diff --git a/pytorch_tabnet/abstract_model.py b/pytorch_tabnet/abstract_model.py
index 29be89f9..f5505fdb 100644
--- a/pytorch_tabnet/abstract_model.py
+++ b/pytorch_tabnet/abstract_model.py
@@ -70,7 +70,7 @@ def __post_init__(self):
         # Defining device
         self.device = torch.device(define_device(self.device_name))
         if self.verbose != 0:
-            print(f"Device used : {self.device}")
+            warnings.warn(f"Device used : {self.device}")
 
     def __update__(self, **kwargs):
         """
@@ -210,9 +210,8 @@ def fit(
         self._set_callbacks(callbacks)
 
         if from_unsupervised is not None:
-            print("Loading weights from unsupervised pretraining")
             self.load_weights_from_unsupervised(from_unsupervised)
-
+            warnings.warn("Loading weights from unsupervised pretraining")
         # Call method on_train_begin for all callbacks
         self._callback_container.on_train_begin()
 
@@ -622,9 +621,9 @@ def _set_callbacks(self, custom_callbacks):
             )
             callbacks.append(early_stopping)
         else:
-            print(
-                "No early stopping will be performed, last training weights will be used."
-            )
+            wrn_msg = "No early stopping will be performed, last training weights will be used."
+            warnings.warn(wrn_msg)
+
         if self.scheduler_fn is not None:
             # Add LR Scheduler call_back
             is_batch_level = self.scheduler_params.pop("is_batch_level", False)
diff --git a/pytorch_tabnet/callbacks.py b/pytorch_tabnet/callbacks.py
index e521c496..cb031d54 100644
--- a/pytorch_tabnet/callbacks.py
+++ b/pytorch_tabnet/callbacks.py
@@ -4,6 +4,7 @@
 import numpy as np
 from dataclasses import dataclass, field
 from typing import List, Any
+import warnings
 
 
 class Callback:
@@ -167,7 +168,8 @@ def on_train_end(self, logs=None):
                 + f"best_{self.early_stopping_metric} = {round(self.best_loss, 5)}"
             )
             print(msg)
-        print("Best weights from best epoch are automatically used!")
+        wrn_msg = "Best weights from best epoch are automatically used!"
+        warnings.warn(wrn_msg)
 
 
 @dataclass