doc: unified docstrings of Whittaker-like classes

chemotools/baseline/_air_pls.py

@@ -61,6 +61,9 @@ class AirPls(
     polynomial_order : int, optional default=1
         The degree of the polynomial used to fit the baseline. A value of 1 corresponds
         to a linear fit, while higher values correspond to higher-order polynomials.
+        Higher values will result in a smoother baseline.
+        Currently, values ``>= 3`` are highly discouraged due to numerical instability
+        that might obscure the smoothing effect.
 
     nr_iterations : int, optional default=15
         The number of iterations used to calculate the baseline. Increasing the number

chemotools/baseline/_ar_pls.py

@@ -59,6 +59,13 @@ class ArPls(
         The lambda parameter that controls the smoothness of the baseline. Higher values
         will result in a smoother baseline.
 
+    differences : int, default=2
+        The order of the differences used for the penalty terms that enforces smoothness
+        of the baseline.
+        Higher values will result in a smoother baseline.
+        Currently, values ``>= 3`` are highly discouraged due to numerical instability
+        that might obscure the smoothing effect.
+
     ratio : float, default=0.01
         The convergence threshold for the weight updating scheme. Lower values will
         result in a more accurate baseline at the cost of computation time and even

chemotools/smooth/_whittaker_smooth.py

@@ -58,7 +58,7 @@ class WhittakerSmooth(
         Please refer to the Notes section for further details.
 
     differences : int, default=1
-        The number of differences to use for the Whittaker smooth. If the aim is to
+        The order of differences to use for the Whittaker smooth. If the aim is to
         obtain a smooth estimate of the ``m``-th order derivative, this should be set to
         at least ``m + 2``.
         Currently, values ``>= 3`` are highly discouraged due to numerical instability