filterAdde to lib.ts_operators

saqc/lib/ts_operators.py

@@ -12,6 +12,7 @@ import numpy as np
 import numba as nb
 from sklearn.neighbors import NearestNeighbors
 from scipy.stats import iqr, median_abs_deviation
+from scipy.signal import filtfilt, butter
 import numpy.polynomial.polynomial as poly
 
 logger = logging.getLogger("SaQC")
@@ -328,6 +329,35 @@ def shift2Freq(data: Union[pd.Series, pd.DataFrame], method: str, freq: str, fil
     return data.reindex(target_ind, method=direction, tolerance=tolerance, fill_value=fill_value)
 
 
+def butterFilter(x, cut_off_freq, nyq_freq=0.5, filter_order=2,
+                 fill_method='linear', filter_type='low'):
+    """
+    Applies butter filter.
+    X is expected to be regularly sampled.
+    
+    Parameters
+    ----------
+    x: pd.Series
+        input timeseries
+    cut_off_freq: float
+        The cutoff frequencie. Relates to multiples of the sampling freq
+    nyq_freq: float
+        The niquist frequencie. relates to multiples if the sampling rate
+    fill_method: str
+        Any method keword, accepted by pandas.Series.interpolate.
+    
+
+    Returns
+    -------
+    """
+    na_mask = x.isna()
+    x = x.interpolate(fill_method).interpolate('ffill').interpolate('bfill')
+    b, a = butter(N=filter_order, Wn=cut_off_freq / nyq_freq, btype=filter_type)
+    y = pd.Series(filtfilt(b, a, x), x.index, name=x.name)
+    y[na_mask] = np.nan
+    return y
+
+
 @nb.njit
 def _coeffMat(x, deg):
     # helper function to construct numba-compatible polynomial fit function