include training loss in mlp

constrained_linear_regression/base.py

@@ -127,6 +127,7 @@ def __init__(
         epsilon=1e-8,
         n_iter_no_change=10,
         max_fun=15000,
+        training_losses=[],
     ):
         super().__init__(
             hidden_layer_sizes=hidden_layer_sizes,
@@ -153,12 +154,16 @@ def __init__(
             n_iter_no_change=n_iter_no_change,
             max_fun=max_fun,
         )
+        self.training_losses = training_losses
         assert shuffle is False, "shuffle should be False in the contrained ML."
 
     @abc.abstractmethod
     def fit(self, X, y):
         pass
 
+    def _save_mse(self, loss):
+        self.training_losses.append(loss)
+
     def _verify_coef(self, feature_count, coef, value, idx=0):
         if coef is not None:
             coef_ = coef

constrained_linear_regression/constrained_multi_layer_perceptron.py

@@ -371,6 +371,7 @@ def _loss_grad_constrained_lbfgs(
         )
         self._update_coef_using_constrain(coef_grads)  # it throws error.
         grad = _pack(coef_grads, intercept_grads)
+        self._save_mse(loss)
         return loss, grad