.

src/thefittest/base/_gpnn.py

@@ -237,7 +237,7 @@ def train_net_structure(
 
     net = optimizer.get_fittest()["phenotype"].copy()
 
-    return net
+    return net, optimizer._stats
 
 
 class BaseGPNN(BaseEstimator, metaclass=ABCMeta):
@@ -278,6 +278,9 @@ def array_like_to_numpy_X_y(
         y = np.array(y, dtype=np.float64)
         return X, y
 
+    def get_net(self) -> Net:
+        return self.net_
+
     @staticmethod
     def genotype_to_phenotype_tree(
         tree: Tree, n_variables: int, n_outputs: int, output_activation: str, offset: bool
@@ -402,7 +405,7 @@ def fit(self, X: ArrayLike, y: ArrayLike):
             task_type = "regression"
             n_outputs = 1
 
-        self.net_ = train_net_structure(
+        self.net_, self.optimizer_stats_ = train_net_structure(
             uniset=uniset,
             X_train=X_train,
             y_train=y_train,

src/thefittest/base/_mlp.py

@@ -170,9 +170,9 @@ def train_net_weights(
     optimizer = weights_optimizer(**weights_optimizer_args)
     optimizer.fit()
 
-    phenotype = optimizer.get_fittest()["phenotype"]
+    net = optimizer.get_fittest()["phenotype"]
 
-    return phenotype, optimizer._stats
+    return net, optimizer._stats
 
 
 class BaseMLPEA(BaseEstimator, metaclass=ABCMeta):