switch from properties to methods

transformers/modeling_utils.py

@@ -87,21 +87,37 @@ class PreTrainedModel(nn.Module):
     def base_model(self):
         return getattr(self, self.base_model_prefix, self)
 
-    @property
-    def input_embeddings(self):
+    def get_input_embeddings(self):
+        """ Get model's input embeddings
+        """
         base_model = getattr(self, self.base_model_prefix, self)
-        return base_model.input_embeddings
+        if base_model is not self:
+            return base_model.get_input_embeddings()
+        else:
+            raise NotImplementedError
 
-    @property
-    def output_embeddings(self):
+    def set_input_embeddings(self, value):
+        """ Set model's input embeddings
+        """
+        base_model = getattr(self, self.base_model_prefix, self)
+        if base_model is not self:
+            base_model.set_input_embeddings(value)
+        else:
+            raise NotImplementedError
+
+    def get_output_embeddings(self):
+        """ Get model's output embeddings
+            Return None if the model doesn't have output embeddings
+        """
         return None  # Overwrite for models with output embeddings
 
     def tie_weights(self):
         """ Make sure we are sharing the input and output embeddings.
             Export to TorchScript can't handle parameter sharing so we are cloning them instead.
         """
-        if self.output_embeddings is not None:
-            self._tie_or_clone_weights(self.output_embeddings, self.input_embeddings)
+        output_embeddings = self.get_output_embeddings()
+        if output_embeddings is not None:
+            self._tie_or_clone_weights(output_embeddings, self.get_input_embeddings())
 
     def _tie_or_clone_weights(self, output_embeddings, input_embeddings):
         """ Tie or clone module weights depending of weither we are using TorchScript or not
@@ -150,9 +166,10 @@ class PreTrainedModel(nn.Module):
         return model_embeds
 
     def _resize_token_embeddings(self, new_num_tokens):
-        old_embeddings = self.input_embeddings
-        self.input_embeddings = self._get_resized_embeddings(old_embeddings, new_num_tokens)
-        return self.input_embeddings
+        old_embeddings = self.get_input_embeddings()
+        new_embeddings = self._get_resized_embeddings(old_embeddings, new_num_tokens)
+        self.set_input_embeddings(new_embeddings)
+        return self.get_input_embeddings()
 
     def _get_resized_embeddings(self, old_embeddings, new_num_tokens=None):
         """ Build a resized Embedding Module from a provided token Embedding Module.