="polynomial", flag_assignment_method='nearest_agg',
- flag_agg_func=max, drop_flags=None)
-
- print("stop")
\ No newline at end of file