Backported fix for selecting sub-optimal optimized parameters for v0.4.1

Upon success the parameter from the last call to _calculate_penalty were used in lieu of the optimized parameters returned by least_square (ls_result.x). Increased version number to v0.4.1
glotaran · Jul 20, 2021 · 0ece818 · 0ece818
1 parent dc00e6d
commit 0ece818
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 2 deletions.
diff --git a/glotaran/__init__.py b/glotaran/__init__.py
@@ -8,7 +8,7 @@
 
 load_plugins()
 
-__version__ = "0.4.0"
+__version__ = "0.4.1"
 
 
 def __getattr__(attribute_name: str):

diff --git a/glotaran/analysis/optimize.py b/glotaran/analysis/optimize.py
@@ -101,7 +101,9 @@ def _create_result(
     root_mean_square_error = np.sqrt(reduced_chi_square) if success else None
     jacobian = ls_result.jac if success else None
 
-    problem.save_parameters_for_history()
+    if success:
+        problem.parameters.set_from_label_and_value_arrays(free_parameter_labels, ls_result.x)
+    problem.reset()
     history_index = None if success else -2
     data = problem.create_result_data(history_index=history_index)
     # the optimized parameters are those of the last run if the optimization has crashed