Merge pull request #247 from imaliyov/master

matscipy/dislocation.py

@@ -1014,7 +1014,7 @@ def cost_function(pos, dislo, bulk, cylinder_r, elastic_param,
 
 def fit_core_position(dislo_image, bulk, elastic_param, hard_core=False,
                       core_radius=10, current_pos=None, bulk_neighbours=None,
-                      origin=(0, 0)):
+                      origin=(0, 0), return_error=False):
     """
     Use `cost_function()` to fit atomic positions to Stroh solution with
 
@@ -1037,6 +1037,9 @@ def fit_core_position(dislo_image, bulk, elastic_param, hard_core=False,
         `matscipy.neigbbours.neighbour_list('ij', bulk, alat)`.
     origin: tuple
         Optionally pass in coordinate origin (x0, y0)
+    return_error: bool
+        Optionally return the `scipy.optimize.minimize` error of the fit.
+        Requires an additional cost_function() call.
 
     Returns
     -------
@@ -1050,12 +1053,19 @@ def fit_core_position(dislo_image, bulk, elastic_param, hard_core=False,
                     dislo_image, bulk, core_radius, elastic_param,
                     hard_core, False, False, bulk_neighbours, origin),
                    method='Powell', options={'xtol': 1e-2, 'ftol': 1e-2})
-    return res.x
+
+    error = cost_function(res.x, dislo_image, bulk, core_radius, elastic_param, hard_core,
+                          False, False, bulk_neighbours, origin)
+
+    if return_error:
+        return res.x, error
+    else:
+        return res.x
 
 
 def fit_core_position_images(images, bulk, elastic_param,
                              bulk_neighbours=None,
-                             origin=(0, 0)):
+                             origin=(0, 0), return_errors=False):
     """
     Call fit_core_position() for a list of Atoms objects, e.g. NEB images
 
@@ -1071,24 +1081,41 @@ def fit_core_position_images(images, bulk, elastic_param,
         as for `fit_core_position()`.
     origin: tuple
         Optionally pass in coordinate origin (x0, y0)
+    return_errors: bool
+        Optionally return the `scipy.optimize.minimize` list of errors 
+        of the fit.
 
     Returns
     -------
     core_positions: array of shape `(len(images), 2)`
     """
     core_positions = []
+
+    if return_errors:
+        core_pos_errors = []
+
     core_position = images[0].info.get('core_position', origin)
     for dislo in images:
         dislo_tmp = dislo.copy()
-        core_position = fit_core_position(dislo_tmp, bulk, elastic_param,
+        result = fit_core_position(dislo_tmp, bulk, elastic_param,
                                           current_pos=dislo.info.get(
                                               'core_position', core_position),
                                           bulk_neighbours=bulk_neighbours,
-                                          origin=origin)
+                                          origin=origin, return_error=return_errors)
+
+        if return_errors:
+            core_position, error = result
+            core_pos_errors.append(error)
+        else:
+            core_position = result
+
         dislo.info['core_position'] = core_position
         core_positions.append(core_position)
 
-    return np.array(core_positions)
+    if return_errors:
+        return np.array(core_positions), np.array(cor_pos_errors)
+    else:
+        return np.array(core_positions)
 
 
 def screw_cyl_tetrahedral(alat, C11, C12, C44,