Davout 0.1.1.dev171__tar.gz → 0.1.1.dev172__tar.gz

{davout-0.1.1.dev171/source/Davout.egg-info → davout-0.1.1.dev172}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: Davout
-Version: 0.1.1.dev171
+Version: 0.1.1.dev172
 Summary: Scientific computing utilities combining FEM, ANN and tooling
 Author: Matheus Janczkowski
 Author-email: Matheus Janczkowski <matheusj2009@hotmail.com>

{davout-0.1.1.dev171 → davout-0.1.1.dev172}/pyproject.toml

@@ -63,7 +63,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "Davout"
-version = "0.1.1.dev171"
+version = "0.1.1.dev172"
 description = "Scientific computing utilities combining FEM, ANN and tooling"
 readme = "README.md"
 authors = [

{davout-0.1.1.dev171 → davout-0.1.1.dev172}/source/Davout/DeepMech/aa_tests/mecsol_2026_quotient_space/neural_network_trainer.py

@@ -131,7 +131,8 @@ class SurrogateModel:
 
         training_class.monte_carlo_training(n_realizations=
         self.n_monte_carlo_realizations, best_models_rank_size=
-        self.n_best_models, show_reinitialization_distance=True)
+        self.n_best_models, show_reinitialization_distance=True,
+        model_base_name=self.saved_model_file)
 
         # Checks the loss again with the best model of the Monte Carlo
         # training
@@ -505,7 +506,9 @@ if __name__=="__main__":
     input_data_file = ("00_successful_complete_data_matrix_pc_matheus."+
     "npy")
 
-    saved_model_file = "saved_model"
+    Lp_norm_exponent = 12
+
+    saved_model_file = "saved_model_lp_norm_"+str(Lp_norm_exponent)
 
     n_training_samples = 10000
 
@@ -521,7 +524,7 @@ if __name__=="__main__":
 
     optimizer = "Adam"
 
-    loss_metric = {"name": "LpNormError", "p": 12, 
+    loss_metric = {"name": "LpNormError", "p": Lp_norm_exponent, 
     "use_stable_implementation": True}
 
     training_flag = True 

{davout-0.1.1.dev171 → davout-0.1.1.dev172}/source/Davout/DeepMech/tool_box/training_tools.py

@@ -778,6 +778,12 @@ class ModelCustomTraining:
 
         max_digits = len(str(self.n_iterations))
 
+        # Initializes the recorders for the gradient statistics
+
+        gradient_value = None 
+
+        gradient_maximum_component = None
+
         # Sets the initial time
 
         self.initial_time = time.time()
@@ -893,11 +899,20 @@ class ModelCustomTraining:
             "#########################################################"+
             "\n")
 
-            print("Initial loss.: "+format(initial_loss, '.5e'))
+            print("Initial loss.......................: "+format(
+            initial_loss, '.5e'))
 
-            print("Final loss...: "+format(final_loss, '.5e'))
+            print("Final loss.........................: "+format(
+            final_loss, '.5e'))
+
+            print("Gradient norm......................: "+format(
+            gradient_value, '.5e'))
+
+            print("Gradient maximum absolute component: "+format(
+            gradient_value, '.5e'))
 
-            print("Training time: "+str(self.elapsed_time)+" seconds.\n")
+            print("Training time......................: "+str(
+            self.elapsed_time)+" seconds.\n")
         
         # Gets the trained parameters and reassigns them to the model if 
         # the model was trained using optimizers from scipy

{davout-0.1.1.dev171 → davout-0.1.1.dev172/source/Davout.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: Davout
-Version: 0.1.1.dev171
+Version: 0.1.1.dev172
 Summary: Scientific computing utilities combining FEM, ANN and tooling
 Author: Matheus Janczkowski
 Author-email: Matheus Janczkowski <matheusj2009@hotmail.com>