[ENH] - Add median absolute error

specparam/measures/gof.py

@@ -76,7 +76,7 @@ def compute_mean_abs_error(power_spectrum, modeled_spectrum):
     Returns
     -------
     error : float
-        Computed MAE.
+        Computed mean absolute error.
     """
 
     error = np.abs(power_spectrum - modeled_spectrum).mean()
@@ -97,7 +97,7 @@ def compute_mean_squared_error(power_spectrum, modeled_spectrum):
     Returns
     -------
     error : float
-        Computed MSE.
+        Computed mean squared error.
     """
 
     error = ((power_spectrum - modeled_spectrum) ** 2).mean()
@@ -118,19 +118,41 @@ def compute_root_mean_squared_error(power_spectrum, modeled_spectrum):
     Returns
     -------
     error : float
-        Computed rMSE.
+        Computed root mean squared error.
     """
 
     error = np.sqrt(((power_spectrum - modeled_spectrum) ** 2).mean())
 
     return error
 
 
+def compute_median_abs_error(power_spectrum, modeled_spectrum):
+    """Calculate the median absolute error.
+
+    Parameters
+    ----------
+    power_spectrum : 1d array
+        Real data power spectrum.
+    modeled_spectrum : 1d array
+        Modelled power spectrum.
+
+    Returns
+    -------
+    error : float
+        Computed median absolute error.
+    """
+
+    error = np.median(np.abs(modeled_spectrum - power_spectrum), axis=0)
+
+    return error
+
+
 # Collect available error functions together
 ERROR_FUNCS = {
     'mae' : compute_mean_abs_error,
     'mse' : compute_mean_squared_error,
     'rmse' : compute_root_mean_squared_error,
+    'medae' : compute_median_abs_error,
 }
 
 
@@ -143,7 +165,7 @@ def compute_error(power_spectrum, modeled_spectrum, error_metric='mae'):
         Real data power spectrum.
     modeled_spectrum : 1d array
         Modelled power spectrum.
-    error_metric : {'mae', 'mse', 'rsme'} or callable
+    error_metric : {'mae', 'mse', 'rsme', 'medae'} or callable
         Which approach to take to compute the error.
 
     Returns

specparam/tests/measures/test_gof.py

@@ -36,8 +36,13 @@ def test_compute_root_mean_squared_error(tfm):
     error = compute_root_mean_squared_error(tfm.power_spectrum, tfm.modeled_spectrum_)
     assert isinstance(error, float)
 
+def test_compute_median_abs_error(tfm):
+
+    error = compute_median_abs_error(tfm.power_spectrum, tfm.modeled_spectrum_)
+    assert isinstance(error, float)
+
 def test_compute_error(tfm):
 
-    for metric in ['mae', 'mse', 'rmse']:
+    for metric in ['mae', 'mse', 'rmse', 'medae']:
         error = compute_error(tfm.power_spectrum, tfm.modeled_spectrum_)
         assert isinstance(error, float)