[docs] window size for locreg detrending

neurokit2/signal/signal_detrend.py

@@ -43,8 +43,10 @@ def signal_detrend(
         Only used if ``method`` is "loess". The parameter which controls the degree of smoothing.
     window : float
         Only used if ``method`` is "locreg". The detrending ``window`` should correspond to the
-        desired low frequency band to remove (for instance, ``1.5`` will remove frequencies below
-        1.5Hz).
+        1 divided by the desired low-frequency band to remove
+        (``window = 1 / detrend_frequency``)
+        For instance, to remove frequencies below ``0.67Hz`` the window should be ``1.5``
+        (``1 / 0.67 = 1.5``).
     stepsize : float
         Only used if ``method`` is ``"locreg"``.
     components : list

studies/ecg_benchmark/README.md

@@ -524,8 +524,8 @@ def tarvainen(ecg, sampling_rate):
 def locreg(ecg, sampling_rate):
     ecg = nk.signal_detrend(ecg, 
                             method="locreg", 
-                            window=0.5*sampling_rate, 
-                            stepsize=0.02*sampling_rate)
+                            window=1/0.5,
+                            stepsize=0.02)
     signal, info = nk.ecg_peaks(ecg, sampling_rate=sampling_rate, method="neurokit")
     return info["ECG_R_Peaks"]
 

studies/ecg_benchmark/make_data.py

@@ -152,8 +152,8 @@ def standardization(ecg, sampling_rate):
 #def locreg(ecg, sampling_rate):
 #    ecg = nk.signal_detrend(ecg,
 #                            method="locreg",
-#                            window=0.5*sampling_rate,
-#                            stepsize=0.02*sampling_rate)
+#                            window=1/0.5,
+#                            stepsize=0.02)
 #    signal, info = nk.ecg_peaks(ecg, sampling_rate=sampling_rate, method="neurokit")
 #    return info["ECG_R_Peaks"]
 #